Search

Treballs Finals de Grau d'Enginyeria Biomèdica. Facultat de Medicina i Ciències de la Salut. Universitat de Barcelona. Curs: 2022-2023. Tutor/Director: Fernández Esparrach, Glòria, Guardiola García, Marta, Microwave imaging (MWI) is an emerging medical imaging technique with promising biomedical applications. In line with this research area, MiWEndo Solutions is a spinoff devoted to the development of a MWI device for colorectal cancer diagnosis called MiWEndo, which is attached to the regular colonoscope. Before its actual commercialization, a fundamental step towards the validation of the MiWEndo prototype is the design of realistic colon tissue-mimicking phantoms. Such phantoms allow the assessment of the imaging system performance under well controlled and reproducible conditions. In this thesis, a new, simple, and highly reproducible phantom recipe based on polyvinylpyrrolidone (PVP) has been developed to mimic healthy colon mucosa. Moreover, a comparative study between the currently used oil-based phantom recipe and the new PVP phantom has been conducted to assess the lifespan and stability of each type of fabricated phantom. It has been concluded that PVP phantoms must be conserved in the fridge for increased stability. With this conservation protocol, the PVP recipe shows a similar lifespan as the oil-based one. Nevertheless, due to both the limited number of samples and time, further studies must be carried out to conclusively determine the recipe with larger stability. Furthermore, a colon phantom motorized measuring system has been designed to improve repeatability and reliability in the measurements of the phantoms. The system consists of an XYZ motorized positioning system, an Arduino Nano, stepper motor drivers and Arduino IDE firmware. Even though further improvements will be required once implemented, a first conception of a system that fulfils the established requirements has been obtained, based on reasonable-cost and simple-implementation components. Besides, a successful proof of concept of the measuring system has been carried out, concluding that a further implementation of the system seems viable.

Treballs Finals de Grau d'Enginyeria Biomèdica. Facultat de Medicina i Ciències de la Salut. Universitat de Barcelona. Curs: 2022-2023. Tutor/Director: Sala Llonch, Roser, Mata Miquel, Christian, Munuera, Josep, Skin disorders are the most common type of cancer in the world and the incident has been lately increasing over the past decades. Even with the most complex and advanced technologies, current image acquisition systems do not permit a reliable identification of the skin lesion by visual examination due to the challenging structure of the malignancy. This promotes the need for the implementation of automatic skin lesion segmentation methods in order to assist in physicians’ diagnostic when determining the lesion's region and to serve as a preliminary step for the classification of the skin lesion. Accurate and precise segmentation is crucial for a rigorous screening and monitoring of the disease's progression. For the purpose of the commented concern, the present project aims to accomplish a state-of-the-art review about the most predominant conventional segmentation models for skin lesion segmentation, alongside with a market analysis examination. With the rise of automatic segmentation tools, a wide number of algorithms are currently being used, but many are the drawbacks when employing them for dermatological disorders due to the high-level presence of artefacts in the image acquired. In light of the above, three segmentation techniques have been selected for the completion of the work: level set method, an algorithm combining GrabCut and k-means methods and an intensity automatic algorithm developed by Hospital Sant Joan de Déu de Barcelona research group. In addition, a validation of their performance is conducted for a further implementation of them in clinical training. The proposals, together with the got outcomes, have been accomplished by means of a publicly available skin lesion image database.

In this study, we examine the suitability of desorption electro-flow focusing ionization (DEFFI) for mass spectrometry imaging (MSI) of biological tissue. We also compare the performance of desorption electrospray ionization (DESI) with and without the flow focusing setup. The main potential advantages of applying the flow focusing mechanism in DESI is its rotationally symmetric electrospray jet, higher intensity, more controllable parameters, and better portability due to the robustness of the sprayer. The parameters for DEFFI have therefore been thoroughly optimized, primarily for spatial resolution but also for intensity. Once the parameters have been optimized, DEFFI produces similar images to the existing DESI. MS images for mouse brain samples, acquired at a nominal pixel size of 50 µm, are comparable for both DESI setups, albeit the new sprayer design yields better sensitivity. Furthermore, the two methods are compared with regard to spectral intensity as well as the area of the desorbed crater on rhodamine-coated slides. Overall, the implementation of a flow focusing mechanism in DESI is shown to be highly suitable for imaging biological tissue and has potential to overcome some of the shortcomings experienced with the current geometrical design of DESI.

Obesity is a preventable risk factor for cerebrovascular disorders and it is associated with cerebral grey and white matter changes. Specifically, individuals with obesity show diminished grey matter volume and thickness, which seems to be more prominent among fronto-temporal regions in the brain. At the same time, obesity is associated with lower microstructural white matter integrity, and it has been found to precede increases in white matter hyperintensity load. To date, however, it is unclear whether these findings can be attributed solely to obesity or whether they are a consequence of cardiometabolic complications that often co-exist with obesity, such as low-grade systemic inflammation, hypertension, insulin resistance, or dyslipidemia. In this narrative review we aim to provide a comprehensive overview of the potential impact of obesity and a number of its cardiometabolic consequences on brain integrity, both separately and in synergy with each other. We also identify current gaps in knowledge and outline recommendations for future research.

Treballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2023, Director: Oliver Díaz, [en] Cancer, together with cardiovascular diseases, are the leading causes of death worldwide. In 2020, breast cancer became the most commonly diagnosed cancer in the world. This disease is mainly diagnosed by imaging techniques such as mammography, but one of the main limitations of this technique is that the compression of the breast, which is necessary to carry it out, often causes discomfort to patients. In addition, there is misinformation about how breast volume and other measures related to this feature can be directly related to breast cancer suffering. This is because the machines used to take breast measurements are generally very expensive and, in practice, these measurements are taken with inaccurate tools such as tape or calipers. But less than ten years ago, with the advent of Microsoft’s Kinect device, studies have emerged that have explored the feasibility of measuring breast volume with this cheaper, simpler and potentially more accurate technology. Using a Pico Flexx camera with characteristics very similar to those of the Kinect, I captured an object in three dimensions with a depth image that I processed to obtain a reconstruction of its surface. I then compared the resulting measurements with the original object to assess the accuracy of the device and demonstrate whether it can be used to take measurements of the breast in a clinical setting.

[spa] INTRODUCCIÓN: El Hematoma intracerebral espontáneo (HIE) es una causa importante de mortalidad y morbilidad en todo el mundo, por ello es fundamental identificar los pacientes con mayor riesgo de mal pronóstico, para intentar estratificarlos y adaptar terapias invasivas en aquellos de mayor riesgo. OBJETIVOS: Analizar los parámetros cualitativos (signos radiológicos) de la TCCSC que predicen un mal pronóstico funcional al alta y desarrollar un modelo cuantitativo (clasificadores de inteligencia artificial basados en radiómica) para predecir mal pronóstico al alta en pacientes con HIE. MÉTODOS: Estudio observacional retrospectivo en el que se analizaron pacientes con HIE confirmado por TCCSC entre enero 2016 y abril 2018. Se incluyeron pacientes de más de de 18 años y con TCCSC realizado dentro de las primeras 24 horas desde el inicio de los síntomas. Se excluyeron los pacientes con HIE secundario. En el estudio cuantitativo además se excluyeron los pacientes en los que no estaban disponibles los variables de radiómica. En la TCCSC fueron analizados los signos radiológicos (parámetros cualitativos) por dos radiólogos independientes, sin acceso a la información clínica y a la evolución. Los signos radiológicos (parámetros cualitativos) evaluados fueron los siguientes: - Signos individuales: Black hole sign (BH), Blend sign (BS), cualquier hipodensidad, Island sign (IS), Swirl sign (SwS) y la clasificación de barras atendiendo a densidad y morfología. - Nuevos parámetros combinados: combined barras total score (CBTS) y hematoma maturity score (HMS). La variable resultado fue el grado de dependencia al alta según la Escala Modificada de Rankin (mRS): no síntomas o incapacidad muy leve-leve (mRS 0-2); Incapacidad moderada o grave (mRS 3-5) y muerte (mRS 6). En segundo lugar, se analizaron los biomarcadores de imagen cuantitativa (clasificadores de inteligencia artificial basados en radiómica) en la TCCSC. Estos biomarcadores se extrajeron utilizando el módulo “Radiomics” del programa 3D-Slicer con segmentación manual de la TCCSC de cada HIE por un radiólogo cualificado, sin acceso a la información clínica. Después de imputar valores perdidos y valores atípicos, los valores se estandarizaron y normalizaron. La muestra se dividió en una cohorte de entrenamiento y otra de validación (70-30% respectivamente). Se usaron diferentes métodos de selección de variables y clasificadores predictivos, para clasificar el pronóstico funcional del paciente al alta en buen/mal pronóstico(mRS0-2/mRS3-6). Se realizaron 10 iteraciones de validación cruzada estratificada en la cohorte de entrenamiento y se calculó la media de los valores de área bajo la curva (AUC). Una vez entrenados los distintos modelos, se calculó la sensibilidad de cada uno en la cohorte de validación. RESULTADOS: En el estudio cualitativo (signos radiológicos) se incluyeron un total de 114. El estudio multivariable se ajustó según las variables que obtuvieron un resultado p, [eng] OBJECTIVES: The purpose of this study was to analyze the qualitative parameters (radiological signs) of non-contrast CT (NCCT) that predict poor functional prognosis at discharge and to develop a quantitative model (radiomics-based artificial intelligence classifiers) to predict poor prognosis at discharge in patients with intracerebral hemorrhage (ICH). METHODS: Patients with spontaneous ICH who underwent baseline CT scan were retrospectively analyzed. Black Hole Sign(BH), Blend Sign(BS), Island sign(IS), Swirl sign(SwS), Barras classification, any hypodensity, any irregularity and two combined novel indicators, Combined Barras Total Score (CBTS) and Hematoma Maturity Score, were assessed independently by two radiologists blinded to clinical information. Patients were trichotomized depending on the disability or dependency at discharge according to modified Rankin Scale (mRS): mRS 0-2; mRS 3-5 and mRS 6. Biomarkers were extracted using the 3D-Slicer Radiomic-module with a manual segmentation by a certified radiologist. After impute missing values and drop out outliers, radiomics values were standardized and normalized to train different classifiers. Sample was split in 70-30% training-test and validation cohort. A variety of linear, nonlinear, and ensemble predictive classifying models, along with several feature selection methods, were used to classify the binary outcome of good/poor outcome at discharge. Tenfold stratified cross validation method was used, and average AUC was calculated. Sensitivity rate was calculated in validation cohort. RESULTS: 114 patients with spontaneous ICH confirmed in NCCT were included in the qualitative analysis (radiological signs). Multivariable statistical analysis was adjusted for anticoagulation, hematoma volume, ventricular expansion, hypertension, blood glucose level at admission, age and history of atrial fibrillation and demonstrated that any hypodensity (OR 4.768, p 0.006), any irregularity(OR 4.768, p 0.006), CBTS ≥ 4 (OR 3.205, p0.025) and the new Hematoma maturity score (Immature)(OR 5.872, p 0.006) are independent predictors of functional outcome at discharge. Regarding quantitative analysis (radiomics-based artificial intelligence classifiers), Polynomial-SVM, KNN-Euclidean and Random Forest-10 in combination with ANOVA correlation features selection method were the best-performing classifiers in training-test cohort(AUC of 0.798, 0.752 and 0.742 respectively). For the best performing models, the sensitivity rate in validation cohort was 0.892(0.778-1;95%IC), with false negative rate of 0%. CONCLUSIONS: The new concept of Hematoma maturity Score was the radiological sign on NCCT with the highest impact on clinical outcome in comparison with the rest of the evaluated radiological signs. The use of radiomic biomarkers with machine learning methods are a promising diagnostic tool for predict functional outcome at discharge in patients with ICH, with a low false negative rate.

Treballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2022, Director: Sergio Escalera Guerrero, Carles Casacuberta i Rubén Ballester Bautista, [en] Cardiovascular diseases are a major cause of death and disability. Deep learning-based segmentation methods could help to reduce their severity by aiding in early diagnosing but high levels of accuracy are necessary. The vast majority of methods focus on correcting local errors and miss the global picture. To ad- dress this issue, researchers have developed techniques that incorporate global context and consider the relationships between pixels. Here, we apply persistent homology, a branch of topology that studies the topological structure of shapes, along with deep learning methods to improve the heart segmentation. We use multidimensional topological losses to avoid spurious components and holes and increase the total accuracy. We evaluate the performance of three different approaches: using the dice and pixel-wise losses with the sum of persistences of label diagrams as a regularizer, using the dice and pixel-wise losses with the bottleneck distance as a regularizer, and using both losses without any regularization. We find that, while more computationally demanding, the methods using topological regularizers outperform the other method in terms of accuracy.

Colon Capsule Endoscopy (CCE) is a minimally invasive procedure which is increasingly being used as an alternative to conventional colonoscopy. Videos recorded by the capsule cameras are long and require one or more experts’ time to review and identify polyps or other potential intestinal problems that can lead to major health issues. We developed and tested a multi-platform web application, AI-Tool, which embeds a Convolution Neural Network (CNN) to help CCE reviewers. With the help of artificial intelligence, AI-Tool is able to detect images with high probability of containing a polyp and prioritize them during the reviewing process. With the collaboration of 3 experts that reviewed 18 videos, we compared the classical linear review method using RAPID Reader Software v9 and the new software we present. Applying the new strategy, reviewing time was reduced by a factor of 6 and polyp detection sensitivity was increased from 81.08% to 87.80%.

Treball Final de Grau de Podologia, Facultat de Medicina i Ciències de la Salut, Universitat de Barcelona, curs: 2021-2022, Tutor: Elena de Planell i Mas, Introducció: Els ossos supernumeraris són estructures ossificades accessòries, generalment congènites, situades pròximes a un altre os o una articulació, inclús es poden trobar, com els sesamoides, en el recorregut d’un tendó. Normalment,són asimptomàtics,s’exposen diferent tipologia d’ossicles que estan més predisposats a patir processos patològics com l’os trigonum, l’os vesalianum, el navicular accessori, el talus secundarius, l’os subtibiale i el peroneus longus. Objectius: Conèixer la diferent simptomatologia dels ossicles accessoris simptomàtics en el peu. Determinar les proves per imatge per realitzar un diagnòstic dels ossicles accessoris. Categoritzar el pla de tractament pels ossicles accessoris simptomàtics del peu. Materials i mètodes: S’han realitzat 6 fórmules de cerca bibliogràfica a les bases de dades mèdiques Pubmed i Scopus dels últims cinc anys, una per os. Resultats: Van incloure’s 39 articles que es disposaren en taules diferenciades per ossicle on es presenta principalment la simptomatologia, l’origen, la clínica, les proves per imatge i el tractament. Conclusions: el dolor i l’edema són presents en tots els ossicles accessoris simptomàtics, amb diferents afectacions circumdants intrínseques a cadascun. La prova per imatge per excel·lència és la radiografia tot i que es secunden d’altres proves complementàries que mostren l’estat dels teixits tous adjacents. El tractament inicial té un abordatge conservador multidisciplinari on l’objectiu principal és fer desaparèixer la simptomatologia. Si aquest enfocament conservador fracassa, és ostensible que la cirurgia artroscòpica amb exèresis de l’ossicle serà d’elecció.

Treball Final de Grau de Podologia, Facultat de Medicina i Ciències de la Salut, Universitat de Barcelona, curs: 2021-2022, Tutor: Artur Crespo, Introducció: La fascitis plantar és una de les patologies més comunes del peu, representa el 80% de les patologies del taló. Als EUA al voltant d'un milió de visites anuals acaben amb el diagnòstic de fascitis plantar. En el 80-90% dels casos els símptomes desapareixen dins dels 10 mesos posteriors. Objectius: Analitzar si l'elastografia és un mètode efectiu pel diagnòstic precoç en les fasciopaties plantars en pacients amb simptomatologia lleu. Material i mètodes: Es va dur a terme una revisió sistemàtica de la literatura en les bases de dades Pubmed i Scopus dels articles que valoressin l’electrografia com a mètode de diagnosis de la fasciopatia plantar. Resultats: En els 9 estudis analitzats es van estudiar un total de 538 fàscies plantars. En 4 dels estudis s’utilitza el mode SWE (Elastografia d’ones de cisallament), en 3 el mode RTSE (Elastografia en temps real) i en dos dels estudis no determinen el mode d’elastografia que empren. Discussió: Tots els autors dels estudis analitzats conclouen que l’elastografia té una valoració positiva en el diagnòstic de la fasciopatia plantar. La caracterització de nous paràmetres ecogràfics de les propietats mecàniques pot ajudar en el diagnòstic precoç i a pacients amb simptomatologia lleu, tot i que, s'ha de determinar en estudis futurs. Conclusions: L'elastografia millora la precisió diagnòstica per la fasciopatia plantar, no obstant això, encara no es pot afirmar que sigui un mètode efectiu per la detecció de les fasciopaties plantar de manera precoç amb pacients amb simptomatologia lleu.

Valencian handball consists in hitting the ball with the hands and it may contribute to injury development on the hands. This study aimed to analyze skin temperature asymmetries and recovery after a cold stress test (CST) in professional players of Valencian handball before and after a competition. Thirteen professional athletes and a control group of ten physically active participants were measured. For both groups, infrared images were taken at the baseline condition; later they underwent a thermal stress test (pressing for 2 min with the palm of the hand on a metal plate) and then recovery images were taken. In athletes, the images were also taken after their competition. Athletes at baseline condition presented lower temperatures (p < 0.05) in the dominant hand compared with the non-dominant hand. There were asymmetries in all regions after their match (p < 0.05). After CST, a higher recovery rate was found after the game. The regions with the most significant differences in variation, asymmetries and recovery patterns were the index, middle and ring fingers, and the palm of the dominant hand. Taking into account that lower temperatures and the absence of temperature variation may be the consequence of a vascular adaptation, thermography could be used as a method to prevent injuries in athletes from Valencian handball.

Finding new ways to cost-effectively facilitate population screening and improve cancer diagnoses at an early stage supported by data-driven AI models provides unprecedented opportunities to reduce cancer related mortality. This work presents the INCISIVE project initiative towards enhancing AI solutions for health imaging by unifying, harmonizing, and securely sharing scattered cancer-related data to ensure large datasets which are critically needed to develop and evaluate trustworthy AI models. The adopted solutions of the INCISIVE project have been outlined in terms of data collection, harmonization, data sharing, and federated data storage in compliance with legal, ethical, and FAIR principles. Experiences and examples feature breast cancer data integration and mammography collection, indicating the current progress, challenges, and future directions. This research received funding mainly from the European Union’s Horizon 2020 research and innovation program under grant agreement no 952179. It was also partially funded by the Ministry of Economy, Industry, and Competitiveness of Spain under contracts PID2019-107255GB and 2017-SGR-1414. Peer Reviewed Article signat per 30 autors/es: Ivan Lazic (1), Ferran Agullo (2), Susanna Ausso (3), Bruno Alves (4), Caroline Barelle (4), Josep Ll. Berral (2), Paschalis Bizopoulos (5), Oana Bunduc (6), Ioanna Chouvarda (7), Didier Dominguez (3), Dimitrios Filos (7), Alberto Gutierrez-Torre (2), Iman Hesso (8), Nikša Jakovljević (1), Reem Kayyali (8), Magdalena Kogut-Czarkowska (9), Alexandra Kosvyra (7), Antonios Lalas (5) , Maria Lavdaniti (10,11), Tatjana Loncar-Turukalo (1),Sara Martinez-Alabart (3), Nassos Michas (4,12), Shereen Nabhani-Gebara (8), Andreas Raptopoulos (6), Yiannis Roussakis (13), Evangelia Stalika (7,11), Chrysostomos Symvoulidis (6,14), Olga Tsave (7), Konstantinos Votis (5) Andreas Charalambous (15) / (1) Faculty of Technical Sciences, University of Novi Sad, 21000 Novi Sad, Serbia; (2) Barcelona Supercomputing Center, 08034 Barcelona, Spain; (3) Fundació TIC Salut Social, Ministry of Health of Catalonia, 08005 Barcelona, Spain; (4) European Dynamics, 1466 Luxembourg, Luxembourg; (5) Centre for Research and Technology Hellas, 57001 Thessaloniki, Greece; (6) Telesto IoT Solutions, London N7 7PX, UK: (7) School of Medicine, Faculty of Health Sciences, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greece; (8) Department of Pharmacy, Kingston University London, London KT1 2EE, UK; (9) Timelex BV/SRL, 1000 Brussels, Belgium; (10) Nursing Department, International Hellenic University, 57400 Thessaloniki, Greece; (11) Hellenic Cancer Society, 11521 Athens, Greece; (12) European Dynamics, 15124 Athens, Greece; (13) German Oncology Center, Department of Medical Physics, Limassol 4108, Cyprus; (14) Department of Digital Systems, University of Piraeus, 18534 Piraeus, Greece; (15) Department of Nursing, Cyprus University of Technology, Limassol 3036, Cyprus

És ben conegut que moltes malalties afecten la morfologia dels vasos sanguinis, pel que és essencial que els metges els estudiïn. Per a un anàlisi més precís, paràmetres quantitatius dels vasos es poden obtenir directament de les imatges mèdiques, encara que avui en dia l'avaluació qualitativa per part d'experts es segueix considerant el mètode estàndard. El principal objectiu d'aquest projecte és recopilar les mètriques existents encaminades a obtenir la tortuositat dels vasos (atribut que defineix la torsió o flexió excessiva de la vascularitat), amb l’objectiu de promoure l'estudi d'índexs quantitatius per que puguin ser aplicats en un futur proper en l’àmbit mèdic. També es presenten índexs basats en altres tipus de característiques dels vasos per comprendre la importància actual de la tortuositat. Per aquest motiu s'ha realitzat un resum de les mètriques dels vasos de la literatura recent. S'han detallat els índexs, classificats per tipus d'atribut (mida, tortuositat i d’altres) i ordenat de més antic a més actual. Tot això, ha permès definir les similituds i les diferències a més de mostrar l'evolució dels índexs de la quantificació de vasos sanguinis. Es destaca que la tortuositat és l'atribut més important ja que la majoria d’articles d'investigació recollits se centren en aquesta propietat, és per això que es posen en rellevància les propietats essencials que haurien de tenir un índex de tortuositat ideal. En resum, les mètriques quantitatives semblen tenir un enorme potencial per diagnosticar de forma correcta patologies, però encara necessiten més investigació si es volen implementar com una eina mèdica. Aquesta revisió pretén ser un punt de partida on es reuneix tot el coneixement actual sobre aquestes mètriques per a poder millorar la futura investigació. Se sabe que muchas enfermedades afectan la morfología de los vasos sanguíneos, por lo que es esencial que los médicos las estudien. Para un análisis más preciso, los parámetros cuantitativos de los vasos se pueden extraer directamente de las imágenes médicas, aunque hoy en día la evaluación cualitativa por parte de expertos se sigue considerando el método estándar. El principal objetivo de este proyecto es recopilar las métricas existentes encaminadas a obtener la tortuosidad del vaso (atributo que define la torsión o flexión excesiva de la vascularidad), con el fin de promover el estudio de índices cuantitativos para que puedan ser aplicados en un futuro próximo en el campo médico. También se presentan índices basados en otro tipo de características de los vasos para comprender la importancia que tiene la tortuosidad actualmente. Para ello, se ha realizado un resumen de las métricas de los vasos de la literatura reciente. Se han detallado los índices, clasificados por tipo de atributo (tamaño, tortuosidad u otros) y ordenado de más antiguo a más actual. Todo ello, permite definir las similitudes y diferencias además de mostrar la evolución de los índices de la cuantificación de vasos sanguíneos. Se destaca que la tortuosidad es el atributo más importante ya que la mayoría de los artículos de investigación recogidos se centran en esta propiedad, por ello se ponen en relevancia las propiedades esenciales que debería tener un índice de tortuosidad ideal. En resumen, las métricas cuantitativas parecen tener un enorme potencial para diagnosticar correctamente patologías, pero aún se necesita más investigación si se quiere implementar como una herramienta m Many diseases are known to affect the morphology of blood vessels, making it essential for clinicians to study them. For a more precise analysis, quantitative parameters of vessels can be extracted from imaging, although nowadays qualitative evaluation by experts remains as the gold standard. The main focus of this project is to collect existing metrics aimed at obtaining vessel tortuosity (attribute that defines excessive torsion or bending of vascularity) in order to promote the study of quantitative indices to be applied in a near future in the medical field. To better understand the importance of tortuosity, other types of vessel characteristic indices are also presented. A summary of the vessel metrics of the recent literature has been carried out. Indexes have been detailed, classified by type of attribute (size, tortuosity or others) and sorted from oldest to current. All this makes it easier to define the similarities and differences between them and allows seeing the evolution of the vessel quantification indices. Vessel diameter, Distance Factor Metric (DFM), Tortuosity index (τ_3), Curve standard deviation (Ksd), Sum of Angles Metric (SOAM), Tortuosity Coefficient (TC) and Tortuosity density (t) are considered to be the most relevant metrics included in this review. It is noted that tortuosity is the most important vessel attribute as most of the research is focused on this property, therefore, the essential properties that an ideal tortuosity index should have are highlighted. To sum up, quantitative metrics seem to have an enormous potential to correctly diagnose pathologies, but there is still a need for more research if they want to be implemented as a medical tool. This review aims to be a starting point where all current knowledge regarding these metrics is brought together in order to enhance research.

[cat] INTRODUCCIÓ: Aproximadament, un 40% dels pacients amb epilèpsia focal son fàrmac-resistents. En aquests casos, la cirurgia de l’epilèpsia pot controlar o reduir significativament les crisis. La zona epileptògena (ZE) és la regió cerebral que la seva resecció implicaria la llibertat de crisis. Malauradament, no hi ha cap prova que determini amb precisió la ZE. HIPÒTESIS: Les alteracions de la DTI que valora la integritat de la substància blanca i les anomalies corticals determinades per la PET-FDG estan relacionades. Per altra banda, el PISCOM té una utilitat similar al SISCOM per detectar la zona d’inici ictal, però evitant la realització de la SPECT interictal. Aquests dos nous processaments de la neuroimatge cerebral permetrien determinar amb major precisió la ZE. OBJECTIUS: Valorar la utilitat de la combinació de la DTI amb la PET-FDG i el PISCOM a l’avaluació pre-quirúrgica de l’epilèpsia fàrmac-resistent. De forma més específica, es pretén determinar la localització de les alteracions de la DTI i la seva relació amb l’hipometabolisme cortical de la PET- FDG i la resecció de la potencial ZE, en pacient adults amb epilèpsia temporal medial amb esclerosi de hipocamp (ETM + EH). En el cas del PISCOM, l’objectiu es valorar la validesa del PISCOM per a identificar la potencial ZE, comparat amb el SISCOM ¡ la PET-FDG, en pacients pediàtrics amb una epilèpsia focal fàrmac-resistent. MÈTODES: De forma retrospectiva, els resultats de la DTI i la PET-FDG en una mostra de 21 pacients adults amb ETM + EH es van comparar amb un grup control sa, i a 18 d’aquests pacients se’ls havia realitzat una cirurgia d’epilèpsia. Respecte al PISCOM, la mostra estava formada per 22 pacients pediàtrics amb epilèpsia focal fàrmac-resistent, majoritàriament extra-temporal. Els resultats del PISCOM es van comparar amb les alteracions determinades pel SISCOM i la PET- FDG, realitzades durant l’avaluació pre-quirúrgica, i amb la ZE determinada per la resecció quirúrgica, la histologia, i el seguiment clínic post-quirúrgic favorable. RESULTATS: Les anomalies determinades per la combinació de la DTI i la PET-FDG van ser extenses, però sent més difuses quan la ETM + EH estava lateralitzada a l’hemisferi esquerre. Segons l’extensió de l’hipometabolisme de la PET-FDG al lòbul temporal epilèptic, es van diferenciar tres grups: anterior, mig i posterior. El grup anterior es relacionava amb alteracions de la DTI més anteriors, a diferència dels altes dos grups on les anomalies de la DTI eren tant anteriors com posteriors. No es va objectivar una relació clara entre les troballes de la DTI i la PET-FDG, i els resultats post-quirúrgics. El PISCOM va localitzar de forma exitosa la ZE en un major número de casos que el SISCOM (86% vs 54%, respectivament), i en una proporció similar a la PET-FDG. La ZE va ser identificada en tots els pacients només sent necessaris els resultats de la PET-FDG i el PISCOM. Es va observar una major concordança global entre observadors a l’analitzar els resultats del PISCOM, comparats amb els de la PET-FDG i el SISCOM. CONCLUSIONS: Els resultats de la combinació de la DTI i la PET-FDG mostren una relació entre ambdues tècniques, permetent identificar diferents xarxes neuronals implicades en una mateixa patologia (ETM + EH), en funció de l’extensió de l’hipometabolisme de la PET-FDG, però sense una clara relació amb el resultats post-quirúrgics. Els resultats del PISCOM suggereixen que podria ser una tècnica complementària per inferir la ZE als casos amb una PET-FDG dubtosa o negativa, substituint al SISCOM i evitant la realització de la SPECT interictal. Ambdues noves tècniques de neuroimatge podrien tenir una utilitat a l’avaluació pre-quirúrgica de l’epilèpsia per ajudar a inferir de forma més precisa la localització de la ZE., [eng] PURPOSE: In drug-resistant epilepsy, surgery can control or significantly reduce seizures. The epileptogenic zone (EZ) is the brain region whose resection would imply freedom from seizures. Unfortunately, there is no evidence to accurately determine EZ. The aim of this study is to determine whether the extent of 18F-FDG-PET hypometabolism combined to DTI abnormalities, and the PISCOM, a new nuclear imaging processing technique, could be useful to try to locate the EZ. METHODS: Twenty-one patients with MTLE-HS underwent comprehensive preoperative evaluation; 18 (86%) of these underwent epilepsy surgery. We analyzed and compared the pattern of white matter (WM) alterations on DTI and cortical hypometabolism on 18F-FDG-PET. Twenty-two children with pharmacorefractory epilepsy, mainly extratemporal, who had undergone presurgical assessment including SISCOM and 18F-FDG-PET and with postsurgical favorable outcome (Engel class I or II) for at least two years, were included. RESULTS: Widespread temporal and extratemporal 18F-FDG-PET and DTI abnormalities were found, involved similar brain regions, being more extensive in the left than the right MTLE-HS. Three groups were determined according to temporal 18F-FDG-PET patterns: hypometabolism restricted to the anterior third, to the middle third, and extending to the posterior third. Patients with anterior temporal hypometabolism showed DTI abnormalities in anterior tracts while patients with posterior hypometabolism showed WM alterations in anterior and posterior tracts. PISCOM concordance with the known EZ was significantly higher than SISCOM (p

Imaging genetic studies aim to test how genetic information influences brain structure and function by combining neuroimaging-based brain features and genetic data from the same individual. Most studies focus on individual correlation and association tests between genetic variants and a single measurement of the brain. Despite the great success of univariate approaches, given the capacity of neuroimaging methods to provide a multiplicity of cerebral phenotypes, the development and application of multivariate methods become crucial. In this article, we review novel methods and strategies focused on the analysis of multiple phenotypes and genetic data. We also discuss relevant aspects of multi-trait modelling in the context of neuroimaging data.

Radiomics is a novel image analysis technique, whereby voxel-level information is extracted from digital images and used to derive multiple numerical quantifiers of shape and tissue character. Cardiac magnetic resonance (CMR) is the reference imaging modality for assessment of cardiac structure and function. Conventional analysis of CMR scans is mostly reliant on qualitative image analysis and basic geometric quantifiers. Small proof-of-concept studies have demonstrated the feasibility and superior diagnostic accuracy of CMR radiomics analysis over conventional reporting. CMR radiomics has the potential to transform our approach to defining image phenotypes and, through this, improve diagnostic accuracy, treatment selection, and prognostication. The purpose of this article is to provide an overview of radiomics concepts for clinicians, with particular consideration of application to CMR. We will also review existing literature on CMR radiomics, discuss challenges, and consider directions for future work.

How the brain represents gender identity is largely unknown, but some neural differences have recently been discovered. We used an intrinsic ignition framework to investigate whether there are gender differences in the propagation of neural activity across the whole-brain and within resting-state networks. Studying 29 trans men and 17 trans women with gender incongruence, 22 cis women, and 19 cis men, we computed the capability of a given brain area in space to propagate activity to other areas (mean-ignition), and the variability across time for each brain area (node-metastability). We found that both measurements differentiated all groups across the whole brain. At the network level, we found that compared to the other groups, cis men showed higher mean-ignition of the dorsal attention network and node-metastability of the dorsal and ventral attention, executive control, and temporal parietal networks. We also found higher mean-ignition values in cis men than in cis women within the executive control network, but higher mean-ignition in cis women than cis men and trans men for the default mode. Node-metastability was higher in cis men than cis women in the somatomotor network, while both mean-ignition and node-metastability were higher for cis men than trans men in the limbic network. Finally, we computed correlations between these measurements and a body image satisfaction score. Trans men's dissatisfaction as well as cis men's and cis women's satisfaction toward their own body image were distinctively associated with specific networks in each group. Overall, the study of the whole-brain network dynamical complexity discriminates gender identity groups, functional dynamic approaches could help disentangle the complex nature of the gender dimension in the brain. H2020 Excellent Science, Grant/Award Number: 945539; H2020 Marie Skłodowska-Curie Actions, Grant/Award Number: 888692; Ministerio de Ciencia e Innovación, Grant/Award Numbers: PGC2018-094919-B-C21, PID2019-105772GB

State-of-the-art machine learning models, and especially deep learning ones, are significantly data-hungry; they require vast amounts of manually labeled samples to function correctly. However, in most medical imaging fields, obtaining said data can be challenging. Not only the volume of data is a problem, but also the imbalances within its classes; it is common to have many more images of healthy patients than of those with pathology. Computer-aided diagnostic systems suffer from these issues, usually over-designing their models to perform accurately. This work proposes using self-supervised learning for wireless endoscopy videos by introducing a custom-tailored method that does not initially need labels or appropriate balance. We prove that using the inferred inherent structure learned by our method, extracted from the temporal axis, improves the detection rate on several domain-specific applications even under severe imbalance. State-of-the-art results are achieved in polyp detection, with 95.00 ± 2.09% Area Under the Curve, and 92.77 ± 1.20% accuracy in the CAD-CAP dataset.

Complete digital pathology transformation for primary histopathological diagnosis is a challenging yet rewarding endeavor. Its advantages are clear with more efficient workflows, but there are many technical and functional difficulties to be faced. The Catalan Health Institute (ICS) has started its DigiPatICS project, aiming to deploy digital pathology in an integrative, holistic, and comprehensive way within a network of 8 hospitals, over 168 pathologists, and over 1 million slides each year. We describe the bidding process and the careful planning that was required, followed by swift implementation in stages. The purpose of the DigiPatICS project is to increase patient safety and quality of care, improving diagnosis and the efficiency of processes in the pathological anatomy departments of the ICS through process improvement, digital pathology, and artificial intelligence tools. This project was funded by European Regional Development Funds, Programa operatiu FEDER de Catalunya 2014–2020 and SA18-014623 DIGIPATICS. UPC activity in this project was partially supported by PID2020-116907RB-I00 and funded by MCIN/AEI/10.13039/501100011033 Peer Reviewed Article signat per 18 autors/es: Jordi Temprana-Salvador (1), Pablo López-García (2), Josep Castellví Vives (1),Lluís de Haro (2), Eudald Ballesta (2), Matias Rojas Abusleme (3), Miquel Arrufat (4), Ferran Marques (5), Josep R. Casas (5),Carlos Gallego (6), Laura Pons (7), José Luis Mate (7), Pedro Luis Fernández (7), Eugeni López-Bonet (8), Ramon Bosch (9), Salomé Martínez (10), Santiago Ramón y Cajal (1), and Xavier Matias-Guiu (11,12) // (1) Department of Pathology, Vall d’Hebron University Hospital, CIBERONC, 08035 Barcelona, Spain; (2) Functional Competence Center, Information Systems, Catalan Health Institute (Institut Català de la Salut), 08006 Barcelona, Spain; (3) Center for Telecommunications and Information Technology (Centre de Telecomunicacions i Tecnologies de la Informació, CTTI), Catalan Health Institute (Institut Català de la Salut), 08006 Barcelona, Spain; (4) Economic and Financial Management, Catalan Health Institute (Institut Català de la Salut), 08006 Barcelona, Spain; (5) Image Processing Group, Technical University of Catalonia (UPC), 08034 Barcelona, Spain; (6) Digital Medical Imaging System of Catalonia (SIMDCAT), TIC Salut, 08005 Barcelona, Spain, (7) Department of Pathology, Germans Trias i Pujol University Hospital, 08916 Badalona, Spain; (8) Department of Pathology, Doctor Josep Trueta Hospital of Girona, 17007 Girona, Spain; (9) Department of Pathology, Verge de la Cinta Hospital of Tortosa, 43500 Tarragona, Spain; (10) Department of Pathology, Joan XXIII University Hospital of Tarragona, 43005 Tarragona, Spain; (11) Department of Pathology, Arnau de Vilanova University Hospital, 25198 Lleida, Spain, (12) Department of Pathology, Bellvitge University Hospital, CIBERONC, 08907 Barcelona, Spain

Treballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2022, Director: Simone Balocco, [en] Intravascular Ultrasound (IVUS) is an intra-operative imaging modality that facilitates observing the vessel wall structure of the human coronary arteries. Segmentation of the arterial wall from IVUS images is crucial for the analysis of the wall’s characteristics and for 3D reconstructed models of the artery. The aim of this project is to learn, study, develop and evaluate a model capable of performing semantic segmentation on medical images using a small IVUS dataset. The dataset used contains 435 pullbacks from 10 different patients acquired by 20 MHz probes. Our proposed model uses the InceptionResNet architecture pretrained with the ImageNet dataset as the encoder. The model is then connected like a U-Net with a decoder which is trained using the images. We have explored different approaches in data augmentation and the best results were given by rotations up to 90 degrees, vertical & horizontal flips and zooming. Among the results, we obtained an average of Dice similarity coefficient, or Dice Index, of 0.75 for the media and 0.92 for the lumen, with pretty low variance. However, the Hausdorff distance (HD) from results show that some misclassified sections appear in average 0.51 and 0.67 mm away from the ground truth class, for the lumen and media respectively. Finally, the Jaccard Measure for the lumen is quite high: 0.85; whereas for the media is 0.60, which is pretty low. The results found in this project present a good start on how to implement a solution for the problem explained.

Treballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2022, Director: Simone Balocco, [en] Ischemic Heart Disease (IHD) is one of the leading causes of mortality in Spain; early diagnosis is key. Intravenous ultrasound imaging (IVUS) can help identify symptoms of IHD, at the cost of segmenting a large volume of frames by medical professionals. While promising, automated image segmentation using Convolutional Neural Networks (CNN) suffer from sample scarcity: a large amount of parameters is often used, and medical imaging datasets are typically small and costly to acquire and label. In this report we study and compare state of the art methods used to deal with sample scarcity. In particular we introduce data augmentation methodologies, specialized training losses and transfer learning methods, and compare their performance on IVUS segmentation of the media and lumen or the artery. Additionally we introduce a promising paradigm, few-shot segmentation, and provide an initial implementation using PFENet. This implementation can avoid significant overfitting, even when trained with a single example, outperforming traditional CNNs on the same segmentation problem.

Many efforts have been made to discover tumor-specific microenvironment elements (TMEs) from immunostained tissue sections. However, the identification of yet unknown but relevant TMEs from multiplex immunostained tissues remains a challenge, due to the number of markers involved (tens) and the complexity of their spatial interactions. We present NaroNet, which uses machine learning to identify and annotate known as well as novel TMEs from self-supervised embeddings of cells, organized at different levels (local cell phenotypes and cellular neighborhoods). Then it uses the abundance of TMEs to classify patients based on biological or clinical features. We validate NaroNet using synthetic patient cohorts with adjustable incidence of different TMEs and two cancer patient datasets. In both synthetic and real datasets, NaroNet unsupervisedly identifies novel TMEs, relevant for the user-defined classification task. As NaroNet requires only patient-level information, it renders state-of-the-art computational methods accessible to a broad audience, accelerating the discovery of biomarker signatures., Comment: 37 pages, 4 figures

Background The development of liver cirrhosis is usually an asymptomatic process until late stages when complications occur. The potential reversibility of the disease is dependent on early diagnosis of liver fibrosis and timely targeted treatment. Recently, the use of non-invasive tools has been suggested for screening of liver fibrosis, especially in subjects with risk factors for chronic liver disease. Nevertheless, large population-based studies with cost-effectiveness analyses are still lacking to support the widespread use of such tools. The aim of this study is to investigate whether non-invasive liver stiffness measurement in the general population is useful to identify subjects with asymptomatic, advanced chronic liver disease. Methods This study aims to include 30,000 subjects from eight European countries. Subjects from the general population aged ≥ 40 years without known liver disease will be invited to participate in the study either through phone calls/letters or through their primary care center. In the first study visit, subjects will undergo bloodwork as well as hepatic fat quantification and liver stiffness measurement (LSM) by vibration-controlled transient elastography. If LSM is ≥ 8 kPa and/or if ALT levels are ≥1.5 x upper limit of normal, subjects will be referred to hospital for further evaluation and consideration of liver biopsy. The primary outcome is the percentage of subjects with LSM ≥ 8kPa. In addition, a health economic evaluation will be performed to assess the cost-effectiveness and budget impact of such an intervention. The project is funded by the European Commission H2020 program. Discussion This study comes at an especially important time, as the burden of chronic liver diseases is expected to increase in the coming years. There is consequently an urgent need to change our current approach, from diagnosing the disease late when the impact of interventions may be limited to diagnosing the disease earlier, when the patient is asymptomatic and free of complications, and the disease potentially reversible. Ultimately, the LiverScreen study will serve as a basis from which diagnostic pathways can be developed and adapted to the specific socio-economic and healthcare conditions in each country. Trial registration This study is registered on Clinicaltrials.gov (NCT03789825).

[cat] INTRODUCCIÓ: La lesió renal aguda postcontrast (LRA-PC) es defineix com un increment de la creatinina sèrica (Cr) ≥ 0,3mg/dl (26,5μmol/l) o ≥ 1,5 vegades comparada amb el valor basal de la Cr que apareix entre les 48-72 hores després de l’administració intravascular del contrast iodat. L’existència d’insuficiència renal crònica (IRC) es considera el factor de risc principal associat a l’aparició de la LRA-PC. La hidratació per via intravenosa (i.v.) és en l’actualitat l’única forma de profilaxi efectiva de la LRA-PC. HIPÒTESI: La hidratació per via oral no és inferior a la hidratació per via i.v. com a mesura profilàctica de la LRA-PC en pacients amb IRC estadi IIIb als que se’ls hi realitzi una tomografia computaritzada (TC) administrant contrast iodat i.v. OBJECTIUS: L’objectiu principal d’aquesta tesi és l’avaluació de forma prospectiva i randomitzada de la no inferioritat de la hidratació per via oral comparada amb la hidratació per via i.v. en la prevenció de la LRA-PC en pacients amb IRC estadi IIIb referits per una TC amb contrast iodat. Els objectius secundaris són: 1) l’anàlisi de la necessitat d’hemodiàlisi durant el mes posterior a la realització de la TC amb contrast iodat en el grup d’hidratació oral i en el d’hidratació i.v.; 2) l’anàlisi de la reversibilitat de la LRA-PC 15 dies després de la realització de la TC amb contrast iodat en els dos grups; 3) l’avaluació de la seguretat en ambdós grups d’hidratació; 4) l’avaluació de la no inferioritat de la hidratació via oral versus la hidratació via i.v. en la prevenció de la LRA-PC en el subgrup de pacients oncològics. MATERIAL I MÈTODES: S’ha realitzat un assaig clínic unicèntric, prospectiu, de distribució aleatòria, no emmascarat, amb dos grups paral·lels, de no inferioritat, l’estudi NICIR. Els pacients s’han randomitzat assignant-se 1:1 per rebre una pauta de profilaxi contra la LRA-PC sigui amb hidratació via oral: 500 ml d’aigua dues hores abans i 2000 ml durant les 24 hores després de l’administració de contrast iodat o amb hidratació i.v.; bicarbonat sòdic (166mmol/l) 3 ml/kg/h començant una hora abans i bicarbonat sòdic (166mmol/l) 1 ml/kg/h durant l’hora posterior a la TC amb contrast iodat. Retrospectivament, també hem analitzat aquesta no inferioritat de la hidratació oral respecte a la hidratació i.v. en el subgrup de pacients oncològics, que suposen el 74% dels pacients de l’estudi NICIR. RESULTATS: Dels 228 pacients que es van randomitzar entre el gener del 2018 fins al gener del 2019, 114 van rebre hidratació i.v. i 114 hidratació oral i van ser avaluables. No es van trobar diferències significatives en els factors de risc i les característiques clíniques associades entre les dues branques de l’estudi (p=0.13->0.95). La ràtio de LRA-PC va ser del 4.4% (95%CI: 1.4-9.9%) a la branca d’hidratació oral i del 5.3% (95%CI: 2.0-11.1%) a la branca d’hidratació i.v. La ràtio de LRA-PC no reversible va ser de 1.8% (95%CI: 0.2-6.2%) a les dues branques. Cap pacient va requerir diàlisi durant el mes posterior a la TC i no hi va haver cap efecte advers relatiu al règim d’hidratació. Dels 174 pacients inclosos en la subanàlisi retrospectiva dels pacients oncològics, 82 van rebre hidratació oral i 92 i.v. La ràtio de LRA-PC va ser del 3.7% a la branca d’hidratació oral i del 5.4% en la i.v. La LRA-PC persistent va ser de 1,8% a la branca oral i de 3.3% a la i.v. CONCLUSIONS: L’estudi NICIR demostra que la hidratació per via oral no és inferior a la hidratació per via i.v. com a mesura profilàctica de la LRA-PC en pacients amb IRC estadi IIIb als que se’ls hi realitza una TC amb contrast iodat. Aquesta hipòtesi també queda demostrada al subgrup de pacients oncològics, que aquest estudi ha avaluat específicament de manera retrospectiva., [eng] TITLE: Oral hydration compared to intravenous hydration in the prevention of post-contrast acute kidney injury in patients with chronic kidney disease stage IIIb: A phase III non-inferiority study (NICIR study) OBJECTIVE: To evaluate the non-inferiority of oral hydration compared to intravenous (i.v.) hydration in the prevention of post-contrast acute kidney injury (PC-AKI) in patients with stage IIIb chronic kidney disease (CKD) referred for an elective contrast-enhanced computed tomography (CE-CT). MATERIAL AND METHODS: This is a prospective, randomized, phase 3, parallel-group, open-label, non-inferiority trial. Patients were randomly assigned 1:1 to receive prophylaxis against PC-AKI either with oral hydration: 500 mL of water two hours before and 2000 mL during the 24 h after performing CE-CT or i.v. hydration: sodium bicarbonate (166 mmol/L) 3 mL/kg/h starting one hour before and sodium bicarbonate (166 mmol/L) 1 mL/kg/h during the first hour after CE-CT. 100 mL of non-ionic iodinated contrast was administered in all cases. The primary outcome was the proportion of PC-AKI in the first 48–72 h after CE-CT. Secondary outcomes were persistent PC-AKI, the need for hemodialysis, and the occurrence of adverse events related to prophylaxis. RESULTS: Of 264 patients randomized between January 2018 and January 2019, 114 received oral hydration, and 114 received i.v. hydration and were evaluable. No significant differences were found (p > 0.05) between arms in clinical characteristics or risk factors. PC-AKI rate was 4.4 % (95 %CI: 1.4-9.9 %) in the oral hydration arm and 5.3 % (95 %CI: 2.0-11.1%) in the i.v. hydration arm. The persistent PC-AKI rate was 1.8 % (95 %CI: 0.2-6.2 %) in both arms. No patient required dialysis during the first month after CE-CT or had adverse effects related to the hydration regime. CONCLUSION: In those with stage IIIb CKD referred for an elective CE-CT, we provide evidence of non-inferiority of oral hydration compared to i.v. hydration in the prevention of PC-AKI.

Background Dural chondrosarcoma is a very rare intracranial tumor, given that meninges do not normally contain cartilaginous tissue from which it can originate. We present a case of primary extraosseous dural chondrosarcoma. Case presentation A 48-year-old woman presented to our tertiary center neurosurgery consultation with progressive headache, vomiting, vertigo, and gait instability of 5 months’ duration. An initial brain CT revealed a large parietal mass with gross calcifications and subtle hyperostosis of the inner table. Subsequent brain MRI showed a heterogeneous expansive lesion with a honey-comb enhancement. Discussion of intra- or extra-axial location was warranted, and finally, initial presurgical suspicion of meningioma arose although some atypical imaging features were detected. The differential diagnosis included solitary fibrous tumor–hemangiopericytoma and dural metastasis. Total resection of the lesion was performed, extra-axial origin was confirmed, and pathology resulted in a primary dural chondrosarcoma. Conclusion The importance of this case presentation lies in the unusual nature of the final diagnosis, the brief literature review and differential diagnosis with emphasis on imaging pearls, as well as the useful reminder for physicians to consider less frequent diseases when key findings do not unambiguously lead to the usual suspects.

Computer-aided detection systems based on deep learning have shown great potential in breast cancer detection. However, the lack of domain generalization of artificial neural networks is an important obstacle to their deployment in changing clinical environments. In this study, we explored the domain generalization of deep learning methods for mass detection in digital mammography and analyzed in-depth the sources of domain shift in a large-scale multi-center setting. To this end, we compared the performance of eight state-of-the-art detection methods, including Transformer based models, trained in a single domain and tested in five unseen domains. Moreover, a single-source mass detection training pipeline was designed to improve the domain generalization without requiring images from the new domain. The results show that our workflow generalized better than state-of-the-art transfer learning based approaches in four out of five domains while reducing the domain shift caused by the different acquisition protocols and scanner manufacturers. Subsequently, an extensive analysis was performed to identify the covariate shifts with the greatest effects on detection performance, such as those due to differences in patient age, breast density, mass size, and mass malignancy. Ultimately, this comprehensive study provides key insights and best practices for future research on domain generalization in deep learning based breast cancer detection.

[eng] Arguably, one of the stimuli that move human the most is music. What physically consists in an array of sounds organized in time and frequency, is capable of evoking a rich palette of emotions, thoughts and behaviors in the listener. Despite the complexity of these psychological phenomena, however, music is also capable of evoking one of the most basic psychophysiological reactions to stimuli in humans: pleasure. Thus, a relevant question is how music, an arbitrary concatenation of sounds, ends up translated into this unique affective experience. As with many other aspects of human behavior, studying the brain function underlying this psychological phenomenon offers an insightful view on this intriguing inquiry. As of today, decades of research have been devoted to elucidate the neuropsychological mechanisms through which music-evoked pleasantness takes place. From psychological models of emotion and affect to evidence-based neuroscientific theories, the question of how music becomes pleasurable has been addressed from many different disciplines, approaches and methodologies. As a result, the brain architecture and mechanics underlying music evoked pleasantness have begun to be understood to great detail. Perhaps the most important breakthrough has been to verify that music engages the brain reward system, validating the idea that musical stimuli are indeed a source of hedonic impact. Nevertheless, along the brain reward system work a number of cortical areas responsible for the perceptual processing of musical stimuli. Altogether, this anatomical and functional interplay conforms an extensive network that actively contributes to the pleasurable experience of listening to music. Furthermore, other brain areas have been observed to be involved in these networks depending on the level of familiarity with the music stimuli, a factor that has been replicated to be intimately associated with music reward processing. As such, these networks utilize precise neural dynamics in order to transfer the information back and forth between the areas involved. A candidate mechanism for this information transfer are neural oscillations. By synchronizing to the same inputs, neuronal populations can pass information and process it efficiently to fulfill a task or demand. The neural oscillations associated with music-evoked pleasantness are just beginning to be understood. However, a common oscillatory signature of musical plea­ sure does appear consistently in the literature: theta oscillations. Nonetheless, whether the different areas engaged in music reward processing are indeed synchronized by these rhythms is yet to be determined. One goal of this thesis is to study whether theta rhythms indeed synchronize relevant brain nodes during pleasant music listening. In order to answer this question, in study 1 we acquired EEG signals from a cohort of participants while they listened and rated a number of musical excerpts. After analysis of the EEG data jointly with the self-reports, we observed that right frontal and temporal signals were more synchronized by theta rhythms the more pleasant music was found. These results converge with previous literature pinpointing right fronto-temporal connectivity to be actively involved in the process of assigning value to music, as well as with literature identifying theta rhythms as a correlate of music-evoked pleasantness and related constructs. Following this line of research, another goal of this thesis is to test whether the nodes synchronized by theta rhythms during pleasant music listening are indeed related to the music processing brain networks consistently replicated in the neuroimaging literature. For this purpose, in study 2 we acquired simultaneous fMRI-EEG recordings with a similar paradigm to that of study l. After replicating previous results identifying a fronto-temporal music-related brain network in our fMRI data, we studied the activity of this network in relation to theta synchronization as measured with EEG and reported pleasantness. We observed that theta synchronization between frontal and temporal signals was positively related to music-related brain function when music was reported to be liked the most, providing empirical evidence for the link between the two phenomena. Yet another intriguing question is whether liking for familiar music synchronizes similar or different brain nodes by theta rhythms. Familiar music has been observed to be liked more than unfamiliar music, suggesting qualitative and/or quantitative differences in its processing. lndeed, previous research has revealed that in addition to music processing areas, familiar music engages the parietal cortex. In study 3 we familiarized a cohort of participants with some music stimuli and had them rate the excerpts 24h later along a new set of stimuli while their EEG was recorded . We found that the right fronto-temporal theta connectivity profile found in the previous studies was only associated with liking for novel music. lnstead, temporo-parietal theta connectivity was more engaged when liking familiar music. These results go in line with previous research and pose the first attempt to our knowledge to study the familiarity effect on musical liking from an oscillatory dynamics perspective. Therefore, the research conducted in this thesis adds up to the state of the art in the neuroscience of music-evoked pleasantness. We provide novel evidence for the role of theta oscillations in synchronizing relevant brain nodes during pleasant music listening, as well as the different cortical nodes these rhythms synchronize depending on the familiarity level with music. Particularly, fronto-temporal theta rhythms were identified to be associated with music-evoked pleasantness overall, and with liking for novel music specifically. We also provide evidence that in the particular case of familiar music, liking is supported by theta temporo-parietal synchronization. Last but not least, our results also link empirically the fronto-temporal brain networks underlying music perception as measured with fMRI with theta oscillations associated with music-evoked pleasantness as measured with EEG. Altogether, the results presented in this thesis contribute to the understanding of the brain networks involved in pleasant and familiar music listening, insofar they corroborate theta rhythms as a means of synchronization dynamics, opening up new windows for future research, [spa] Podría decirse que uno de los estímulos que más conmueven al ser humano es la música. Lo que físicamente consiste en una serie de sonidos organizados en tiempo y frecuencia, es capaz de evocar una rica amalgama de emociones, pensamientos y comportamientos en el oyente. Sin embargo, a pesar de la complejidad de estos fenómenos psicológicos, la música también es capaz de evocar una de las reacciones psicofisiológicas más básicas que los estímulos pueden despertar en los seres humanos: placer. Por tanto, una cuestión relevante es cómo la música, una concatenación arbitraria de sonidos, acaba traduciéndose en esta experiencia afectiva única. Como ocurre con muchos otros aspectos del comportamiento humano, estudiar la función cerebral asociada a este fenómeno psicológico ofrece un punto de vista esclarecedor sobre esta intrigante cuestión. A día de hoy, se han dedicado décadas de investigación para dilucidar los mecanismos neuropsicológicos a través de los cuales se produce el placer evocado por la música. Desde modelos psicológicos de emoción y afecto hasta teorías neurocientíficas basadas en evidencia empírica, la cuestión de cómo la música se vuelve placentera se ha abordado desde muchas disciplinas, enfoques y metodologías diferentes. Como resultado, la arquitectura del cerebro y la mecánica subyacente a el placer a la música han comenzado a entenderse con gran detalle. Quizás el avance más importante ha sido verificar que la música activa el sistema de recompensa del cerebro, validando la idea de que los estímulos musicales son en efecto una fuente de impacto hedónico. Sin embargo, junto al sistema de recompensa cerebral funcionan una serie de áreas corticales responsables del procesamiento perceptual de los estímulos musicales. En conjunto, esta interacción anatómica y funcional conforma una red extensa que contribuye activamente a la experiencia de escuchar música placentera. Además, se ha observado que otras áreas del cerebro están involucradas en estas redes dependiendo del nivel de familiaridad con los estímulos musicales, un factor íntimamente asociado con el procesamiento de la recompensa musical, como así se ha replicado en la literatura . Como tal, estas redes utilizan dinámicas neuronales concretas para transferir la información entre las áreas involucradas. Un mecanismo candidato para explicar esta transferencia de información son las oscilaciones neuronales. Al sincronizarse con los mismos inputs, las poblaciones neuronales pueden comunicar información y procesarla de manera eficiente para cumplir con una tarea o demanda. Las oscilaciones neuronales asociadas al placer musical apenas han comenzado a entenderse. No obstante, hay un correlato oscilatorio del placer musical que aparece consistentemente en la literatura: las oscilaciones theta. Sin embargo, aún no se ha determinado si las diferentes áreas involucradas en el procesamiento de recompensas musicales están realmente sincronizadas a través de estos ritmos. Uno de los objetivos de esta tesis es estudiar si los ritmos theta realmente sincronizan los nodos cerebrales relevantes durante la escucha de música placentera. Para contestar esta pregunta, en el primer estudio adquirimos señales de EEG en un grupo de participantes mientras escuchaban y evaluaban una serie de fragmentos musicales. Después del análisis de los datos de EEG junto a las evaluaciones de placer, observamos que las señales frontales y temporales derechas estaban más sincronizadas a través de ritmos theta cuanto más placentera resultaba la música. Estos resultados convergen con literatura previa que señala que las redes fronto-temporales derechas participan activamente en el proceso de valoración musical, así como con literatura que identifica los ritmos theta como un correlato del placer evocado por la música y constructos relacionados. Siguiendo esta línea de investigación, otro objetivo de esta tesis es probar si los nodos sincronizados por ritmos theta durante la escucha de música placentera están real­ mente relacionados con las redes cerebrales de procesamiento musical normalmente halladas en la literatura de neuroimagen. Para ello, en el estudio 2 adquirimos registros simultáneos de fMRI-EEG en un paradigma similar. Después de replicar resultados anteriores identificando una red cerebral fronto-temporal relacionada con la música en nuestros datos de resonancia magnética funcional, estudiamos la actividad de esta red en relación con la sincronización theta medida con EEG, y el placer evocado. Observamos que la sincronización theta entre las señales frontales y temporales se relaciona positivamente con la función cerebral relacionada con la música cuando se reporta que ésta gusta más, proporcionando evidencia empírica del vínculo entre estos dos fenómenos. Otra cuestión importante es si el placer evocado por la música familiar sincroniza los mismos o diferentes nodos cerebrales por ritmos theta. Se ha observado que la música familiar gusta más que la música desconocida, lo que sugiere diferencias cualitativas y/o cuantitativas en su procesamiento. De hecho, investigaciones anteriores han revelado que, además de las áreas de procesamiento musical, la música familiar activa la corteza parietal. En el estudio 3 familiarizamos a un grupo participantes con algunos estímulos musicales y les pedimos que evaluaran los mismos fragmentos 24 horas después junto a un nuevo conjunto de estímulos mientras se registraba su EEG. Encontramos que la conectividad theta fronto-temporal derecha encontrada en los estudios anteriores solo se encontró asociada al placer evocado por música novel. En cambio, un perfil de conectividad theta temporo-parietal se encontró más asociado al placer cuando éste fue evocado por música familiar. Estos resultados están en línea con investigaciones previas y suponen el primer intento, a nuestro entender, en estudiar el efecto de familiaridad en el placer musical desde una perspectiva de dinámicas oscilatorias. Por todo esto, la investigación realizada en esta tesis contribuye a la vanguardia en la neurociencia del placer musical. Proporcionamos evidencia novedosa sobre el papel de las oscilaciones theta en la sincronización de los nodos cerebrales relevantes durante la escucha de música agradable, así como los diferentes nodos corticales que estos ritmos sincronizan según el nivel de familiaridad con la música. Particularmente, se identificó que los ritmos theta fronto-temporales están asociados con el placer evocado por la música en general, y específicamente con el placer evocado por la música nueva. También proporcionamos evidencia que en el caso particular de la música familiar, el placer está asociado con sincronización theta temporo-parietal. Por último, aunque no menos importante, nuestros resultados también vinculan empíricamente las redes cerebrales fronto-temporales relacionadas que subyacen a la percepción de la música (medidas con fMRI) con oscilaciones theta asociadas al placer musical (medidas con EEG). En conjunto, los resultados presentados en esta tesis contribuyen a la comprensión de las redes cerebrales involucradas en la escucha de música placentera y familiar, en la medida en que corroboran los ritmos theta como medio de comunicación neuronal, abriendo nuevas oportunidades para futuras investigaciones.

Cardiovascular research is in an ongoing quest for a superior imaging method to integrate gross-anatomical information with microanatomy, combined with quantifiable parameters of cardiac structure. In recent years, synchrotron radiation-based X-ray Phase Contrast Imaging (X-PCI) has been extensively used to characterize soft tissue in detail. The objective was to use X-PCI to comprehensively quantify ischemic remodeling of different myocardial structures, from cell to organ level, in a rat model of myocardial infarction. Myocardial infarction-induced remodeling was recreated in a well-established rodent model. Ex vivo rodent hearts were imaged by propagation based X-PCI using two configurations resulting in 5.8 µm and 0.65 µm effective pixel size images. The acquired datasets were used for a comprehensive assessment of macrostructural changes including the whole heart and vascular tree morphology, and quantification of left ventricular myocardial thickness, mass, volume, and organization. On the meso-scale, tissue characteristics were explored and compared with histopathological methods, while microstructural changes were quantified by segmentation of cardiomyocytes and calculation of cross-sectional areas. Propagation based X-PCI provides detailed visualization and quantification of morphological changes on whole organ, tissue, vascular as well as individual cellular level of the ex vivo heart, with a single, non-destructive 3D imaging modality., Scientific Reports, 11, ISSN:2045-2322

Deep learning models can enable accurate and efficient disease diagnosis, but have thus far been hampered by the data scarcity present in the medical world. Automated diagnosis studies have been constrained by underpowered single-center datasets, and although some results have shown promise, their generalizability to other institutions remains questionable as the data heterogeneity between institutions is not taken into account. By allowing models to be trained in a distributed manner that preserves patients’ privacy, federated learning promises to alleviate these issues, by enabling diligent multi-center studies. We present the first simulated federated learning study on the modality of cardiovascular magnetic resonance and use four centers derived from subsets of the M&M and ACDC datasets, focusing on the diagnosis of hypertrophic cardiomyopathy. We adapt a 3D-CNN network pretrained on action recognition and explore two different ways of incorporating shape prior information to the model, and four different data augmentation set-ups, systematically analyzing their impact on the different collaborative learning choices. We show that despite the small size of data (180 subjects derived from four centers), the privacy preserving federated learning achieves promising results that are competitive with traditional centralized learning. We further find that federatively trained models exhibit increased robustness and are more sensitive to domain shift effects.

Treballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2021, Director: Karim Lekadir, [en] Big data has benefits and complications. Analysing big data has the capability of obtaining great knowledge which can be used for taking the best decisions, but it is an execution of high complexity. In medicine, the amount of data generated every day is more than enough to be called big data. In this project, the data extracted from radiomics, an image analysis technique that generates about 650 different variables, is treated with the objective of being visualized in an understandable and intuitive way. These visualizations are compiled in the Radiomie wide association study web page, a tool that will help investigators formulating, confirming or refuting their theories, which could be used to support new health politics.

Background: Individuals with obesity are known to present cognitive deficits, especially in executive functions. Executive functions play an important role in health and success throughout the whole life and have been related to food decision-making and to the ability to maintain energy balance. It is possible to improve executive functions through targeted training. This would involve brain plasticity changes that could be studied through connectivity MRI. The general hypothesis of this study is that executive functions training in children with obesity can improve food choices and produce cognitive and neuroimaging changes (structural and functional connectivity), as well as improve emotional state and quality of life.Methods: Randomized controlled double-blind trial with 12-month follow-up. Thirty children with obesity will be randomly allocated into “executive training” (Cognifit with adaptive difficulty + Cogmed) or “control task” group (Cognifit without adaptive difficulty). Both groups will attend 30–45 min of individual gamified training (Cogmed and/or Cognifit systems) by iPad, five times per week during 6 weeks. Cogmed and Cognifit software are commercially available from Pearson and Cognifit, respectively. Participants will receive an iPad with both apps installed for a 6-week use. Participants will also receive counseling diet information via presentations sent to the iPad and will wear a Fitbit Flex 2 tracker to monitor daily activity and sleep patterns. Main outcomes will be cognitive, emotional, food decision, and quality-of-life measures, as well as neuroimaging measures. Participants are evaluated at baseline (T0), after treatment (T1), and 12 months since baseline (T2).Discussion: Longitudinal study with active control group and 3 time points: baseline, immediately after treatment, and 1 year after baseline. Threefold treatment: executive function training, psychoeducation, and feedback on activity/sleep tracking. We will evaluate the transfer effects of the intervention, including emotional and functional outcomes, as well as the effects on neural plasticity by connectivity MRI.Trial registration: This project has been registered in ClinicalTrials.gov (trial registration number NCT03615274), August 3, 2018.

[spa] Los trastornos neurocognitivos (TNC) son un grupo heterogéneo de entidades clínicas cuya característica principal es el deterioro de todas o algunas de las capacidades funcionales y cognitivas previamente adquiridas por el individuo. Entre ellos encontramos el TNC mayor, tener múltiples etiologías, su causa más común es la enfermedad de Alzheimer (EA). Las marcas neuropatológicas distintivas de esta patología son la acumulación de placas de proteína β-amiloide y la de proteína tau hiperfosforilada que forma, entre otros, ovillos neurofibrilares. El primer lugar en el que se han detectado cambios en el proceso de formación de estas acumulaciones es el locus coeruleus (LC). Este es un núcleo situado en el tegmento pontino y es la mayor fuente de noradrenalina del sistema nervioso central. La integridad estructural del LC se ha investigado tanto en EA como en población en riesgo de desarrollar EA, sin embargo la preservación funcional de este núcleo a lo largo del continuum del TNC restaba sin estudiar. En esta tesis se presentan dos estudios centrados en la evaluación de la conectividad funcional entre el LC y diferentes regiones corticales y subcorticales, en grupos en riesgo de EA: hijos de pacientes con EA de inicio tardío (H-EAIT), individuos con deterioro cognitivo leve (DCL) y pacientes con trastorno depresivo mayor en edad avanzada (TDM); con el fin de comprobar si la medición de la funcionalidad del LC puede constituir un marcador precoz del comienzo del proceso neurodegenerativo en estas etapas tempranas del TNC. El primer estudio se centra en analizar la conectividad del LC en un grupo de H- EAIT, en comparación con sujetos sanos, durante una secuencia de resonancia magnética funcional en estado de reposo. En el segundo estudio se incluyó un grupo de DCL, de TDM y un grupo control; y se estudió la conectividad del LC durante la realización de una tarea oddball o de detección de estímulos, que permite estudiar la actividad neuronal del LC. En ambos casos correlacionamos los resultados de los análisis de neuroimagen con medidas clínicas, neuropsicológicas y conductuales, con el fin de caracterizar los patrones de alteración funcional y su asociación con posibles variables clínicas o cognitivas. Nuestros resultados apuntan a la disminución de la conectividad global del LC en el grupo de H-EAIT durante el estado de reposo y en TDM ante la detección de estímulos salientes (oddball). En ambos grupos observamos una disminución de la conectividad entre el LC y regiones del cerebelo posterior, implicadas en la modulación de la actividad de la red por defecto, de control ejecutivo y de saliencia. Igualmente, el grupo de TDM se caracteriza por presentar una disminución de la conectividad entre el LC y el giro fusiforme y la corteza cingulada anterior (CCA) ante los estímulos oddball. En los H-EAIT encontramos una correlación positiva entre la conectividad del LC-cerebelo posterior y medidas de recuerdo diferido que señala la implicación de este patrón en la función mnésica. Además, encontramos una asociación negativa entre la severidad de síntomas depresivos y la conectividad LC-cerebelo posterior. En los TDM, la conectividad LC-CCA se asocia negativamente con medidas de severidad y carga del trastorno depresivo, reflejo de la diferenciación entre dos subgrupos de pacientes depresivos: en episodio activo de depresión y en remisión. Los resultados de este estudio confirman que existen alteraciones en la conectividad del LC en estadios iniciales del TNC y en población con un riesgo incrementado de desarrollo de EA. Además, apoyan la modulación del LC de regiones corticales y subcorticales implicadas en función mnésica, atencional y regulación del arousal. Esta tesis supone un punto de base para la investigación más profunda del LC a lo largo del continuo del TNC, con el fin de poder desarrollar herramientas diagnósticas no invasivas que permitan detectar precozmente procesos neurodegenerativos y el desarrollo de tratamientos más personalizados., [eng] Neurocognitive disorders (NCD) are a heterogeneous group of clinical entities that share as a principal hallmark the impairment in all or some of the previously acquired cognitive and functional capacities. Among this group, Alzheimer’s disease (AD) is the most prevalent cause of NCD. The two biological hallmarks of AD are the β-amyloid plaques and hyperphosphorylated tau neurofibrillary tangles. The locus coeruleus (LC) nucleus, the major source of noradrenaline to the central nervous system, is the region where pathological changes start in the continuum of AD. Structural integrity of this nucleus has been investigated in AD and AD at-risk population, but its functional preservation remained unstudied. In this thesis, we present two studies focused in the LC functional connectivity. The first study included a group of offspring of late onset AD (O-LOAD) and we used a resting-state functional magnetic resonance (rs-fMRI). For the second study a group of mild cognitive impairment (MCI) and late-life major depressive disorder (MDD) were recruited and they performed an oddball task during the acquisition of functional magnetic resonance imaging (fMRI). In both works, neuroimaging results were correlated with clinical, neuropsychological and behavioral data. Firstly, we found that both O-LOAD and MDD groups showed a diminished LC global connectivity, and these groups also presented a lower connectivity between LC and posterior cerebellum regions, during rs-fMRI and the oddball task, respectively. Moreover, MDD showed a lower connectivity between LC and fusiform gyrus and anterior cingulate cortex (ACC) in front of oddball stimuli. In O-LOAD we observed a negative association between LC-posterior cerebellum and depression symptoms and memory measures; and MDD showed an inverse correlation between LC-CCA connectivity and depression severity and the number of episodes measures. Our results point to the presence of connectivity alterations of LC in early stages of NCD and in AD at- risk population. A deeper investigation of functional activity and connectivity of the LC through the NCD continuum would be useful for the development of non-invasive diagnostic tools and the development of personalized treatments.

International audience; Deep learning techniques for 3D brain vessel image segmentation have not been as successful as in the segmentation of other organs and tissues. This can be explained by two factors. First, deep learning techniques tend to show poor performances at the segmentation of relatively small objects compared to the size of the full image. Second, due to the complexity of vascular trees and the small size of vessels, it is challenging to obtain the amount of annotated training data typically needed by deep learning methods. To address these problems, we propose a novel annotation-efficient deep learning vessel segmentation framework. The framework avoids pixel-wise annotations, only requiring weak patch-level labels to discriminate between vessel and non-vessel 2D patches in the training set, in a setup similar to the CAPTCHAs used to differentiate humans from bots in web applications. The user-provided weak annotations are used for two tasks: 1) to synthesize pixel-wise pseudo-labels for vessels and background in each patch, which are used to train a segmentation network, and 2) to train a classifier network. The classifier network allows to generate additional weak patch labels, further reducing the annotation burden, and it acts as a second opinion for poor quality images. We use this framework for the segmentation of the cerebrovascular tree in Time-of-Flight angiography (TOF) and Susceptibility-Weighted Images (SWI). The results show that the framework achieves state-of-the-art accuracy, while reducing the annotation time by 77% w.r.t. learning-based segmentation methods using pixel-wise labels for training.

[spa] Esta tesis busca contribuir al estudio de los correlatos neurobiológicos de los síntomas obsesivos-compulsivos subclínicos (SOCS) con la esperanza de que sus hallazgos puedan ser eventualmente vinculados al desarrollo del trastorno obsesivo-compulsivo (TOC) y servir para identificar individuos vulnerables a desarrollarlo. Para ello, presentamos tres estudios realizados en una muestra de niños sanos, en los que utilizamos un enfoque multidimensional de los síntomas e implementamos distintas modalidades de neuroimagen, así como la combinación de datos de neuroimagen y genética, para aportar una visión integral de los factores que subyacen los SOCS. En el primer estudio, analizamos cambios volumétricos en la sustancia gris y blanca asociados a los SOCS y, en el segundo, evaluamos alteraciones de conectividad funcional asociadas a estos síntomas. En ambos, estudiamos el efecto de la edad y el sexo en dichas asociaciones. Finalmente, en el tercer estudio, identificamos nodos cerebrales en los que la conectividad funcional dinámica tiende a reverberar (atractores) y estudiamos su relación con los SOCS y su base genética. Nuestros resultados vinculan los SOCS totales con alteraciones en el circuito cortico-estriato- talamo-cortical (CSTC), hecho que concuerda con el modelo neurobiológico del TOC, pero también en regiones límbicas fuera de este circuito. También observamos que diferentes síntomas se asocian a alteraciones neurobiológicas distintas, lo que coincide con el modelo multidimensional del trastorno. Además, encontramos que los cambios estructurales relacionados con los síntomas de orden eran específicos de niños menores de 10 años, mientras que los cambios estructurales y funcionales asociados con los síntomas de acumulación eran específicos de sujetos de más de 10 años, especialmente en niñas, lo que sugiere que estos grupos demográficos pueden ser especialmente vulnerables a desarrollar dichos síntomas. Finalmente, identificamos polimorfismos en los genes GRM7 y GNAQ, implicados en la modulación de la neurotransmisión glutamatérgica, y PARVA, asociado a la regulación del citoesqueleto de actina, que predisponen al aumento de la capacidad atractora del hipocampo, vinculada a los SOCS totales. Del mismo modo, hallamos polimorfismos en los genes ATP1B1 y TESC, implicados en el mantenimiento de gradientes electroquímicos, que predisponen y protegen, respectivamente, al aumento de las propiedades atractoras del córtex parietal superior, relacionadas con los síntomas de orden., [eng] Obsessive-compulsive disorder (OCD) is characterized by the presence of intrusive thoughts that cause anxiety (obsessions) and lead to repetitive behaviors or mental acts aimed at reducing this anxiety (compulsions). Although OCD has been associated with alterations in the cortico-striato- thalamo-cortical (CSTC) circuits, the clinical heterogeneity of the disorder makes its neurobiology difficult to investigate. Given that the evidence suggests that different OCD symptoms could arise from distinct neurobiological alterations, the use of the multidimensional model, which allows characterizing different symptom dimensions, could help define more homogeneous subgroups of patients and identify more robust endophenotypes. However, the studies that have used this approach have given heterogeneous findings, possibly due to differences between patients in terms of medication, chronicity, or comorbidities. Notably, epidemiological studies show that obsessive-compulsive symptoms (OCSs) in childhood predispose to developing OCD in adulthood. Furthermore, the symptom profile observed in adulthood is consistent with the same symptom precursors in childhood. Therefore, OCSs could be studied in samples of healthy children from a dimensional perspective, thus avoiding the confounding factors typically observed in clinical samples. This thesis seeks to contribute to the study of the neurobiological correlates of OCSs in hopes that its findings may eventually be linked to the development of OCD and serve to characterize at-risk individuals. With that aim, we present three studies assessing a sample of healthy children, in which we use a multidimensional approach and implement different neuroimaging modalities and techniques, as well as the combination of neuroimaging and genetic data, to provide a comprehensive view of the factors underlying OCSs. In the first study, we analyzed volumetric changes in gray and white matter associated with total and dimensional OCSs whereas, in the second one, we evaluated functional connectivity alterations associated with total and dimensional OCSs. In both studies, we also assessed the effect of age and sex on these associations. In the third one, we identified brain nodes in which dynamic functional connectivity tends to reverberate (attractors) and we studied their relationship with total and dimensional OCSs. Furthermore, by combining neuroimaging and gene expression data, we identified genetic variants that moderated the relationship between OCSs and attractors. Our results link total OCSs with alterations in the CSTC circuit, which concurs with the prevailing neurobiological model of OCD, but also in limbic regions outside of this circuit. We also observed that different symptoms were associated with distinct neurobiological alterations: obsessing symptoms were related to alterations in limbic regions; doubt/checking symptoms were associated with changes in the ventral cognitive CSTC loop, the insula and regions mediating frontal processing; ordering symptoms were associated with alterations in the ventral cognitive and sensorimotor CSTC loops and the superior parietal cortex; hoarding symptoms were related to alterations in different CSTC loops, suggesting a larger CSTC alteration. Structural changes related to ordering symptoms were specific to boys under 10 years of age, whereas structural and functional changes associated with hoarding symptoms were specific to children over 10 years of age, especially to girls, which suggests that these demographic groups may be more susceptible to developing such symptoms. Finally, we identified polymorphisms in the GRM7 and GNAQ genes, involved in the modulation of glutamate neurotransmission, and in the PARVA gene, associated with the regulation of the actin cytoskeleton, that predisposed to an increase in the attractor properties of the hippocampus, linked to total OCSs. We also found polymorphisms in the ATP1B1 and TESC genes, involved in the maintenance of electrochemical gradients, that predisposed and protected, respectively, to an increase in the attractor properties of the superior parietal cortex, related to ordering symptoms.

A system to integrate microwave imaging with optical colonoscopy is presented. The overarching goal is to improve the prevention and early diagnosis of one of the main health and economic burdens of an increasingly aging population, i.e., colorectal cancer. For a colonoscopy, the gold standard for colorectal cancer diagnosis, 22% of polyps are not detected, and the risk of cancer after a negative colonoscopy can be up to 7.9%. To remedy this, a microwave imaging system able to generate an alarm when a polyp is detected is designed, manufactured and validated with a colon phantom composed of tissue-mimicking oil-gelatin materials reproducing the anatomy and dielectric properties of a human colon with a polyp. The acquisition was performed by a miniaturized ring-shaped switched array of 16 antennas attachable at the tip of a conventional colonoscope. This has been conceived to satisfy endoscopy size restrictions, patient safety and intercompatibility with current clinical practice. A Modified Monofocusing imaging method preceded by a previous frame average subtraction as a calibration technique shows a perfect detection of a 10-mm polyp (100% sensitivity and specificity) in the eight analyzed trajectories. The phantom results demonstrate the feasibility of the system in future preclinical trials. The work of Alejandra Garrido was supported by DIN2019-010857. The work of Roberto Sont, Ignasi Belda, and Marta Guardiola was supported in part by the European Union’s Horizon 2020 Research and Innovation Programme under Grant 960251 and in part by the European Institute of Innovation and Technology (EIT). The work of Jordi Romeu was supported by PID2019-107885GB-C31/AEI/10.13039/501100011033.

Treballs Finals de Grau d'Enginyeria Biomèdica. Facultat de Medicina i Ciències de la Salut. Universitat de Barcelona. Curs: 2020-2021. Directors: Roser Sala Llonch, Agnès Pérez Millan, The use of automated or semi-automated approaches based on imaging data has been suggested to support the diagnoses of some diseases. In this context, Machine Learning (ML) appears as a useful emerging tool for this purpose, allowing from feature extraction to automatic classification. Alzheimer Disease (AD) and Frontotemporal Dementia (FTD) are two common and prevalent forms of early-onset dementia with different, but partly overlapping, symptoms and brain patterns of atrophy. Because of the similarities, there is a need to establish an accurate diagnosis and to obtain good markers for prognosis. This work combines both supervised and unsupervised ML algorithms to classify AD and FTD. The data used consisted of gray matter volumes and cortical thicknesses (CTh) extracted from 3TT1 MRI of 44 healthy controls (HC, age: 57.8±5.4 years), 53 Early-Onset Alzheimer Disease patients (EOAD, age: 59.4±4.4 years) and 64 FTD patients (FTD, age: 64.4±8.8 years). A principal component analysis (PCA) of all volumes and thicknesses was performed and a number of principal components (PC) that accumulated at least 80% of the data variance were entered into a Support Vector Machine (SVM). Overall performance was assessed using a 5-fold crossvalidation...

Resumen: La incontinencia fecal es una patología con importantes implicaciones sociosanitarias, con un tratamiento complejo y no siempre satisfactorio, especialmente la incontinencia fecal idiopática. El sistema nervioso central regula los procesos de continencia y defecación. Los estudios de neuroimagen han demostrado ser útiles para caracterizar las áreas cerebrales que controlan el área anorrectal. A partir de un grupo de voluntarias sanas, se ha creado un modelo de caracterización de estas áreas cerebrales anorrectales, que podrá ser utilizado posteriormente para compararlo con un grupo de pacientes con incontinencia fecal idiopática, estudiando posibles diferencias y posibles opciones terapéuticas. La incontinència fecal és una patologia amb importants implicacions sociosanitàries, amb un tractament complex i no sempre satisfactori, especialment la incontinència fecal idiopàtica. El sistema nerviòs central regula els processos de continència i defecació. Els estudis de neuroimatge han demostrat ser útils per caracteritzar les àrees cerebrals que controlen l'àrea anorrectal. A partir d'un grup de voluntàries sanes, s'ha creat un model de caracterització d'aquestes àrees cerebrals anorrectals, que podrà ser utilitzat posteriorment per comparar-lo amb un grup de pacients amb incontinència fecal idiopàtica, estudiant possibles diferències i possibles opcions terapèutiques.

[eng] Time-of-Flight Positron Emission Tomography (TOF-PET) scanners demand fast and efficient photo-sensors and scintillators coupled to fast readout electronics. Nowadays, there are two main configurations regarding the scintillator crystal geometry: the segmented or pixelated and the monolithic approach. Depending on the cost, spatial resolution and time requirements of the PET module, one can choose between one or another. The pixelated crystal is the most extensive configuration on TOF-PET scanners as the coincidence time resolution is better compared to the monolithic. On the contrary, monolithic scintillator crystals for Time-of-Flight Positron Emission Tomography (ToF-PET) are increasing in popularity this last years due to their performance potential and price in front of the commonly used segmented crystals. On one hand, monolithic blocks allows to determine 3D information of the gamma-ray interaction inside the crystal, which enables the possibility to correct the parallax error (radial astigmatism) at off-center positions within a PET scanner, resulting in an improvement of the spatial resolution of the device. On the other hand, due to the simplicity during the crystal manufacturing process as well as for the detector design, the price is reduced compared to a regular pixelated detector. The thesis starts with the use of HRFlexToT, an ASIC developed in this group, as the readout electronics for measurements with single pixelated crystals coupled to different SiPMs. These measurements show an energy linearity error of 3% and an energy resolution below 10% of the 511 keV photopeak. Single Photon Time Resolution (SPTR) measurements performed using an FBK SiPM NUV-HD (4 mm x 4 mm pixel size) and a Hamamatsu SiPM S13360-3050CS gave a 141 ps and 167 ps FWHM respectively. Coincidence Time Resolution (CTR) measurements with small cross-section pixelated crystals (LFS crystal, 3 m x 3 mm x 20 mm ) coupled to a single Hamamatsu SiPM S13360-3050CS provides a CTR of 180 ps FWHM. Shorter crystals (LSO:Ce Ca 0.4%) coupled to a Hamamatsu S13360-3050CS SiPM or FBK-NUVHD yields a CTR of 117 ps and 119 ps respectively. Then, the results with different monolithic crystals and SiPM sensors using HRFlexToT ASIC will be presented. A Lutetium Fine Silicate (LFS) of 25 mm x 25 mm x 20 mm, a small LSO:Ce Ca 0.2% of 8 mm x 8 mm x 5 mm and a Lutetium-Yttrium Oxyorthosilicate (LYSO) of 25 mm x 25 mm x 10 mm has been experimentally tested. After subtracting the TDC contribution (82 ps FWHM), a coincidence time resolution of 244 ps FWHM for the small LFS crystal and 333 ps FWHM for the largest LFS one is reported. Additionally, a novel time calibration correction method for CTR improvement that involves a pico-second pulsed laser will be detailed. In the last part of the dissertation, a new developed simulation framework that will enable the cross-optimization of the whole PET system will be explained. It takes into consideration the photon physics interaction in the scintillator crystal, the sensor response (sensor size, pixel pitch, dead area, capacitance) and the readout electronics behavior (input impedance, noise, bandwidth, summation). This framework has allowed us to study a new promising approach that will help reducing the CTR parameter by segmenting a large area SiPM into "m" smaller SiPMs and then summing them to recover all the signal spread along these smaller sensors. A 15% improvement on time resolution is expected by segmenting a 4 mm x 4 mm single sensor into 9 sensors of 1.3 mm x 1.3 mm with respect to the case where no segmentation is applied., [cat] Aquesta tesi tenia com a objectiu la fabricació i avaluació d'un prototip per a detecció de fotons gamma en aplicació per imatge mèdica, més concretament en Tomografia per Emissió de Positrons amb mesura de temps de vol (TOF-PET). L'avaluació del mòdul va començar fent una caracterització completa del chip (ASIC) anomenat HRFlexToT, una versió nova i millorada de l'antic chip FlexToT, desenvolupat i fabricat pel grup de la Unitat Tecnològica del ICC de la Universitat de Barcelona. Aquesta avaluació inicial del chip compren des de la comprovació de les funcionalitats bàsiques fins a la generació d'un test automàtic per generar les gràfiques de linealitat corresponents durant el test elèctric. Un cop donat per bo, es va muntar en una placa demostrada, també ideada per l'equip d'enginyers del grup, i ja quedava llesta per realitzar les mesures pertinents. Tot seguit, es varen realitzar les mesures òptiques, que incloïa mesures de Singe Photon Time Resolution (SPTR) i de Coincidence Time Resolution (CTR). Aquest valors actuen com a figures de mèrit a l'hora de comparar les prestacions amb d'altres ASICs competidors del HRFlexToT. Es van obtenir valors de 60 ps de resposta pel que respecta al SPTR i de 115 ps de CTR en cristalls segmentats, una millora entorn del 20-30% respecte a la versió predecessora del chip. Aquests valors mostren ser el límit de l'estat de l'art actual i amb aquesta idea es van començar a fer altres mesures, en aquest cas amb cristall monolítics, blocs grans llegits per diversos fotosensors de les empreses Hamamatsu i FBK. Per altra banda, es va provar el funcionament del ASIC en configuració anomenada monolítica, on el cristall centellejador s'utilitza en blocs grans en coptes d’emprar cristalls segmentats, això abarateix el cost total del detector. Aquesta configuració degrada les propietats de CTR, un paràmetre crític a l'hora de tenir un producte bo i eficient. S’han obtingut mesures de 250 ps de CTR amb aquesta configuració, d’on es pot dir que l’HRFlexToT es trobar a l’estat de l’art de la tecnologia electrònica dedicada a TOF-PET amb cristalls segmentats i monolítics. Finalment, es va desenvolupar una nova eina simulació que consisteix en un sistema híbrid entre un simulador físic i un electrònic per tal de tenir una idea del comportament complet del mòdul detector. Una solució que ningú havia provat fins ara o que no es pot trobar en la literatura.

Background One common denominator to the clinical phenotypes of borderline personality disorder (BPD) and major depressive disorder (MDD) is emotion regulation impairment. Although these two conditions have been extensively studied separately, it remains unclear whether their emotion regulation impairments are underpinned by shared or distinct neurobiological alterations. Methods We contrasted the neural correlates of negative emotion regulation across an adult sample of BPD patients (n = 19), MDD patients (n = 20), and healthy controls (HCs; n = 19). Emotion regulation was assessed using an established functional magnetic resonance imaging cognitive reappraisal paradigm. We assessed both task-related activations and modulations of interregional connectivity. Results When compared to HCs, patients with BPD and MDD displayed homologous decreased activation in the right ventrolateral prefrontal cortex (vlPFC) during cognitive reappraisal. In addition, the MDD group presented decreased activations in other prefrontal areas (i.e., left dorsolateral and bilateral orbitofrontal cortices), while the BPD group was characterized by a more extended pattern of alteration in the connectivity between the vlPFC and cortices of the visual ventral stream during reappraisal. Conclusions This study identified, for the first time, a shared neurobiological contributor to emotion regulation deficits in MDD and BPD characterized by decreased vlPFC activity, although we also observed disorder-specific alterations. In MDD, results suggest a primary deficit in the strength of prefrontal activations, while BPD is better defined by connectivity disruptions between the vlPFC and temporal emotion processing regions. These findings substantiate, in neurobiological terms, the different profiles of emotion regulation alterations observed in these disorders.

The Jake Mandell Tone Deaf Test (JMT) is an online measure designed to evaluate pitch discrimination in the normal population. The test has been available online since 2006, but no psychometric data have been published. The present research was designed to provide validity and reliability data for this test. Study 1 was conducted in university students, and the performance on the JMT was compared to the AMUSIA online screening tests. Correlations were high with the subtests of the AMUSIA online screening tests, but especially with the Tone discrimination test. Study 2 was conducted in children, and the JMT was compared to different tests of musical abilities. The JMT showed a good relationship with the Tone Discrimination and Rhythm Imitation subtest, but especially with the global score of musical abilities. In addition, the test showed good one-year test-retest reliability. Finally, external validation of the JMT was obtained by noting that musicians obtained higher scores than non-musicians. Discussion is focused on the possibility of using the JMTas a rapid screening measure of individual differences in pitch discrimination in the normal population. El test de Jake Mandell (JMT) es un test online diseñado para evaluar la discriminación tonal en la población normal. El test está disponible online desde 2006, pero no se han publicado datos psicométricos. La presente investigación fue diseñada para proporcionar datos de validez y fiabilidad para este test. El estudio 1 se realizó en estudiantes universitarios, y el rendimiento en el JMT se comparó con las pruebas de detección online denominadas AMUSIA. Las correlaciones fueron altas con los subtests de AMUSIA, pero especialmente con el de discriminación de tonos. El estudio 2 se realizó en niños, y el JMT se comparó con diferentes pruebas de habilidades musicales. El JMT mostró una buena relación con el subtest de discriminación de tono e imitación de ritmo, pero especialmente con la puntuación global de las habilidades musicales. Además, el test mostró una buena fiabilidad test-retest al cabo de un año. Finalmente, la validación externa del JMT se obtuvo al señalar que los músicos obtuvieron puntuaciones más altas que los no músicos. La discusión se centra en la posibilidad de utilizar el JMT como una medida de detección rápida de las diferencias individuales en la discriminación de tono en la población normal. O teste de surdez de tom de Jake Mandell (JMT) é uma medida on-line projetada para avaliar a discriminação de tom na população normal. O teste está disponível online desde 2006, mas nenhum dado psicométrico foi publicado. A presente pesquisa foi projetada para fornecer dados de validade e confiabilidade para este teste. O estudo 1 foi realizado em estudantes universitários, e o desempenho no JMT foi comparado aos testes de triagem on-line da AMUSIA. As correlações foram altas com os subtestes dos testes de triagem on-line da AMUSIA, mas principalmente com o teste de discriminação de tons. O estudo 2 foi realizado em crianças e o JMT foi comparado a diferentes testes de habilidades musicais. O JMT mostrou uma boa relação com o subteste Discriminação de Tom e Imitação de Ritmo, mas principalmente com o escore global de habilidades musicais. Além disso, o teste mostrou boa confiabilidade teste-reteste de um ano. Finalmente, a validação externa do JMT foi obtida observando que os músicos obtiveram pontuações mais altas que os não músicos. A discussão está focada na possibilidade de usar o JMT como uma medida de triagem rápida das diferenças individuais na discriminação de arremesso na população normal. &nbsp

Treballs finals del Màster de Fonaments de Ciència de Dades, Facultat de matemàtiques, Universitat de Barcelona, Any: 2020, Tutor: Eloi Puertas i Prats i Joaquim Raduà, [en] This master’s thesis seeks to review and objectively evaluate the current white matter hyperintensities (WMH) automatic segmentation methods published journals. To this end, the methods have been systematically searched in scientific databases, and those meeting inclusion criteria have been evaluated. The evaluation has consisted in applying the method to detect WMH in our dataset of patients with bipolar disorder and healthy controls, in which an experienced neuroradiologist had manually coded all WMH. After the systematic search, we selected all available methods that were ready for use with standard MRI data by a standard user. Four methods met these criteria. We then applied these methods to detect WMH in our dataset, and compared the results with the neuroradiologist-based ground truth deriving several evaluation metrics. This master’s thesis also include a discussion section, in which we compare the results of our evaluations with the results of the WMH Segmentation Challenge held in 2017, which included substantially different datasets. The most relevant conclusion of this master’s thesis is that no method seems to be accurate enough for clinical implementation, although the low performance of the methods may be related to the differences between our data and the data that were used to train them. Besides, realizing the huge improvement made in the field during the last few years after the appearance of deep neural networks, we anticipate that a method with sufficient accuracy might be available soon. The codes used to obtain the results and graphs displayed in this project together with some guidelines to run them are available through PFM-WMH 1.

International audience; BACKGROUND: MRI studies show that obese adults have reduced grey matter (GM) and white matter (WM) tissue density as well as altered WM integrity. Bariatric surgery can lead to substantial weight loss and improvements in metabolic parameters, but it remains to be examined if it induces structural brain changes. The aim of this study was to characterize GM and WM density changes measured with MRI in a longitudinal setting following sleeve gastrectomy, and to determine whether any changes are related to inflammation and cardiometabolic blood markers. METHODS: 29 participants with obesity (age: 45.9 ​± ​7.8 years) scheduled to undergo sleeve gastrectomy were recruited. High-resolution T1-weighted anatomical images were acquired 1 month prior to as well as 4 and 12 months after surgery. GM and WM densities were quantified using voxel-based morphometry (VBM). Circulating lipid profile, glucose, insulin and inflammatory markers (interleukin-6, C-reactive protein and lipopolysaccharide-binding protein) were measured at each time point. A linear mixed effect model was used to compare brain changes before and after SG, controlling for age, sex, initial BMI and diabetic status. To assess the associations between changes in adiposity, metabolism and inflammation and changes in GM or WM density, the mean GM and WM densities were extracted across all the participants using atlas-derived regions of interest, and linear mixed-effect models were used. RESULTS: As expected, weight, BMI, waist circumference and neck circumference significantly decreased after SG compared with baseline (p ​\textless ​0.001 for all). A widespread increase in WM density was observed after surgery, particularly in the cerebellum, brain stem, cerebellar peduncle, cingulum, corpus callosum and corona radiata (p ​\textless ​0.05, after FDR correction). Significant increases in GM density were observed 4 months after SG compared to baseline in several brain regions such as the bilateral occipital cortex, temporal cortex, postcentral gyrus, cerebellum, hippocampus and insula as well as right fusiform gyrus, right parahippocampal gyrus, right lingual gyrus and right amygdala. These GM and WM increases were more pronounced and widespread after 12 months and were significantly associated with post-operative weight loss and the improvement of metabolic alterations. A linear mixed-effect model also showed associations between post-operative reductions in lipopolysaccharide-binding protein, a marker of inflammation, and increased WM density. To confirm our results, we tested whether the peak of each significant region showed BMI-related differences in an independent dataset (Human Connectome Project). We matched a group of individuals who were severely obese with a group of individuals who were lean for age, sex and ethnicity. Severe obesity was associated with reduced WM density in the brain stem and cerebellar peduncle as well as reduced GM density in cerebellum, regions that significantly changed after surgery (p ​\textless ​0.01 for all clusters). CONCLUSIONS: Bariatric surgery-induced weight loss and improvement in metabolic alterations is associated with widespread increases in WM and GM densities. These post-operative changes overlapped with baseline brain differences between participants who were severely obese and those who were normal-weight in a separate dataset, which may suggest a recovery of WM and GM alterations after bariatric surgery.

Background: Artificial intelligence (AI) is rapidly infiltrating multiple areas in medicine, with gastrointestinal endoscopy paving the way in both research and clinical applications. Multiple challenges associated with the incorporation of AI in endoscopy are being addressed in recent consensus documents. Objectives: In the current paper, we aimed to map future challenges and areas of research for the incorporation of AI in capsule endoscopy (CE) practice. Design: Modified three-round Delphi consensus online survey. Methods: The study design was based on a modified three-round Delphi consensus online survey distributed to a group of CE and AI experts. Round one aimed to map out key research statements and challenges for the implementation of AI in CE. All queries addressing the same questions were merged into a single issue. The second round aimed to rank all generated questions during round one and to identify the top-ranked statements with the highest total score. Finally, the third round aimed to redistribute and rescore the top-ranked statements. Results: Twenty-one (16 gastroenterologists and 5 data scientists) experts participated in the survey. In the first round, 48 statements divided into seven themes were generated. After scoring all statements and rescoring the top 12, the question of AI use for identification and grading of small bowel pathologies was scored the highest (mean score 9.15), correlation of AI and human expert reading-second (9.05), and real-life feasibility-third (9.0). Conclusion: In summary, our current study points out a roadmap for future challenges and research areas on our way to fully incorporating AI in CE reading.