Search

The Information Processing in Medical Imaging (IPMI) Conference series has a longstanding history and tradition. Since 1969, the conference has taken place on either side of the Atlantic and was organised in Asia for the first time in 2019. IPMI 2023 (www.ipmi2023.org), the twenty-eighth conference in the series, was the first in Latin America. The conference was held from June 18th to 23rd at Huinid Pioneros Hotel in San Carlos de Bariloche, Argentina. Through the years, IPMI became a biennial beacon of foundational computational methods for medical imaging. IPMI's distinctive small-audience format enables extensive discussion time for selected oral presentations. The conference has a retreat-like ambience, which promotes a more intimate exchange and building personal and mentoring relationships across generations from our community. IPMI 2023 featured traditional study groups, small groups of delegates who scrutinise manuscripts before their oral presentation. This leads to thorough debates enriching the authors and broader audience. Historically, discussions on each paper kick-start with early career delegates in each study group to elicit inclusive participation. Thus, every session becomes a forum for sharing, encouraging, and challenging new ideas. For this year’s edition, we received 240 submissions, of which 169 papers went through the double-blind peer review system being evaluated by at least three reviewers each. Following the reviewing process, each manuscript having at least one weak accept in its reviews was evaluated by two or more members of the Paper Selection Committee and subsequently discussed at the paper selection meeting. Finally, 63 submissions were accepted for publication and presentation at the conference, resulting in a 26% acceptance rate. At the conference, 23 oral and 40 poster presentations were given on topics ranging from classical registration and segmentation to novel machine-learning approaches for self-supervised learning and domain adaptation. Three excellent keynote speakers joined us virtually. Prof Gitta Kutyniok (Ludwig Maximilians Universität München), Dr Petar Veličković (DeepMind), and Prof Jong Chul Ye (Korea Advanced Institute of Science & Technology) enlightened us with various talks on the foundations of modern information processing ranging from the mathematics of machine learning to geometric deep learning and diffusion models in inverse problems in imaging. The François Erbsmann Prize is awarded every edition to a young scientist of age 35 or below, the first author of a paper, giving their first oral presentation at IPMI. The prize was awarded to one of the 18 eligible oral presenters. Additionally, awards were given for the best poster presentations. The IPMI 2023 Organising Team wants to express special gratitude to the Paper Selection Committee and the 186 reviewers who did an outstanding job reviewing submissions, consolidating decisions and designing the programme. We thank Albert Chung, Archana Venkataraman, Diana Mateus, Herve Lombaert, Ipek Oguz, Juan Eugenio Iglesias, Kaleem Siddiqi, Kevin Zhou, and Xavier Pennec, who decided on each paper based on the available reviews and thorough committee discussions. We also thank Andreas Maier, Ipek Oguz and Archana Venkattakaraman for their leadership in organising the Study Groups and all the study group leaders. We also express our gratitude to Haoran Dou, Azade Farshad and Yan Xia for their tireless work as Executive Assistants to the chairs. We are also indebted to Enzo Ferrante and Ariel Curiale for their engagement in IPMI 2023 from its conception in IPMI 2023 and for their support with all the challenges of organising IPMI across the Atlantic. Finally, we are indebted to the team of Dekon Group, particularly to Mustafa Bay, Meltem Çakmak, and Mehmet Eldegez for their organisational support. This conference would not have been possible without the help of such a great team of colleagues. One of the defining priorities throughout the organisation was to maximise diversity across career stages, gender and geographical spread in our Organising Committee, Paper Selection Committee, Study Groups, and generally wherever possible. We have been delighted to organise IPMI 2023 and hope that the scientific and social activities programme will make this conference an unforgettable experience. We wish that IPMI 2023, surrounded by the Andes mountain range and unique Patagonian lakes and forests, will be an event conducive to scientific exchange and collegial sharing and challenging of ideas, which will contribute to international cooperation, scientific progress and understanding, and to nurturing of the next generation of researchers in information processing in medical imaging. General Chairs Alejandro Frangi Marleen de Bruijne Program Chairs Demian Wassermann Nassir Navab

Background and objective: Klinefelter Syndrome (KS) is the most common sex chromosome aneuploidy (47, XXY) and cause of male hypergonadotropic hypogonadism. It is characterized by an extreme clinical heterogeneity in presentation, including infertility, hypogonadism, language delay, metabolic comorbidities, and neurocognitive and psychiatric disorders. Since testosterone is known to have organizational, neurotrophic and neuroprotective effects on brain, the condition of primary hypogonadism could play a role. Moreover, given that KS subjects have an additional X, genes on the extra-chromosome could also exert a significant impact. The aim of this narrative review is to analyze the available literature on the relationship between KS and neuropsychiatric disorders.Methods:To extend to the best of published literature on the topic, appropriate keywords and MeSH terms were identified and searched in Pubmed. Finally, references of original articles and reviews were examined.Results:Both morphological and functional studies focusing on the brain showed that there were important differences in brain structure of KS subjects. Different psychiatric disorders such as Schizophrenia, autism, attention deficit hyperactivity disorder, depression and anxiety were frequently reported in KS patients according to a broad spectrum of phenotypes. T supplementation (TRT) was not able to improve the psychotic disorders in KS men with or without overt hypogonadism.Conclusion:Although the risk of psychosis, depression and autism is increased in subjects with KS, no definitive evidence has been found in studies aiming at identifying the relationship between aneuploidy, T deficit and the risk of psychiatric and cognitive disorders in subjects affected by KS.

Dynamic testing represents the mainstay in the differential diagnosis of ACTH-dependent Cushing's syndrome. However, in case of undetectable or detectable lesion 6 mm on MRI, bilateral inferior petrosal sinus sampling (BIPSS) is suggested by current guidelines. Aim of this study was to analyze the performance of CRH, desmopressin and high-dose dexamethasone suppression test (HDDST) in the differential diagnosis of ACTH-dependent Cushing's syndrome as well as the impact of invasive and noninvasive tests on surgical outcome in patients affected by Cushing's disease (CD).Retrospective analysis on 148 patients with CD and 26 patients with ectopic ACTH syndrome.Among CD patients, negative MRI/lesion 6 mm was detected in 97 patients (Group A); 29 had a 6-10 mm lesion (Group B) and 22 a macroadenoma (Group C). A positive response to CRH test, HDSST and desmopressin test was recorded in 89.4%, 91·4% and 70.1% of cases, respectively. Concordant positive response to both CRH/HDDST and CRH/desmopressin tests showed a positive predictive value of 100% for the diagnosis of CD. Among Group A patients with concordant CRH test and HDDST, no difference in surgical outcome was found between patients who performed BIPSS and those who did not (66.6% vs 70.4%, p = 0.78).CRH, desmopressin test and HDDST have high accuracy in the differential diagnosis of ACTH-dependent CS. In patients with microadenoma 6 mm or non-visible lesion, a concordant positive response to noninvasive tests seems sufficient to diagnose CD, irrespective of MRI finding. In these patients, BIPSS should be reserved to discordant tests.

Purpose: The prevalence and the etiopathogenesis of thyroid dysfunctions in Klinefelter syndrome (KS) are still unclear. The primary aim of this study was to evaluate the pathogenetic role of hypogonadism in the thyroid disorders described in KS, with the scope to distinguish between patients with KS and hypogonadism due to other causes (Kallmann syndrome, idiopathic hypogonadotropic hypogonadism, iatrogenic hypogonadism and acquired hypogonadotropic hypogonadism after surgical removal of pituitary adenomas) called non-KS. Therefore, we evaluated thyroid function in KS and in non-KS hypogonadal patients. Methods: This is a case–control multicentre study from KING group: Endocrinology clinics in university-affiliated medical centres. One hundred and seventy four KS, and sixty-two non-KS hypogonadal men were enrolled. The primary outcome was the prevalence of thyroid diseases in KS and in non-KS. Changes in hormonal parameters were evaluated. Exclusion criterion was secondary hypothyroidism. Analyses were performed using Student’s t test. Mann–Whitney test and Chi-square test. Results: FT4 was significantly lower in KS vs non-KS. KS and non-KS presented similar TSH and testosterone levels. Hashimoto’s thyroiditis (HT) was diagnosed in 7% of KS. Five KS developed hypothyroidism. The ratio FT3/FT4 was similar in both groups. TSH index was 1.9 in KS and 2.3 in non-KS. Adjustment for differences in age, sample size and concomitant disease in multivariate models did not alter the results. Conclusions: We demonstrated in KS no etiopathogenic link to hypogonadism or change in the set point of thyrotrophic control in the altered FT4 production. The prevalence of HT in KS was similar to normal male population, showing absence of increased risk of HT associated with the XXY karyotype.

Decompressive craniectomy (DC) is a common surgical procedure consisting of the removal of a portion of the skull that is performed after incidents such as stroke, traumatic brain injury (TBI) or other events that could result in acute subdural hemorrhage and/or increasing intracranial pressure. In these cases, CT scans are obtained to diagnose and assess injuries, or guide a certain therapy and intervention. We propose a deep learning based method to reconstruct the skull defect removed during DC performed after TBI from post-operative CT images. This reconstruction is useful in multiple scenarios, e.g. to support the creation of cranioplasty plates, accurate measurements of bone flap volume and total intracranial volume, important for studies that aim to relate later atrophy to patient outcome. We propose and compare alternative self-supervised methods where an encoder-decoder convolutional neural network (CNN) estimates the missing bone flap on post-operative CTs. The self-supervised learning strategy only requires images with complete skulls and avoids the need for annotated DC images. For evaluation, we employ real and simulated images with DC, comparing the results with other state-of-the-art approaches. The experiments show that the proposed model outperforms current manual methods, enabling reconstruction even in highly challenging cases where big skull defects have been removed during surgery.