Search

Highlights: What are the main findings? This multicenter retrospective study, encompassing two hospital sites for training and leveraging data from forty-seven different hospitals for testing, aimed to develop machine learning models utilising both clinical and radiomics features (individually and in combination) extracted from pre-treatment CT scans. The objective was to predict PFS and OS in patients with PDAC. Among the various models developed, the model based on the combination of clinical and CT-derived tumour radiomics features demonstrated superior performance for OS with a C-index of 0.72 in the testing dataset. Conversely, the model based solely on clinical features excelled in predicting PFS, achieving a C-index of 0.70 in the testing dataset. What is the implication of the main finding? The analysis further highlighted the high relevance of clinical features both for OS and PFS outcomes in patients with PDAC. Purpose. This multicenter retrospective study aims to identify reliable clinical and radiomic features to build machine learning models that predict progression-free survival (PFS) and overall survival (OS) in pancreatic ductal adenocarcinoma (PDAC) patients. Methods. Between 2010 and 2020 pre-treatment contrast-enhanced CT scans of 287 pathology-confirmed PDAC patients from two sites of the Hopital Universitaire de Bruxelles (HUB) and from 47 hospitals within the HUB network were retrospectively analysed. Demographic, clinical, and survival data were also collected. Gross tumour volume (GTV) and non-tumoral pancreas (RPV) were semi-manually segmented and radiomics features were extracted. Patients from two HUB sites comprised the training dataset, while those from the remaining 47 hospitals of the HUB network constituted the testing dataset. A three-step method was used for feature selection. Based on the GradientBoostingSurvivalAnalysis classifier, different machine learning models were trained and tested to predict OS and PFS. Model performances, SCOPUS: ar.j, info:eu-repo/semantics/published

Définir précisément l’agressivité d’une maladie est la clé pour une médecine personnalisée et de précision. La biopsie de prostate est un examen fondamental car il est le seul à confirmer la présence d’un cancer et permet de déterminer son grade d’agressivité qui est connu pour être le facteur pronostic le plus important. La fiabilité des résultats obtenus est donc primordiale car cela influencera le choix de traitement proposé. En 2019, la Société Européenne d’Urologie a recommandé l’implémentation de l’imagerie par résonance magnétique (IRM) avant la biopsie de prostate étant donné l’amélioration de la détection de cancers significatifs grâce aux biopsies ciblées et la diminution de la détection de cancers non-significatifs. Cependant, plusieurs questions ont été soulevées comme le schéma de biopsie à réaliser et l’impact sur les modèles prédictifs qui ont été initialement développés à l’époque de biopsies systématisées. L’hypothèse de cette thèse de doctorat est que l’intégration de la biopsie ciblée par IRM dans le processus diagnostic améliore la prise en charge du cancer de prostate.Dans un premier temps, nous avons défini un schéma de biopsie personnalisé après analyse tridimensionnelle de la prostate et des trajets des aiguilles de biopsie en utilisant un algorithme d’intelligence artificielle. Ensuite nous avons évalué les modèles statistiques prédisant le grade et le stade de la maladie en exploitant une des plus grandes bases de données européennes multicentriques de patients diagnostiqués par biopsies ciblées et opérés par prostatectomie radicale. Nous avons démontré la plus-value de l’IRM et des biopsies ciblées qui permettent de mieux définir l’avancée de la maladie. Par la suite, nous avons investigué leur impact sur la définition des groupes de risque qui servent à guider le choix de traitement. Nous avons remarqué la faible fiabilité des modèles de classification actuels, entrainant un risque de traitement inapproprié. Enfin, nous avons cherché à sa, Doctorat en Sciences médicales (Médecine), info:eu-repo/semantics/nonPublished

Peritoneal carcinomatosis (PC) refers to malignant epithelial cells that spread to the peritoneum, principally from abdominal malignancies. Until recently, PC prognosis has been considered ill-fated, with palliative therapies serving as the only treatment option. New locoregional treatments are changing the outcome of PC, and imaging modalities have a critical role in early diagnosis and disease staging, determining treatment decision making strategies. The aim of this review is to provide a practical approach for detecting and characterizing peritoneal deposits in cross-sectional imaging modalities, taking into account their appearances, including the secondary complications, the anatomical characteristics of the peritoneal cavity, together with the differential diagnosis with other benign and malignant peritoneal conditions. Among the cross-sectional imaging modalities, computed tomography (CT) is widely available and fast; however, magnetic resonance (MR) performs better in terms of sensitivity (92% vs. 68%), due to its higher contrast resolution. The appearance of peritoneal deposits on CT and MR mainly depends on the primary tumour histology; in case of unknown primary tumour (3-5% of cases), their behaviour at imaging may provide insights into the tumour origin. The timepoint of tumour evolution, previous or ongoing treatments, and the peritoneal spaces in which they occur also play an important role in determining the appearance of peritoneal deposits. Thus, knowledge of peritoneal anatomy and fluid circulation is essential in the detection and characterisation of peritoneal deposits. Several benign and malignant conditions show similar imaging features that overlap those of PC, making differential diagnosis challenging. Knowledge of peritoneal anatomy and primary tumour histology is crucial, but one must also consider clinical history, laboratory findings, and previous imaging examinations to achieve a correct diagnosis. In conclusion, to correctly diagnose, SCOPUS: re.j, info:eu-repo/semantics/published

Multidisciplinary team (MDT) meetings aim to optimize patient management. We evaluated the impact of MDT discussions on the management and diagnosis of focal pancreatic lesions in a single tertiary center., SCOPUS: ar.j, info:eu-repo/semantics/published

Diffusion-weighted magnetic resonance imaging (MRI) is a proven, quick, and noninvasive tech-nique which requires no ionizing radiation or contrast media administration and is able to de-tect, characterize and stage malignant tumors, and to monitor treatment response. Over the two last decades, many clinical applications of diffusion-weighted imaging (DWI) have emerged in the field of women’s oncology. This technique provides both qualitative and quanti-tative information offering a promising new tumor biomarker in clinical routine. This thesis work was therefore conducted to demonstrate the critical role of DWI in the man-agement of adnexal masses and breast cancers. The dissertation is divided into 5 chapters. Chapter 1 presents the general background. Chapter 2 is dedicated to the contribution of DWI in assessing adnexal masses. Chapter 3 is focused on the role of DWI in the evaluation of breast cancer response to neoadjuvant treatments with emphasis on the early response to neoadju-vant chemotherapy (NAC) and to neoadjuvant radiotherapy. Chapter 4 explores the performance of whole-body MRI (WB-MRI) including DWI in the initial staging of breast cancer patients at high risk of metastasis. Chapter 5 contains the discussion, future perspectives, and general con-clusion.Initially, we validated prospectively the ADNEX MR scoring system in the assessment of inde-terminate ovarian and adnexal masses with ultrasound (US) in 323 patients. We also evaluated a new, modified ADNEX MR scoring system that incorporates DWI including apparent diffusion coefficient (ADC) mapping. It showed an improvement of the classical ADNEX MR scoring sys-tem in diagnosing lesions with a score of 4 or 5 when adding DWI data analysis including ADC mapping, from 95.5% sensitivity and 86.6% specificity to 95.7% and 93.3%, respectively.Afterwards, we evaluated the added value of quantitative DWI in the diagnosis of indetermi-nate adnexal masses with US in a cohort of 163 patients with 201, Doctorat en Sciences médicales (Médecine), info:eu-repo/semantics/nonPublished

Objective: Soft-tissue sarcomas (STSs) of the extremities are a group of malignancies arising from the mesenchymal cells that may develop distant metastases or local recurrence. In this article, we propose a novel methodology aimed to predict metastases and recurrence risk in patients with these malignancies by evaluating magnetic resonance radiomic features that will be formally verified through formal logic models. Materials and Methods: This is a retrospective study based on a public dataset evaluating MRI scans T2-weighted fat-saturated or short tau inversion recovery and patients having “metastases/local recurrence” (group B) or “no metastases/no local recurrence” (group A) as clinical outcomes. Once radiomic features are extracted, they are included in formal models, on which is automatically verified the logic property written by a radiologist and his computer scientists coworkers. Results: Evaluating the Formal Methods efficacy in predicting distant metastases/local recurrence in STSs (group A vs group B), our methodology showed a sensitivity and specificity of 0.81 and 0.67, respectively; this suggests that radiomics and formal verification may be useful in predicting future metastases or local recurrence development in soft tissue sarcoma. Discussion: Authors discussed about the literature to consider Formal Methods as a valid alternative to other Artificial Intelligence techniques. Conclusions: An innovative and noninvasive rigourous methodology can be significant in predicting local recurrence and metastases development in STSs. Future works can be the assessment on multicentric studies to extract objective disease information, enriching the connection between the radiomic quantitative analysis and the radiological clinical evidences., SCOPUS: ar.j, info:eu-repo/semantics/published

Highlights: What are the main findings? Both the model utilizing edema-related features and the model utilizing mass-related features demonstrated promising results in predicting the occurrence of lung metastases, with similar performances. What is the implication of the main finding? The findings suggest that the analysis of radiomic features extracted exclusively from edema can offer valuable insights into the prediction of lung metastases. Introduction: This study aimed to evaluate whether radiomic features extracted solely from the edema of soft tissue sarcomas (STS) could predict the occurrence of lung metastasis in comparison with features extracted solely from the tumoral mass. Materials and Methods: We retrospectively analyzed magnetic resonance imaging (MRI) scans of 32 STSs, including 14 with lung metastasis and 18 without. A segmentation of the tumor mass and edema was assessed for each MRI examination. A total of 107 radiomic features were extracted for each mass segmentation and 107 radiomic features for each edema segmentation. A two-step feature selection process was applied. Two predictive features for the development of lung metastasis were selected from the mass-related features, as well as two predictive features from the edema-related features. Two Random Forest models were created based on these selected features; 100 random subsampling runs were performed. Key performance metrics, including accuracy and area under the ROC curve (AUC), were calculated, and the resulting accuracies were compared. Results: The model based on mass-related features achieved a median accuracy of 0.83 and a median AUC of 0.88, while the model based on edema-related features achieved a median accuracy of 0.75 and a median AUC of 0.79. A statistical analysis comparing the accuracies of the two models revealed no significant difference. Conclusion: Both models showed promise in predicting the occurrence of lung metastasis in soft tissue sarcomas. These findings suggest t, SCOPUS: ar.j, info:eu-repo/semantics/published

Several solid lesions can be found within the pancreas mainly arising from the exocrine and endocrine pancreatic tissue. Among all pancreatic malignancies, the most common subtype is pancreatic ductal adenocarcinoma (PDAC), to a point that pancreatic cancer and PDAC are used interchangeably. But, in addition to PDAC, and to the other most common and well-known solid lesions, either related to benign conditions, such as pancreatitis, or not so benign, such as pancreatic neuroendocrine neoplasms (pNENs), there are solid pancreatic lesions considered rare due to their low incidence. These lesions may originate from a cell line with a differentiation other than exocrine/endocrine, such as from the nerve sheath as for pancreatic schwannoma or from mesenchymal cells as for solitary fibrous tumour. These rare solid pancreatic lesions may show a behaviour that ranges in a benign to highly aggressive malignant spectrum. This review includes cases of an intrapancreatic accessory spleen, pancreatic tuberculosis, solid serous cystadenoma, solid pseudopapillary tumour, pancreatic schwannoma, purely intraductal neuroendocrine tumour, pancreatic fibrous solitary tumour, acinar cell carcinoma, undifferentiated carcinoma with osteoclastic-like giant cells, adenosquamous carcinoma, colloid carcinoma of the pancreas, primary leiomyosarcoma of the pancreas, primary and secondary pancreatic lymphoma and metastases within the pancreas. Therefore, it is important to determine the correct diagnosis to ensure optimal patient management. Because of their rarity, their existence is less well known and, when depicted, in most cases incidentally, the correct diagnosis remains challenging. However, there are some typical imaging features present on cross-sectional imaging modalities that, taken into account with the clinical and biological context, contribute substantially to achieve the correct diagnosis., SCOPUS: re.j, info:eu-repo/semantics/published

Introduction. Radiomics is a promising imaging-based tool which could enhance clinical observation and identify representative features. To avoid different interpretations, the Image Biomarker Standardisation Initiative (IBSI) imposed conditions for harmonisation. This study evaluates IBSI-compliant radiomics applications against a known benchmark and clinical datasets for agreements. Materials and methods. The three radiomics platforms compared were RadiomiX Research Toolbox, LIFEx v7.0.0, and syngo.via Frontier Radiomics v1.2.5 (based on PyRadiomics v2.1). Basic assessment included comparing feature names and their formulas. The IBSI digital phantom was used for evaluation against reference values. For agreement evaluation (including same software but different versions), two clinical datasets were used: 27 contrast-enhanced computed tomography (CECT) of colorectal liver metastases and 39 magnetic resonance imaging (MRI) of breast cancer, including intravoxel incoherent motion (IVIM) and dynamic contrast-enhanced (DCE) MRI. The intraclass correlation coefficient (ICC, lower 95% confidence interval) was used, with 0.9 as the threshold for excellent agreement. Results. The three radiomics applications share 41 (3 shape, 8 intensity, 30 texture) out of 172, 84 and 110 features for RadiomiX, LIFEx and syngo.via, respectively, as well as wavelet filtering. The naming convention is, however, different between them. Syngo.via had excellent agreement with the IBSI benchmark, while LIFEx and RadiomiX showed slightly worse agreement. Excellent reproducibility was achieved for shape features only, while intensity and texture features varied considerably with the imaging type. For intensity, excellent agreement ranged from 46% for the DCE maps to 100% for CECT, while this lowered to 44% and 73% for texture features, respectively. Wavelet features produced the greatest variation between applications, with an excellent agreement for only 3% to 11% features. Conclusion., SCOPUS: ar.j, info:eu-repo/semantics/published

Given the lack of direct comparative evidence, we aimed to compare the oncological outcomes of localized pancreatic ductal adenocarcinoma (PDAC) treated in the same tertiary cancer center with isotoxic high-dose stereotactic body radiotherapy (iHD-SBRT) or conventional chemoradiotherapy (CRT). Biopsy-proven borderline/locally advanced patients treated with iHD-SBRT (35 Gy in 5 fractions with a simultaneous integrated boost up to 53 Gy) or CRT (45-60 Gy in 25-30 fractions) were retrospectively included from January 2006 to January 2021. The median overall survival (mOS) was evaluated trough uni- and multivariate analyses. The progression free survival (PFS) and the 1-year local control (1-yLC) were also reported. Eighty-two patients were included. The median follow-up was 19.7 months. The mOS was in favour of the iHD-SBRT group (22.5 vs. 15.9 months, p < 0.001), including after multivariate analysis (HR 0.39 [CI95% 0.18-0.83], p = 0.014). The median PFS and the 1-yLC were also significantly better for iHD-SBRT (median PFS: 16.7 vs. 11.5 months, p = 0.011; 1-yLC: 75.8 vs. 39.3%, p = 0.004). In conclusion, iHD-SBRT is a promising RT option and may offer an improvement in OS in comparison to CRT for localized PDAC. Further validation is required to confirm the exact role of iHD-SBRT and the optimal therapeutic sequence for the treatment of localized PDAC., SCOPUS: ar.j, info:eu-repo/semantics/published