RESUMEN
PURPOSE: The purpose of the 2023 SFR data challenge was to invite researchers to develop artificial intelligence (AI) models to identify the presence of a pancreatic mass and distinguish between benign and malignant pancreatic masses on abdominal computed tomography (CT) examinations. MATERIALS AND METHODS: Anonymized abdominal CT examinations acquired during the portal venous phase were collected from 18 French centers. Abdominal CT examinations were divided into three groups including CT examinations with no lesion, CT examinations with benign pancreatic mass, or CT examinations with malignant pancreatic mass. Each team included at least one radiologist, one data scientist, and one engineer. Pancreatic lesions were annotated by expert radiologists. CT examinations were distributed in balanced batches via a Health Data Hosting certified platform. Data were distributed into four batches, two for training, one for internal evaluation, and one for the external evaluation. Training used 83 % of the data from 14 centers and external evaluation used data from the other four centers. The metric (i.e., final score) used to rank the participants was a weighted average of mean sensitivity, mean precision and mean area under the curve. RESULTS: A total of 1037 abdominal CT examinations were divided into two training sets (including 500 and 232 CT examinations), an internal evaluation set (including 139 CT examinations), and an external evaluation set (including 166 CT examinations). The training sets were distributed on September 7 and October 13, 2023, and evaluation sets on October 15, 2023. Ten teams with a total of 93 members participated to the data challenge, with the best final score being 0.72. CONCLUSION: This SFR 2023 data challenge based on multicenter CT data suggests that the use of AI for pancreatic lesions detection is possible on real data, but the distinction between benign and malignant pancreatic lesions remains challenging.
RESUMEN
OBJECTIVE: To develop a simple scoring system in order to predict the risk of severe (death and/or surgery) ischemic colitis METHODS: In this retrospective study, 205 patients diagnosed with ischemic colitis in a tertiary hospital were consecutively included over a 6-year period. The study sample was sequentially divided into a training cohort (n = 103) and a validation cohort (n = 102). In the training cohort, multivariable analysis was used to identify clinical, biological, and CT variables associated with poor outcome and to build a risk scoring system. The discriminative ability of the score (sensitivity, specificity, positive predictive value, negative predictive value) was estimated in the two cohorts to externally validate the score, and a receiver operating characteristic curve was established to estimate the area under the curve of the score. Bootstrapping was used to validate the score internally. RESULTS: In the training cohort, four independent variables were associated with unfavorable outcome: hemodynamic instability (2 pts), involvement of the small bowel (1 pt), paper-thin wall pattern (3 pts), no stratified enhancement pattern (1 pt). The score was used to categorize patients into low risk (score: 0, 1), high risk (score: 2-3), and very high risk (score: 4-7) groups with sensitivity and specificity of 97% and 67%, respectively, and a good discriminating capability, with a C-statistic of 0.94. Internal and external validation showed good discrimination capability (C-statistics of 0.9 and 0.84, respectively). CONCLUSION: A simple risk score can stratify patients into three distinct prognosis groups, which can optimize patient management. CLINICAL TRIAL NUMBER: NCT04662268 KEY POINTS: ⢠Simple scoring system predicting the risk of severe ischemic colitis ⢠First study to include CT findings to the clinical and biological data used to determine a severity score.