Detection and characterization of pancreatic lesion with artificial intelligence: The SFR 2023 artificial intelligence data challenge.

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.

Fat-containing hepatocellular carcinoma in patients with cirrhosis: proposal of a diagnostic modification regarding enhancement characteristics.

OBJECTIVES: The aim of this study was to develop and validate an algorithm for the non-invasive diagnosis of these fat-containing HCCs. METHODS: Eighty-four cirrhotic patients with 77 fat-containing HCCs and 11 non-HCC fat-containing nodules were retrospectively included. All MRIs were reviewed; nodule characteristics, European Association for the Study of the Liver (EASL) and LI-RADS classifications, and survival were collected. One of the major features of LI-RADS v2018 (non-rim-like arterial phase hyperenhancement [APHE]) was changed to include different enhancing patterns at arterial phase and a new fat-LI-RADS algorithm was created for fat-containing nodules in cirrhosis. Its diagnostic performance was evaluated in both a derivation and external validation cohort (external cohort including 58 fat-containing HCCs and 10 non-HCC fat nodules). Reproducibility of this new algorithm was assessed. RESULTS: In the derivation cohort, 54/77 (70.1%) fat-containing HCCs had APHE, 62/77 (80.5%) had enhancement compared to the nodule itself at arterial phase (APE), 43/77 (55.8%) had washout, and 20/77 (26.0%) had an enhancing capsule. EASL and LI-RADS had a sensitivity of 37.7% (29/77) and 36.4% (28/77), respectively, for the diagnosis of fat-containing HCC and both had a specificity of 100% (11/11). The new fat-LI-RADS algorithm increased sensitivity to 50.6% (39/77) without decreasing the specificity of 100% (11/11). The validation cohort confirmed the increased sensitivity, with a slight decrease in specificity. The concordance for the diagnosis of HCC for fat-LR5 was 85.3% (58/68). CONCLUSION: The new fat-LI-RADS algorithm proposed here significantly improves the performance of the non-invasive diagnosis of fat-containing HCC and thus could reduce the number of biopsies conducted for fat-containing HCCs. CLINICAL RELEVANCE STATEMENT: The European Association for the Study of the Liver and LI-RADS guidelines are poorly sensitive for the diagnosis of fat-containing HCC, mainly because of the low rate of arterial phase hyperenhancement (APHE) displayed by fat-containing HCC. Using all types of enhancement instead of APHE improves sensitivity of LI-RADS. KEY POINTS: • Fat-containing HCCs on MRI account for 7.5% of HCCs and have different imaging characteristics from non-fatty HCCs. • The European Association for the Study of the Liver and LI-RADS algorithms for the non-invasive diagnosis of HCC have low sensitivity for the diagnosis of fat-containing HCC with MRI (37.7% and 36.4%, respectively). • The new fat-LI-RADS, which includes a slight modification of the "arterial enhancement" criterion, improves the sensitivity for the diagnosis of fat-containing HCC using MRI, without degrading the specificity.

Imaging after percutaneous thermal and non-thermal ablation of hepatic tumour: normal appearances, progression and complications.

The increasing number of liver tumours treated by percutaneous ablation leads all radiologists to be confronted with the difficult interpretation of post-ablation imaging. Radiofrequency and microwave techniques are most commonly used. Recently, irreversible electroporation treatments that do not induce coagulation necrosis but cellular apoptose and respect the collagen architecture of bile ducts and vessels have been introduced and lead to specific post-ablation features and evolution. Ablations cause 'normal' changes in ablation and periablation zones. It is necessary to know these post-ablation features to avoid the misinterpretation of recurrence or complication that would lead to unnecessary treatments. Another challenge for the radiologist is to detect as early as possible the residual unablated tumour or the disease progression (local progression and tumour seeding) that will require a new treatment. Finally, the complications, frequent or rarer, should be recognised to be managed adequately. The purpose of this article is therefore to describe the large spectrum of normal and pathological aspects related to the treatment of hepatic tumour by percutaneous thermal ablation and irreversible electroporation ablation.