PURPOSE:Pancreatic ductal adenocarcinoma (PDA) is a leading cause of cancer-related deaths, with early diagnosis hampered by nonspecific symptoms and limitations of existing imaging techniques. This study aimed to develop a deep learning (DL) algorithm to automatically classify pancreas lesions on contrast-enhanced CT scans as normal, benign, or malignant, to assist radiologists in detecting early-stage pancreatic cancer. MATERIALS AND METHODS:A dataset of 1,037 portal-phase CT scans was compiled from 18 institutions. The dataset was divided into a training set (N = 732) and a test set (N = 305), which was further divided into an internal validation test set (N = 139) and an external validation test set (N = 166). After segmentation using the TotalSegmentator algorithm, the pancreas was isolated from each CT scan. A DL model combining TotalSegmentator's pre-trained encoder and nnUNet decoder layers was developed to classify pancreas lesions. Ten-fold cross-validation was applied, and model performance was assessed using precision, recall, area under the curve (AUC) and a final score (FS) representing a weighted average of the previous three metrics. The final prediction combined the outputs of ten models. RESULTS:Across all validation datasets (139 and 166 patients, respectively, in the internal and external dataset), precision and recall were 0.57 and 0.63, respectively, while AUC was 0.84. In the external validation dataset, malignant lesions were detected with an AUC of 0.97. The model achieved an FS of 0.72 in both internal and external validation datasets, indicating consistent performance across datasets. CONCLUSION:This study demonstrated the feasibility of using a DL algorithm for automated pancreas lesion classification in CT scans. The model showed strong performance, particularly in detecting malignant lesions. Further research is needed to assess the model's clinical applicability and performance in real-world settings.

The co-primary aims of this study were to assess the diagnostic and prognostic performance of Whole-body-2-[18F]-fluorodeoxyglucose-positron emission tomography coupled with MRI (WB-2-[18F]FDG-PET/MRI) imaging in the smoldering multiple myeloma (SMM) workup for detection of MM-related medullary or extra-medullary disease and for the prediction of progression-free survival (PFS) defined as time to progression to symptomatic MM requiring therapy. A total of 116 patients with SMM (without CRAB or SLiM criteria before imaging) were prospectively included in the study and underwent full multi-parametric WB-2-[18F]FDG-PET/MRI imaging. PET detected at least one FL > 5 mm in 9% of patients compared to 20% on MRI (p = 0.02). PET detected diffuse bone marrow involvement (BMI) in 20% of patients and MRI in 53% (p < 10−3). A total of 98 patients with true SMM not requiring treatment were then followed up. The presence of diffuse BMI on MRI was the strongest adverse prognostic parameter for PFS (univariate: hazard ratio (HR), 6.12; p < 10−2. Multivariate : HR, 4.16; p = 0.03) and could be proposed as a new high-risk biomarker for progression to symptomatic MM. Dynamic contrast-enhanced (DCE)-MRI based increased peak enhancement intensity (PEI) and maximum intensity time ratio (MITR) values were other strong adverse prognostic factors.

Abstract Background Intestinal ultrasound (IUS) has emerged as a valuable tool for monitoring inflammatory bowel disease (IBD), but its reliability depends on the technical skills of the operator. This study aimed to assess the IUS learning curve of a complete novice and a gastroenterologist with formal general ultrasound training and regular IUS experience, using an expert radiologist as the gold standard. Methods This was a single-center, prospective study conducted between April 2023 and January 2024, which consecutively included patients with IBD (Crohn’s disease [CD], ulcerative colitis [UC]). Three IUS scans were performed successively and in a standardized manner by the three operators in each included patient. Each operator assessed disease activity using quantitative and semi-quantitative scales, and ultrasound activity scores based on bowel wall thickness and stratification, Doppler signal intensity according to the Limberg score, and appearance of mesenteric fat. Inter-operator agreement was calculated using the intraclass correlation coefficient (ICC). Results A total of 50 patients were included (76.6% CD; 23.4% UC), of whom 31.9% had active disease and 23.4% a history of IBD-related surgery. Mean BMI (standard deviation [SD]) was 24.2 (4.2) kg/m². Mean examination time was 4’12’’ for the expert radiologist (gold standard), 3’45’’ for the experienced gastroenterologist, and 5’17’’ for the novice (p<0.001). Regarding quantitative assessment of IUS activity (visual analog scale), the experienced gastroenterologist had good agreement with the radiologist throughout the study (ICC ranging from 0.61 at the start of the study to 0.79 at the end; overall ICC = 0.71 [95% CI 0.54-0.83]), while the novice maintained poor agreement, with no learning curve (ICC increasing from 0.21 to 0.33; overall ICC = 0.33 [0.05-0.55]). Similar results were obtained with the other activity measurements. The ileocolonic location of CD and surgical history significantly impaired novice performance. Excluding surgical patients enabled a learning curve to be obtained for the novice, with poor initial agreement (ICC 0.21) then moderate to good agreement at the end of the study (ICC 0.59). Conclusion A formal general ultrasound training and 200 IUS examinations of experience allowed good agreement with an expert radiologist in IBD patients. The novice failed to go beyond poor agreement throughout the study, although a faster learning curve was observed in patients without history of IBD surgery. These findings highlight the difficulty of the technical performance of IUS rather than its interpretation. Comprehensive training beforehand and between 50 and 200 examinations of experience are necessary before embarking on IUS.

Background Intestinal ultrasound (IUS) has emerged as a valuable tool for monitoring inflammatory bowel disease (IBD), but its reliability depends on the technical skills of the operator. Aims This study aimed to assess the IUS learning curve of a novice briefly trained to get him familiarized with the ultrasound machine and normal/abnormal IUS images, and a gastroenterologist with formal general ultrasound training and regular IUS experience (>200 exams), using an expert radiologist as the gold standard. Methods The ULTRA-IBD study was a single-center, prospective study involving 50 consecutive IBD patients. Three successive IUS examinations were performed on each patient by the 3 operators. Inter-rater agreement on IUS disease activity was calculated using the intraclass correlation coefficient. Results The experienced gastroenterologist demonstrated consistently good agreement with the radiologist throughout the study, while the novice failed to go beyond poor agreement after 50 examinations, although a faster learning curve was observed in patients without history of IBD surgery. Conclusion These findings highlight the difficulty of the technical performance of IUS rather than its interpretation. Comprehensive ultrasound training beforehand and between 50 and 200 examinations of experience are necessary before embarking on IUS.

PurposeTo assess the efficacy of the gelatin torpedoes embolization technique after lung neoplastic lesions percutaneous radiofrequency ablation (PRFA) to reduce chest tube placement rate and hospital length of stay, and the safety of this embolization technique.Materials and methodsA total of 114 PRFA of lung neoplastic lesions performed in two centers between January 2017 and December 2022 were retrospectively reviewed. Two groups were compared, with 42 PRFA with gelatin torpedoes embolization technique (gelatin group) and 72 procedures without (control group). Procedures were performed by one of seven interventional radiologists using LeVeen CoAccess™ probe. Multivariate analyses were performed to identify risk factors for chest tube placement and hospital length of stay.ResultsThere was a significantly lower chest tube placement rate in the gelatin group compared to the control group (3 [7.1 %] vs. 27 [37.5 %], p < 0,001). Multivariate analysis showed a significant association between chest tube placement and gelatin torpedoes embolization technique (OR: 0.09; 95 % CI: 0.02–0.32; p = 0.0006). No significant difference was found in hospital length of stay between the two groups. Multivariate analysis did not show a significant relationship between hospital length of stay and gelatin torpedoes embolization technique. No embolic complication occurred in the gelatin group.ConclusionGelatin torpedoes embolization technique after PRFA of lung neoplastic lesions resulted in significantly reduced chest tube placement rate in our patient population. No significant reduction in hospital length of stay was observed. No major complication occurred in the gelatin group.

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.

Adrenal lesions are common incidental findings in clinical practice, which are mostly benign and harmless; adenomas are the most common benign adrenal tumors, representing more than 75% adrenal lesions. Medical information provided by CT scans such as lesions' dimension, attenuation values, etc., are crucial for the diagnosis of adenoma. Measurements of percentage washout of injected contrast material from contrast-enhanced CT provide reproducible means to distinguish adenomas from malignant masses. Despite of the 3D volume CT data, only selected 2D slices are used in this diagnosis process, which introduces uncertainty of clinical decision and requires high expertise of medical professionals. To alleviate this problem and to facilitate the diagnosis, we proposed an region-growing based 3-Slices washout calculation method, as a preliminary study of our further work of automatic 3D adrenal lesion characterisation. Comparing with the expert's diagnosis, our method showed a significant (more than 10%) improvement on the accuracy, revealing that computer-based 3D lesion characterisation could become a promising and reliable tool for the diagnosis of adrenal lesions.

To determine whether instillation of normal saline solution for sealing the needle track reduces incidence of pneumothorax and chest tube placement after computed tomography-guided percutaneous lung biopsy. A total of 242 computed tomography-guided percutaneous lung biopsies performed at a single institution were retrospectively reviewed, including 93 biopsies in which the needle track was sealed by instillation of 3–5 ml of normal saline solution during needle withdrawal (water seal group) and 149 biopsies without sealing (control group). Patient and lesion characteristics, procedure-specific variables, pneumothorax and chest tube placement rates were recorded. Baseline characteristics were comparable in both groups. There was a statistically significant decrease in the pneumothorax rate (19.4

Managing hepatocellular carcinoma (HCC) in patients with a transjugular intrahepatic portosystemic shunt (TIPS) is becoming increasingly common. This study aimed to evaluate the safety and efficacy of percutaneous thermal ablation for treating HCC in patients with TIPS. This retrospective longitudinal study was conducted at Nantes University Hospital. The main inclusion criteria were patients undergoing percutaneous thermal ablation for HCC. Patients with a pre-existing TIPS were included in the ‘TIPS group’. A 1:1 control group without TIPS, the ‘n-TIPS group’, was created for this case-control study. The primary endpoints were overall survival and progression-free survival over 12 months. Safety was assessed by comparing complications between the groups. Between 2008 and 2020, 371 patients underwent percutaneous thermal ablation for HCC. The ‘TIPS group’ included 34 patients (66 nodules), while 34 patients (84 nodules) were randomly assigned to the ‘n-TIPS group.’ Overall survival rates at 1 year were 97

Although peptide radionuclide therapy (PRRT) using a somatostatin analog (SSA) radiolabeled with a beta- emitter: [177Lu]Lu-DOTATATE has shown a good clinical efficacy in neuroendocrine tumors (NETs), most of the patients only achieved tumoral stabilization and rare but severe long-term hematological toxicities have been reported. One of the promising options to improve PRRT is targeted alpha therapy. It is therefore essential to propose animal models that can mimic systemic spread disease, especially microscopic disease such as early stage of NET liver metastases to explore targeted alpha therapy. Herein, we report the evaluation of efficacy and toxicity of [225Ac]Ac-DOTATOC in an original preclinical murine model simulating the development of well-characterized liver metastases of pancreatic NETs with SSTR overexpression. A mouse model of liver metastases of pancreatic NETs was developed by intraportal injection of AR42J cells and explored using [68 Ga]Ga-DOTATOC and [18F]F-FDG PET/MRI. Biodistribution study and radiation dosimetry of [225Ac]Ac-DOTATOC were determined in subcutaneous tumor-bearing NMRI-nude mice. Efficacy and toxicity were determined by intravenous injection of increasing activities of [225Ac]Ac-DOTATOC 10 days after intraportal graft. Liver tumors showed a high uptake of [68 Ga]Ga-DOTATOC and no uptake of [18F]F-FDG confirming the well-differentiated phenotype. All groups treated with [225Ac]Ac-DOTATOC showed a significant increase in overall survival compared with DOTATOC-treated mice, especially those treated with the highest activities: 53 days with 240 kBq (p = 0.0001), and 58 days with 2 × 120 kBq (p < 0.0001) vs 28 days with non-radiolabeled DOTATOC. On blood tests, a transient and moderate decreased in white blood cells count after treatment and no severe hepatic or renal toxicity were observed after treatment which was consistent with pathological and radiation dosimetry findings. [225Ac]Ac-DOTATOC exhibit a favorable efficacy and toxicity profile in a mouse model of liver micrometastatic pancreatic NET.

Abstract Background Dynamic contrast-enhanced-MRI (DCE-MRI) is able to study bone marrow angiogenesis in patients with multiple myeloma (MM) and asymptomatic precursor diseases but its role in the management of MM has not yet been established. The aims of this prospective study was to compare DCE-MRI-based parameters between all monoclonal plasma cell disease stages in order to find out discriminatory parameters and to seek correlations with other diffusion-weighted MRI and positron emission tomography (PET)-based biomarkers in a hybrid simultaneous whole-body-2-[18F]fluorodeoxyglucose (FDG)-PET/MRI (WB-2-[18F]FDG-PET/MRI) imaging approach. Methods Patients with newly diagnosed Monoclonal gammopathy of undetermined significance (MGUS), smoldering multiple myeloma (SMM) or symptomatic MM according to international myeloma working group and underwent WB-2-[18F]FDG-PET/MRI imaging including bone marrow DCE sequences at the Nantes University Hospital were prospectively enrolled in this study before receiving treatment. Results One hundred and sixty-seven patients (N = 167, mean age: 64 years ± 11 [Standard deviation], 66 males) were considered for the analysis. DCE-MRI-based Peak Enhancement Intensity (PEI), Time to PEI (TPEI) and their maximum intensity time ratio (MITR: PEI/TPEI) values were significantly different between the different monoclonal plasma cell disease stages, PEI values increasing and TPEI values decreasing progressively along the spectrum of plasma cell disorders, from MGUS stage to symptomatic multiple myeloma. PEI values were significantly higher in patients with diffuse bone marrow involvement (either in PET or in MRI images) than in those without diffuse bone marrow involvement, unlike TPEI values. PEI and TPEI values were not significantly different between patients with or without focal bone lesions. Conclusion Different DCE-MRI-based parameters (PEI, TPEI, MITR) could significantly differentiate all monoclonal plasma cell disease stages and complemented conventional MRI and PET-based biomarkers.

Abstract Background Abdominal pain is common in patients visiting the emergency department (ED). The aim of this study was to assess the diagnostic contribution of point-of-care ultrasound (POCUS) in patients presenting to the ED with acute abdominal pain. Methods We designed an interventional randomized, controlled, open label, parallel-group, trial in two French EDs. We included adult patients presenting to the ED with acute abdominal pain. Exclusion criteria were a documented end-of-life, an immediate need of life-support therapy and pregnant or breast-feeding women. Patients were randomized in the experimental group (i.e., workup including POCUS) or control group (usual care). The primary objective of the study was to assess the added value of POCUS on diagnostic pathway in the ED, according to the diagnostic established a posteriori by an adjudication committee. The primary endpoint was the proportion of exact preliminary diagnosis between the 2 groups. The preliminary diagnosis made after clinical examination and biological results with POCUS (intervention arm) or without POCUS (usual care) was considered exact if it was similar to the adjudication committee diagnosis. Results Between June 2021 11th and June 2022 23th, 256 patients were randomized, but five were not included in the primary analysis, leaving 125 patients in the POCUS group and 126 patients in the usual care group (130 women and 121 men, median [Q1-Q3] age: 42 [30;57]). There was no difference for exact diagnosis between the two groups (POCUS 70/125, 56% versus control 78/126 (62%), RD 1.23 [95% CI 0.74–2.04]). There was no difference in the accuracy for the diagnosis of non-specific abdominal pain nor number of biological or radiological exams. Diagnostic delays and length of stay in the ED were also similar. Conclusions In this trial, systematic POCUS did not improve the rate of diagnostic accuracy in unselected patients presenting to the ED with acute abdominal pain. However, as it was a safe procedure, further research should focus on patients with suspected etiologies where POCUS is particularly useful. Trial registration: This trial was registered on ClinicalTrials.gov on 2022/07/20 ( https://clinicaltrials.gov/study/NCT04912206?id=NCT04912206&rank=1 ) (NCT04912206).

To validate the proton density fat fraction (PDFF) obtained by the MRQuantif software from 2D chemical shift encoded MR (CSE–MR) data in comparison with the histological steatosis data. This study, pooling data from 3 prospective studies spread over time between January 2007 and July 2020, analyzed 445 patients who underwent 2D CSE–MR and liver biopsy. MR derived liver iron concentration (MR–LIC) and PDFF was calculated using the MRQuantif software. The histological standard steatosis score (SS) served as reference. In order to get a value more comparable to PDFF, histomorphometry fat fraction (HFF) were centrally determined for 281 patients. Spearman correlation and the Bland and Altman method were used for comparison. Strong correlations were found between PDFF and SS (rs = 0.84, p < 0.001) or HFF (rs = 0.87, p < 0.001). Spearman’s coefficients increased to 0.88 (n = 324) and 0.94 (n = 202) when selecting only the patients without liver iron overload. The Bland and Altman analysis between PDFF and HFF found a mean bias of 5.4

Après la ménopause, la décompensation des troubles de la statique pelvienne s’explique par le sevrage de l’imprégnation œstrogènique.