Newsletter
Every new paper in this field, in your inbox.
A newly developed deep learning auto-segmentation model effectively segments pancreatic cancer and surrounding structures to assist surgical planning. In a study involving 275 patients, the model achieved a Dice similarity coefficient of 75.4 in internal and 75.6 in external validation. High segmentation accuracy was noted…
Endoscopic management has emerged as a key strategy for treating biliary leakages resulting from surgery or trauma. Researchers emphasize the importance of accurately classifying leaks to choose appropriate treatments, which range from conservative care to surgical intervention. Notably, endoscopic techniques, especially endoscopic retrograde cholangiopancreatography with…
Findings indicate that AI-driven imaging models significantly improve the detection of pancreatic ductal adenocarcinoma (PDA) through advanced radiomics and deep learning techniques. These models can identify subtle pre-diagnostic signs on CT scans months to years before clinical symptoms arise, addressing current limitations in sensitivity and…
A narrative review underscores the significance of radiological assessments following pancreaticoduodenectomy in detecting postoperative complications. Utilizing cross-sectional imaging, particularly contrast-enhanced computed tomography, allows for prompt diagnosis and better patient outcomes. The review highlights the complementary role of magnetic resonance imaging in identifying specific biliary and…
A study evaluating augmented reality navigation combined with indocyanine green fluorescence in laparoscopic right hemihepatectomy found that this approach significantly reduced intraoperative blood loss (150 ml vs 250 ml) and lowered blood transfusion rates (13.5% vs 39.5%). Additionally, it improved one- and three-year recurrence-free survival…
A novel risk score, PRETREAT, was developed to estimate post-transplant recurrence in patients with perihilar cholangiocarcinoma (pcca) following neoadjuvant therapy and liver transplantation. In a study involving 399 patients, the five-year recurrence rate was 29.7% in the development cohort and 27.6% in the validation cohort.…
A novel CAP score, combining type-IV collagen, albumin, and prothrombin time, effectively predicts 180-day surgery-related mortality in liver resection for hepatocellular carcinoma. With a predictive accuracy (AUC: 0.728), the score identifies risk levels in patients, showing mortality rates of 0.8% for low risk to 26.7%…
A novel AI-assisted endoscopic ultrasound (EUS) system demonstrated superior performance in recognizing biliopancreatic stations and segmenting anatomical structures. Testing with 45,737 EUS images revealed the mean teacher algorithm achieved 95.60% accuracy in internal tests and 92.72% in external validation for station recognition. Additionally, U-Net v2…
A systematic review evaluated enhanced recovery after surgery (ERAS) protocols specifically for cirrhotic patients undergoing hepatectomy. From 3,123 screened articles, 81 were analyzed, revealing five perioperative items with moderate or high evidence levels. Key recommendations include preoperative ascites control, postoperative venous thromboembolism prophylaxis if coagulation…
A predictive model for lymph node metastases (lnm) in intrahepatic cholangiocarcinoma patients undergoing adequate lymphadenectomy has been developed, revealing significant associations with preoperative CA 19-9 levels, suspicious lymph nodes, patient age, and tumor burden score. Analysis of 693 patients resulted in a model with high…
