Pancreatic ductal adenocarcinoma (PDAC) stands out as the predominant malignant neoplasm affecting the pancreas, characterized by a poor prognosis, in most cases patients being diagnosed in a nonresectable stage. Image-based artificial intelligence (AI) models implemented in tumor detection, segmentation, and classification could improve diagnosis with better treatment options and increased survival. This review included papers published in the last five years and describes the current trends in AI algorithms used in PDAC. We analyzed the applications of AI in the detection of PDAC, segmentation of the lesion, and classification algorithms used in differential diagnosis, prognosis, and histopathological and genomic prediction. The results show a lack of multi-institutional collaboration and stresses the need for bigger datasets in order for AI models to be implemented in a clinically relevant manner.

Background & Aims: To evaluate the predictive factors for recurrence of the disease and overall survival(OS) after achieving complete response (CR) in patients with hepatocellular carcinoma (HCC) treated withtransarterial chemoembolization (TACE).Methods: From January 2013 to December 2017, 168 treatment-naïve patients diagnosed with HCCunderwent TACE as a first-line therapy and the gathered data was retrospectively reviewed. We determined the predictive factors for complete response (CR), for recurrence after CR and for survival using the Cox proportional hazard model.Results: Median follow-up was 27.4 months (range 4-65 months). The mean patient age was 62.2±7.9 years. Eighty-three patients had an α-fetoprotein (AFP) level > 20ng/mL. The median maximal diameter of the tumors was 3.5 cm. Sixty-three patients (37.5%) achieved CR after TACE, and recurrence after CR was detected in 37 patients (58.7%). In multivariate analysis, tumor size (≤4.5 cm) and a single tumor were found to be predictive factors for CR, with hazard ratios (HRs) of 2.352 (p=0.022) and 3.964 (p 25 ng/mL and multiple tumors were demonstrated to have a significant relationship with recurrence after CR, with HRs of 1.650 (p=0.05) and 3.932 (p=0.038), respectively. Increased initial serum AFP > 22 ng/mL, tumorsize > 4.5 cm, outside Milan criteria, not receiving a liver transplant and presence of portal vein thrombosis (PVT) were correlated with poor survival.Conclusions: In patients treated with TACE as an initial therapy, tumor size (≤4.5 cm) and single tumor were predictive factors for CR. Multiple nodules and an elevated serum AFP > 25 ng/mL were predictive factors for recurrence after CR. Outside Milan criteria tumors, elevated AFP levels and the presence of PVT were significantly correlated with decreased survival.

The liver is a frequent focus in radiology due to its diverse pathology, and artificial intelligence (AI) could improve diagnosis and management. This systematic review aimed to assess and categorize research studies on AI applications in liver radiology from 2018 to 2024, classifying them according to areas of interest (AOIs), AI task and imaging modality used. We excluded reviews and non-liver and non-radiology studies. Using the PRISMA guidelines, we identified 6680 articles from the PubMed/Medline, Scopus and Web of Science databases; 1232 were found to be eligible. A further analysis of a subgroup of 329 studies focused on detection and/or segmentation tasks was performed. Liver lesions were the main AOI and CT was the most popular modality, while classification was the predominant AI task. Most detection and/or segmentation studies (48.02%) used only public datasets, and 27.65% used only one public dataset. Code sharing was practiced by 10.94% of these articles. This review highlights the predominance of classification tasks, especially applied to liver lesion imaging, most often using CT imaging. Detection and/or segmentation tasks relied mostly on public datasets, while external testing and code sharing were lacking. Future research should explore multi-task models and improve dataset availability to enhance AI’s clinical impact in liver imaging.

Background and Aims: The evidence regarding the use of anticoagulant (AC) agents in portal vein thrombosis (PVT) is increasing and, most patients undergo chronic treatment with low molecular weight heparin (LMWH) or vitamin K antagonists (VKA). Nevertheless, there are no clear data about who should receive antithrombotic therapy, when to initiate it, how long and what dose should be used for this set of patients. The aim of the study was to assess the outcome of patients with cirrhosis and portal vein thrombosis who received AC therapy, in terms of thrombus regression, bleeding events and survival rates. Methods: This observational and retrospective study included 107 cirrhotic patients diagnosed with PVT in a single tertiary center between 2010-2019. 54 received low molecular weight heparin or vitamin K antagonist (AC treatment group) and 53 were untreated. All patients were periodically follow-up to assess the evolution of PVT (regression, progression, stable thrombus) and potential occurrence of bleeding events. Results: The regression of portal vein thrombosis was significantly higher in the AC treatment group (OR=2.430; 95% CI=1.11-6.167; p=0.026), more than 50% of on-treatment patients experiencing regression of the thrombus. However, bleeding events were significantly more frequent in the AC treatment group (18.5% vs. 7.5%) and the risk of bleeding was associated with thrombocytes less than 50x103/mm3 (OR=8.266; 95%CI: 2.310-39.211; p=0.002). Survival was better in the AC treatment group (68.4% vs 48.7% at 5 years and 92.7% vs 77.8% at 1 year, p=0.038) and was lower in patients that experienced bleeding events (37.22% survival at 5 years, mean time survival 44 months, p=0.008). Conclusions: In our cohort of cirrhotic patients with PVT more than 50% of patients receiving AC therapy presented regression of the thrombus; most of them obtained partial recanalization. The bleeding complication rate was higher than expected, reaching 18%. The overall mortality was lower in the treated group.

Background: Liver resection (LR) is the treatment of choice for most benign and malignant focal liver lesions, as well as in selected patients with liver trauma. Few other therapies can compete with LR in selected cases, such as liver transplantation in hepatocellular carcinoma (HCC) and ablative therapies in small HCCs or liver metastases. The present paper analyses a single center experience in LR, reviewing the indications of LR, the operative techniques and their short-term results. Between January 2000 and December 2016, in "œDan Setlacec" Center of General Surgery and Liver Transplantation were performed 3165 LRs in 3016 patients, for pathologic conditions of the liver. In the present series, liver resections for living-donor liver transplantation were excluded. The median age of the patients was 56 years (mean 58 years; range 1-88), with male/female ratio 1524/1492 and adult/pediatric patient ratio 2973/43. Results: Malignant lesions were the main indication for LR (2372 LRs; 74.9%). Among these, colorectal liver metastases were the most frequent indication (952 LRs; 30.1%), followed by hepatocellular carcinoma (575 patients, 18.2%). The highest number of resected tumors per patient was 21, and the median diameter of the largest tumor was 40 mm (mean 51 mm; range 3-250). Major resections rate was 18.6% (588 LRs) and anatomical LRs were performed in 789 patients (24.9%). The median operative time was 180 minutes (mean 204 minutes; range 45-920). The median blood loss was 500 ml (mean 850 ml; range 500-9500), with a transfusion rate of 41.6% (1316 LRs). The morbidity rate was 40.1% (1270 LRs) and the rate of major complications (Dindo-Clavien IIIa or more) was 13.2% (418 LRs). Mortality rate was 4.2% (127 pts). LRs should be performed in specialized high-volume centers to achieve the best results (low morbidity and mortality rates).