Search

To develop and externally validate a multiphase computed tomography (CT)-based machine learning (ML) model for staging liver fibrosis (LF) by using whole liver slices. The development dataset comprised 232 patients with pathological analysis for LF, and the test dataset comprised 100 patients from an independent outside institution. Feature extraction was performed based on the precontrast (PCP), arterial (AP), portal vein (PVP) phase, and three-phase CT images. CatBoost was utilized for ML model investigation by using the features with good reproducibility. The diagnostic performance of ML models based on each single- and three-phase CT image was compared with that of radiologists’ interpretations, the aminotransferase-to-platelet ratio index, and the fibrosis index based on four factors (FIB-4) by using the receiver operating characteristic curve with the area under the curve (AUC) value. Although the ML model based on three-phase CT image (AUC = 0.65–0.80) achieved higher AUC value than that based on PCP (AUC = 0.56–0.69) and PVP (AUC = 0.51–0.74) in predicting various stage of LF, significant difference was not found. The best CT-based ML model (AUC = 0.65–0.80) outperformed the FIB-4 in differentiating advanced LF and cirrhosis and radiologists’ interpretation (AUC = 0.50–0.76) in the diagnosis of significant and advanced LF. All PCP, PVP, and three-phase CT-based ML models can be an acceptable in assessing LF, and the performance of the PCP-based ML model is comparable to that of the enhanced CT image-based ML model.

To investigate the performance of the combined hepatocyte fraction (HepF) and apparent diffusion coefficient (ADC) values to stage hepatic fibrosis (HF) in patients with hepatitis B/C.A total of 281 patients with hepatitis B/C prospectively underwent gadoxetate disodium-based T1 mapping and diffusion-weighted imaging. HepF was determined from pre and postcontrast T1 mapping with pharmacokinetics. The independent predictors of the HF stage (S0-4) were identified from HepF, ADC, conventional T1-based parameters, and age using a logistic regression analysis. The performances of independent and combined predictors in diagnosing various HF stages were compared by analyzing receiver operating characteristic curves. The intraclass correlation coefficient (ICC) was used to assess the interobserver reproducibility of each predictor.In total, 167 patients with various stages of HF were included. All measurements had excellent interobserver agreement (ICC ≥ 0.75). The hepatic relative enhancement, HepF ,and ADC values were significantly different among various HF stages (p 0.05). The HepF and ADC were independent predictors of S0, S1, S2 , and S3 disease (p 0.05). T1The combined predictor of HepF and ADC shows acceptable performance for staging HF.

Background Polypoid lesion of gallbladder (PLG) size larger than 10 mm is considered to be one of the surgical indications, but the final pathological results are mostly non-neoplastic polyps. The aim of the study was to define the risk factors to discriminate neoplastic PLG and create more precise criteria for surgical indications. Methods A large scale, case-series study based on 2704 patients who underwent cholecystectomy for PLG was designed. Logistic regression analysis and receiver operating characteristic curve (ROC) was adopted to identify risk factors and the optimal size criteria for predicting neoplastic PLG. Results Patients in the neoplastic group were significantly older than those in the non-neoplastic group and the average PLG size is much larger in the neoplastic group (18.5 ± 4.7 mm vs 12.6 ± 3.6 mm). Neoplastic PLGs are prone to be single and non-neoplastic polyps are usually multiple. On Multivariate logistic regression analysis, PLG size larger than 15 mm and age older than 43 years were found to be the independent risk factors to discriminate neoplastic PLG （Odds ratio 3.546 and 2.77 respectively）. The ROC curve showed that 12 mm might be the more reasonable PLG size threshold for the surgical suggestion. Conclusions Considering its moderate diagnostic accuracy, the size of gallbladder polyp larger than 10 mm is insufficient to indicate surgical therapy for PLG and 12 mm should be the more optimal polyp's size threshold. Patients older than 43 years have a higher risk of having neoplastic polyps.

Information on the stage of liver cirrhosis is essential for prognostication and decisions on surgical planning for hepatocellular carcinoma (HCC) patients. But a non-invasive liver cirrhosis staging model is still lacking. The aim of our study was to develop a non-invasive model based on routine clinical parameters to evaluate the severity of cirrhosis in hepatitis B related HCC patients. A total of 226 HCC patients with chronic hepatitis B virus (HBV) infection who had liver resection were analyzed in this retrospective study. We found that platelets, prothrombin activity, maximum oblique diameter of right hepatic lobe and spleen length were the independent predictors of liver cirrhosis in HCC patients. By cumulating the weight of risk scores of independent variables, we constructed the PPMS (PLT/PTA/maximum oblique diameter of right hepatic lob/spleen length) index. The areas under the receiver operating characteristic curves (AUROC) of PPMS index were 0.820, 0.667, and 0.650 in predicting ≥cirrhosis 1 (C1), ≥cirrhosis 2 (C2), and ≥cirrhosis 3 (C3), respectively. The optimal cut-off value of the PPMS index for predicting ≥C1, ≥C2, and ≥C3 was 4.392, 4.471, and 4.784, respectively. And the corresponding sensitivity was 63.1%, 63.2%, and 64.7%, the corresponding specificity was 89.4%, 64.3%, and 62.5%, respectively. Our study constructed a non-invasive liver cirrhosis index (PPMS) could distinguish patients from different stages of liver cirrhosis, which might add more preoperative information for HCC patients.