RESUMEN
Esophageal varices (EV) in liver cirrhosis carry high mortality risks. Traditional endoscopy, which is costly and subjective, prompts a shift towards machine learning (ML). This review critically evaluates ML applications in predicting bleeding risks and grading EV in patients with liver cirrhosis. Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we conducted a systematic review of studies using ML to predict the risk of variceal bleeding and/or grade EV in liver disease patients. Data extraction and bias assessment followed the CHARMS (CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modeling Studies) checklist and PROBAST (Prediction model Risk Of Bias Assessment Tool) tool, respectively. Due to the heterogeneity of the study, a meta-analysis was not feasible; instead, descriptive statistics summarized the findings. Twelve studies were included, highlighting the use of various ML models such as extreme gradient boosting, artificial neural networks, and convolutional neural networks. These studies demonstrated high predictive accuracy, with some models achieving area under the curve values above 99%. However, significant heterogeneity was noted in input variables, methodologies, and outcome measures. Moreover, a substantial portion of the studies exhibited unclear or high risk of bias, mainly due to insufficient participant numbers, unclear handling of missing data, and a lack of detailed reporting on endoscopic procedures. ML models show significant promise in predicting the risk of variceal bleeding and grading EV in patients with cirrhosis, potentially reducing the need for invasive procedures. Nonetheless, the current literature reveals considerable heterogeneity and methodological limitations, including high or unclear risks of bias. Future research should focus on larger, prospective trials and the standardization of ML assessment criteria to confirm these models' practical utility in clinical settings.
RESUMEN
Background: Bowel ultrasonography (BUS) is emerging as a promising noninvasive tool for assessing disease activity in inflammatory bowel disease (IBD) patients. We evaluated the diagnostic accuracy of BUS in IBD patients against the gold standard diagnostic method, standard colonoscopy. Methods: Major databases were searched from inception to May 2023 for studies on BUS diagnostic accuracy in IBD. Outcomes of interest were pooled sensitivity, specificity, positive (PPV), and negative (NPV) predictive values. Endoscopic confirmation served as ground truth. Standard meta-analysis methods with a random-effects model and I2 statistics were applied. Risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 tool. Results: Twenty studies (1094 patients) were included in the final analysis. The majority (75%) of studies considered bowel wall thickness >3 mm as abnormal. Endoscopic evaluation was performed between days 3 and 180. The pooled diagnostic accuracy of BUS in IBD was 66% (95% confidence interval [CI] 58-72%; I2=78%), sensitivity was 88.6% (95%CI 85-91%; I2=77%), and specificity 86% (95%CI 81-90%; I2=95%). PPV and NPV were 94% (95%CI 93-96%; I2=25%) and 74% (95%CI 66-80%; I2=95%), respectively. On subgroup analysis, small-intestine contrast-enhanced ultrasonography (SICUS) demonstrated high sensitivity (97%, 95%CI 91-99%; I2=83%), whereas BUS exhibited high specificity (94%, 95%CI 92-96%; I2=0%) and NPV (76%, 95%CI 68-83%; I2=80.9%). Meta-regression revealed a significant relation between side-to-side anastomosis and BUS specificity (P=0.02) and NPV (P=0.004). Conclusion: The high diagnostic accuracy of BUS in detecting bowel wall inflammation suggests utilizing regular BUS as the primary modality, with subsequent consideration of SICUS if clinically warranted.