Answers are grounded in this paper and cite their sources.
Cirrhosis and portal hypertension remain major challenges in interventional radiology, with current diagnostic and therapeutic approaches limited by requiring invasive assessment, variability inherent to imaging modalities, and suboptimal risk stratification. For instance, traditional risk scoring systems, such as the Model for End-Stage Liver Disease, are not sufficient to predict complications after interventions such as transjugular intrahepatic portosystemic shunt (TIPS) insertion. Similarly, imaging via Doppler ultrasound, computed tomography, and magnetic resonance imaging provide valuable insights but suffer from operator dependence, inconsistent sensitivity, and accessibility constraints. This narrative review summarizes the current state of the art with regard to leveraging artificial intelligence (AI) to diagnose and predict outcomes in patients with cirrhosis or portal hypertension as well as to optimize outcomes in patients undergoing TIPS placement for the treatment of these conditions. The implementation of AI into clinical practice could potentially significantly transform the treatment of cirrhosis and portal hypertension by optimizing diagnostic accuracy, procedural safety, and long-term outcomes.
No takes yet. Share an insight, caveat, or question.
Khooblall et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: