Combining CT-based radiomics and deep learning features with clinical data enhances prediction of hepatocellular carcinoma risk in cirrhosis patients.
Key Details
- 1Study used a multicenter, prospective cohort of 2,411 cirrhosis patients in China (2018–2023).
- 2All patients underwent 3-phase contrast-enhanced abdominal CT at baseline.
- 3AI model extracted radiomics (PyRadiomics) and deep learning (ResNet-18) features from liver and spleen on CT.
- 4The integrated aMAP-CT model significantly outperformed standard clinical models (AUC 0.809–0.869).
- 5Model stratified patients into high- (26.3% incidence) and low-risk (1.7%) groups over three years.
- 6Stepwise application identified 7% of patients at very high risk for HCC (27.2% three-year incidence).
Why It Matters
This study demonstrates how AI-driven analysis of imaging data can meaningfully improve personalized risk stratification in a major oncologic domain, supporting earlier HCC detection and potentially better clinical outcomes for high-risk cirrhosis patients.
Source
EurekAlert
Related News
•EurekAlert
AI-Powered OCT Enables Rapid 'Optical Biopsy' for Early Endometrial Cancer Detection
A team at Washington University has developed a catheter-based 3D OCT system with AI to quickly and noninvasively detect early endometrial cancers.
•EurekAlert
AI Clinical Reasoning in Diagnostics and Digital Fatigue in Healthcare
Recent JMIR features explore large language models in clinical diagnostics and digital fatigue among healthcare professionals.
•EurekAlert
KAIST, MIT, Microsoft Develop Efficient AI Image Upsampling for Robotics
KAIST, MIT, and Microsoft have created 'Upsample Anything,' a training-free AI method to restore high-resolution visual data from compressed images with up to 16x improved GPU memory efficiency.