ML-based semi-automatic quantification of hepatobiliary lesion-to-liver enhancement ratio for benign hepatocellular tumor characterization

Semi-automatic ML quantification of hepatobiliary lesion-to-liver enhancement ratio in benign liver tumors (n=23) correlated strongly with manual values (r=0.69–0.87) and effectively differentiated FNH from HCA, showing feasibility.

• Strong correlation with manual lesion-to-liver contrast enhancement ratio (Pearson r 0.69–0.87).
• Highest OATP expression correlation with 2D ring ROI LLCER (r=0.56).
• Best FNH vs HCA classification with ring-shaped or uniform liver ROI (inverse Fisher ratio 0.28–0.37).

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.