Intratumoral and Peritumoral Features for Predicting Microvascular Invasion in Hepatocellular Carcinoma: Systematic Review and Meta-Analysis.
Benqi Zhao, Yibei Yu, Yuling Yang et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review and meta-analysis assessed the effectiveness of radiomics-based models in predicting microvascular invasion (MVI) in hepatocellular carcinoma (HCC). The analysis included 26 studies and found that these models had a pooled AUC of 0.85 (95% CI 0.82-0.88), indicating good diagnostic performance, especially when combined with clinical data.
Key findings
- Radiomics models achieved a pooled AUC of 0.85 for predicting MVI in HCC.
- Clinicoradiomics models showed higher performance than radiomics-only models (MRI: 0.87 vs 0.85; CT: 0.87 vs 0.81).
- 3D-based models also demonstrated a pooled AUC of 0.85.
Why it matters
Accurate preoperative prediction of microvascular invasion is crucial for determining prognosis and treatment strategies in hepatocellular carcinoma. Radiomics offers a quantitative approach that may enhance diagnostic accuracy beyond traditional methods.
What to keep in mind
Considerable heterogeneity and limited external validation were noted, indicating a need for methodological standardization.