Assessing MRI-based Artificial Intelligence Models for Preoperative Prediction of Microvascular Invasion in
Xin Han1, Lianfeng Shan2, Rui Xu3
1Shandong University of Traditional Chinese Medicine, Jinan 250355, Shandong, China (X.H., M.L.).
Purpose:
To evaluate the performance of magnetic resonance imaging (MRI)-based artificial intelligence (AI) in the preoperative prediction of microvascular invasion (MVI) in patients with hepatocellular carcinoma (HCC).
Methods:
A systematic search of PubMed, Embase, and Web of Science was conducted up to May 2025, following PRISMA guidelines. Studies using MRI-based AI models with histopathologically confirmed MVI were included. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool and the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework. Statistical synthesis used bivariate random-effects models.
Results:
Twenty-nine studies were included, totaling 2838 internal and 1161 external validation cases. Pooled internal validation showed a sensitivity of 0.81 (95% CI: 0.76-0.85), specificity of 0.82 (95% CI: 0.78-0.85), diagnostic odds ratio (DOR) of 19.33 (95% CI: 13.15-28.42), and area under the curve (AUC) of 0.88 (95% CI: 0.85-0.91). External validation yielded a comparable AUC of 0.85. Traditional machine learning methods achieved higher sensitivity than deep learning approaches in both internal and external validation cohorts (both P < 0.05). Studies incorporating both radiomics and clinical features demonstrated superior sensitivity and specificity compared to radiomics-only models (P < 0.01).
Conclusions:
MRI-based AI demonstrates high performance for preoperative prediction of MVI in HCC, particularly for MRI-based models that combine multimodal imaging and clinical variables. However, substantial heterogeneity and low GRADE levels may affect the strength of the evidence, highlighting the need for methodological standardization and multicenter prospective validation to ensure clinical applicability.
Insights
Magnetic resonance imaging (MRI)-based artificial intelligence (AI) shows high accuracy in predicting microvascular invasion (MVI) in liver cancer (HCC) before surgery. Combining imaging and clinical data improves prediction, though more standardized research is needed.
Area of Science:
- Radiology
- Artificial Intelligence
- Oncology
Background:
- Microvascular invasion (MVI) is a critical prognostic factor in hepatocellular carcinoma (HCC).
- Accurate preoperative prediction of MVI is essential for optimal treatment planning and patient management.
- Current diagnostic methods for MVI have limitations, necessitating advanced predictive tools.
Purpose of the Study:
- To systematically evaluate the diagnostic performance of artificial intelligence (AI) models utilizing magnetic resonance imaging (MRI) for the preoperative prediction of MVI in HCC.
- To assess the pooled accuracy metrics including sensitivity, specificity, and AUC of MRI-based AI models.
- To compare the performance of different AI approaches (e.g., deep learning vs. traditional machine learning) and feature combinations.
Main Methods:
- A systematic literature review was performed across PubMed, Embase, and Web of Science databases up to May 2025, adhering to PRISMA guidelines.
- Studies involving MRI-based AI models for MVI prediction in HCC with histopathological confirmation were included.
- Quality assessment was conducted using QUADAS-2, and statistical synthesis employed bivariate random-effects models.
Main Results:
- Twenty-nine studies with 2838 internal and 1161 external validation cases were analyzed.
- Pooled internal validation demonstrated high performance: sensitivity 0.81, specificity 0.82, and AUC 0.88.
- External validation confirmed comparable performance (AUC 0.85). Traditional machine learning outperformed deep learning, and models combining radiomics with clinical features showed superior accuracy.
Conclusions:
- MRI-based AI models exhibit significant potential for accurate preoperative MVI prediction in HCC.
- Models integrating multimodal imaging data and clinical variables demonstrate enhanced predictive capabilities.
- Substantial heterogeneity and limitations in evidence quality necessitate methodological standardization and prospective multicenter validation for clinical translation.
