Development and Validation of an Interpretable Machine Learning Model Based on Gd-EOB-DTPA-enhanced MRI for Evaluating Small HCC (≤2 cm): A Multicenter Cohort Study.

Published in Academic radiology. Abstract: This study aims to develop and externally validate an interpretable machine-learning model that integrates gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid magnetic resonance imaging (MRI) radiomics, qualitative MRI features, and clinical data to accurately distinguish small (≤2 cm) hepatocellular carcinomas from benign hepatic lesions. This retrospective, multicenter cohort included 296 lesions (Center A: 209 [146 training, 63 internal test]; Centers B + C: 87 external validation). Three feature categories were extracted and integrated. A multistep feature selection and model tuning pipeline was performed exclusively within the training set using nested cross-validation to prevent data leakage. Seven machine learning (ML) algorithms were compared. In the external validation set, a reader study was performed to compare the diagnostic performance of the fusion model against that of radiologists. SHapley Additive exPlanations (SHAP) was employed to interpret the optimal model. The fusion model combining three feature categories achieved the best diagnostic performance. In the internal test set, the fusion model yielded an area under the receiver operating characteristic curve (AUC) of 0.913, compared to 0.791 for the radiomics-only model. Its discriminative ability remained robust in an independent external validation cohort (AUC = 0.824), and remained stable after mitigating interscanner variability with ComBat harmonization (AUC = 0.806). The fusion model showed nu...

5 Key Takeaways

1. The fusion model integrating radiomics, qualitative MRI features, and clinical data achieved superior diagnostic performance compared to radiomics-only models in internal testing.
2. The model maintained robust discriminative ability in external validation and after harmonization for interscanner variability.
3. SHAP analysis identified hepatitis B virus status, alpha-fetoprotein, and hepatobiliary phase-derived radiomics features as key predictors of the model's decisions.

OpenRounds transcript synthesis · Verify against the source