The Capabilities of Machine Learning Radiomics Based Models in the MRI Diagnosis of Early HCC

Purpose : To evaluate machine-learning radiomics based models on enhanced MR images in diagnostics of early HCC. Material and methods : Data from 72 patients with 93 masses who underwent Gadoxetic acid-enhanced MRI scans was retrospectively analyzed. Results : Binary classification models were produced for the differential diagnosis of regenerative and dysplastic nodes, early HCC and HCC nodes with an atypical enhancement with high discriminatory capabilities; the area under the ROC-curve ranged from 0.89 to 0.95 in various models. Conclusion : The performed radiomic models can be used as an effective method for differential diagnostics of HCC with typical and atypical enhancement, dysplastic and regenerative nodes.