Radiomics and Its Application in Predicting Microvascular Invasion of Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is a global health concern, with increasing morbidity and mortality rates in the world. Microvascular invasion (MVI) is a major cause of early postoperative recurrence of HCC. Predicting MVI before surgery can help doctors develop treatment plans. However, the diagnosis of MVI depends on postoperative pathological verification, which is difficult to predict before surgery. In recent years, more researchers have used radiomics to solve clinical problems. In this study, we used radiomics to predict MVI. We propose a fusion model that combines clinical and radiomics magnetic resonance imaging data to predict MVI of HCC. At present, our fusion prediction model achieves 72.60% accuracy and 0.7607 area under the curve (AUC). In this chapter, we first introduce fundamentals of radiomics and then we present our MVI prediction method using radiomics.