Differential diagnosis of pathologic grade of hepatocellular carcinoma (HCC) using whole-tumor histogram and texture analysis based on apparent diffusion coefficient (ADC) map

This work aimed for ADC histogram and texture features based strategy to identify poorly differentiated hepatocellular carcinoma (HCC) which may own a high risk of recurrence or metastasis. The results showed that ADC histogram and texture features can discriminate HCC grade. The 95th percentile achieved the best result (AUC: 0.772; sensitivity: 90%; specificity: 72%) on ADC signal intensity map, forming a valuable strategy for clinical practice.