Related Experiment Video
Updated: Oct 11, 2025
Author Spotlight: Advancing Hepatic Fibrosis Diagnosis Using Magnetic Resonance Elastography and AI
Published on: July 21, 2023
MRI-based radiomics distinguish different pathological types of hepatocellular carcinoma
Xuehu Wang1, Shuping Wang1, Xiaoping Yin2
1College of Electronic and Information Engineering, Hebei University, Baoding, 071002, China; Research Center of Machine Vision Engineering & Technology of Hebei Province, Baoding, 071002, China; Key Laboratory of Digital Medical Engineering of Hebei Province, Baoding, 071002, China.
Object:
To distinguish combined hepatocellular cholangiocarcinoma (cHCC-CC), hepatocellular carcinoma (HCC) and cholangiocarcinoma (CC) before operation using MRI radiomics.
Method:
This study retrospectively analyzed 196 liver cancers: 33 cHCC-CC, 88 HCC and 75 CC. They had confirmed by pathological analysis in the Affiliated Hospital of Hebei University. MRI lesions were manually segmented by a radiologist.1316 features were extracted from MRI lesions by Pyradiomics. Useful features were retained through two-level feature selection to establish a classification model. Receiver operating characteristic (ROC), area under curve (AUC) and F1-score were used to evaluate the performance of the model.
Results:
Compared with low-order image features, the performance of the model based on high-order features was improved by about 10%. The model showed better performance in identifying HCC tumors during the delay phase (AUC = 0.91, sensitivity = 0.88, specificity = 0.89, accuracy = 0.89, F1-Score = 0.88).
Conclusion:
The classification ability of cHCC-CC, HCC and CC can be further improved by extracting MRI high-order features and using a two-level feature selection method.
