Differential diagnosis of hepatocellular carcinoma and hepatic hemangioma based on maximum wavelet-coefficient statistics: Novel radiomics features from plain CT

作者:

Highlights:

• A novel radiomics technique to differentially diagnose HCC and hepatic hemangioma.

• Extracting features from plain CT to represent histopathological characteristics.

• Achieved state-of-the-art performance in radiomics diagnoses of HCC and hemangioma.

• Exploration of the associations between the features and histopathological characteristics.

• The features can highlight histopathological differences and enhanced diagnosability.

摘要

•A novel radiomics technique to differentially diagnose HCC and hepatic hemangioma.•Extracting features from plain CT to represent histopathological characteristics.•Achieved state-of-the-art performance in radiomics diagnoses of HCC and hemangioma.•Exploration of the associations between the features and histopathological characteristics.•The features can highlight histopathological differences and enhanced diagnosability.

论文关键词:Medical image processing,Radiomics,Computer vision,Wavelet representation,Liver tumor,Feature engineering

论文评审过程:Received 27 January 2022, Revised 25 May 2022, Accepted 23 July 2022, Available online 5 August 2022, Version of Record 5 August 2022.

论文官网地址:https://doi.org/10.1016/j.ipm.2022.103046