United adversarial learning for liver tumor segmentation and detection of multi-modality non-contrast MRI

• Published in: Medical image analysis Vol. 73; p. 102154
• Main Authors: Zhao, Jianfeng, Li, Dengwang, Xiao, Xiaojiao, Accorsi, Fabio, Marshall, Harry, Cossetto, Tyler, Kim, Dongkeun, McCarthy, Daniel, Dawson, Cameron, Knezevic, Stefan, Chen, Bo, Li, Shuo
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.10.2021; Elsevier BV
• Subjects: Coders; Decision making; Diagnosis; Feature extraction; Hemangioma; Hepatocellular carcinoma; Image processing; Image segmentation; Information processing; Learning; Liver; Liver cancer; Liver tumors segmentation and detection; Magnetic resonance imaging; Medical imaging; Multi-modality NCMRI; Multi-phase radiomics feature; Radiomics; Simultaneous discrimination learning; Tumors; United adversarial learning; United adversarial learning; Liver tumors segmentation and detection; 41A10; 65D05; 65D17; Multi-modality NCMRI; 41A05; Multi-phase radiomics feature
• ISSN: 1361-8415; 1361-8423; 1361-8423
• DOI: 10.1016/j.media.2021.102154

•For the first time, a time-saving, safe, and inexpensive tool achieved simultaneous liver tumors segmentation and detection using multi-modality NCMRI only.•The novel edge dissimilarity feature pyramid module (EDFPM) innovatively extracts the multi-size edge dissimilarity maps to enhance the multi-modality NCMRI feature.•The innovative fusion and selection channel (FSC) fuses the multi-modality NCMRI feature and adaptively makes the final decision of feature selection.•The proposed coordinate sharing with padding (CSWP) mechanism enables the MPRG-D achieves the united adversarial learning for liver tumors segmentation and detection for the first time.•The newly designed multi-phase radiomics guided discriminator (MPRG-D) enhances discrimination by adding the multi-phase radiomics feature. [Display omitted] Simultaneous segmentation and detection of liver tumors (hemangioma and hepatocellular carcinoma (HCC)) by using multi-modality non-contrast magnetic resonance imaging (NCMRI) are crucial for the clinical diagnosis. However, it is still a challenging task due to: (1) the HCC information on NCMRI is insufficient makes extraction of liver tumors feature difficult; (2) diverse imaging characteristics in multi-modality NCMRI causes feature fusion and selection difficult; (3) no specific information between hemangioma and HCC on NCMRI cause liver tumors detection difficult. In this study, we propose a united adversarial learning framework (UAL) for simultaneous liver tumors segmentation and detection using multi-modality NCMRI. The UAL first utilizes a multi-view aware encoder to extract multi-modality NCMRI information for liver tumor segmentation and detection. In this encoder, a novel edge dissimilarity feature pyramid module is designed to facilitate the complementary multi-modality feature extraction. Secondly, the newly designed fusion and selection channel is used to fuse the multi-modality feature and make the decision of the feature selection. Then, the proposed mechanism of coordinate sharing with padding integrates the multi-task of segmentation and detection so that it enables multi-task to perform united adversarial learning in one discriminator. Lastly, an innovative multi-phase radiomics guided discriminator exploits the clear and specific tumor information to improve the multi-task performance via the adversarial learning strategy. The UAL is validated in corresponding multi-modality NCMRI (i.e. T1FS pre-contrast MRI, T2FS MRI, and DWI) and three phases contrast-enhanced MRI of 255 clinical subjects. The experiments show that UAL gains high performance with the dice similarity coefficient of 83.63%, the pixel accuracy of 97.75%, the intersection-over-union of 81.30%, the sensitivity of 92.13%, the specificity of 93.75%, and the detection accuracy of 92.94%, which demonstrate that UAL has great potential in the clinical diagnosis of liver tumors.