Domain-specific information preservation for Alzheimer’s disease diagnosis with incomplete multi-modality neuroimages

• Published in: Medical image analysis Vol. 101; p. 103448
• Main Authors: Xu, Haozhe, Wang, Jian, Feng, Qianjin, Zhang, Yu, Ning, Zhenyuan
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.04.2025
• Subjects: Algorithms; Alzheimer Disease - diagnostic imaging; Alzheimer’s disease; Domain-specific; Generative adversarial network; Humans; Image Interpretation, Computer-Assisted - methods; Incomplete modality; Magnetic Resonance Imaging; Multimodal Imaging - methods; Neural Networks, Computer; Neuroimaging - methods; Incomplete modality; Alzheimer’s disease; Domain-specific; Generative adversarial network
• ISSN: 1361-8415; 1361-8423; 1361-8423
• DOI: 10.1016/j.media.2024.103448

Although multi-modality neuroimages have advanced the early diagnosis of Alzheimer’s Disease (AD), missing modality issue still poses a unique challenge in the clinical practice. Recent studies have tried to impute the missing data so as to utilize all available subjects for training robust multi-modality models. However, these studies may overlook the modality-specific information inherent in multi-modality data, that is, different modalities possess distinct imaging characteristics and focus on different aspects of the disease. In this paper, we propose a domain-specific information preservation (DSIP) framework, consisting of modality imputation stage and status identification stage, for AD diagnosis with incomplete multi-modality neuroimages. In the first stage, a specificity-induced generative adversarial network (SIGAN) is developed to bridge the modality gap and capture modality-specific details for imputing high-quality neuroimages. In the second stage, a specificity-promoted diagnosis network (SPDN) is designed to promote the inter-modality feature interaction and the classifier robustness for identifying disease status accurately. Extensive experiments demonstrate the proposed method significantly outperforms state-of-the-art methods in both modality imputation and status identification tasks. •Proposing a DSIP framework for AD diagnosis with incomplete multi-modal neuroimages.•Devising SIGAN to preserve details and mitigate style gaps for imputing missing data.•Developing SPDN for disease identification via modality-specific information interaction.•Experiments validate DSIP’s efficacy in both imputation and identification tasks.