Source Free Semi-Supervised Transfer Learning for Diagnosis of Mental Disorders on fMRI Scans

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. PP; no. 11; pp. 1 - 18
• Main Authors: Hu, Yao, Huang, Zhi-An, Liu, Rui, Xue, Xiaoming, Sun, Xiaoyan, Song, Linqi, Tan, Kay Chen
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.11.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adaptation models; Attention deficit/hyperactivity disorder; autism spectrum disorder; CAI; Computational modeling; Computer assisted instruction; computer-aided diagnosis; Datasets; Deep learning; Diagnosis; domain shift; Functional magnetic resonance imaging; Health care; Health services; Knowledge management; Labels; Medical diagnostic imaging; Medical imaging; Mental disorders; Neuroimaging; Semi-supervised learning; semi-supervised transfer learning; Solid modeling; source free; Training
• DOI: 10.1109/TPAMI.2023.3298332
• ISSN: 0162-8828; 1939-3539

The high prevalence of mental disorders gradually poses a huge pressure on the public healthcare services. Deep learning-based computer-aided diagnosis (CAD) has emerged to relieve the tension in healthcare institutions by detecting abnormal neuroimaging-derived phenotypes. However, training deep learning models relies on sufficient annotated datasets, which can be costly and laborious. Semi-supervised learning (SSL) and transfer learning (TL) can mitigate this challenge by leveraging unlabeled data within the same institution and advantageous information from source domain, respectively. This work is the first attempt to propose an effective semi-supervised transfer learning (SSTL) framework dubbed S3TL for CAD of mental disorders on fMRI data. Within S3TL, a secure cross-domain feature alignment method is developed to generate target-related source model in SSL. Subsequently, we propose an enhanced dual-stage pseudo-labeling approach to assign pseudo-labels for unlabeled samples in target domain. Finally, an advantageous knowledge transfer method is conducted to improve the generalization capability of the target model. Comprehensive experimental results demonstrate that S3TL achieves competitive accuracies of 69.14%, 69.65%, and 72.62% on ABIDE-I, ABIDE-II, and ADHD-200 datasets, respectively. Furthermore, the simulation experiments also demonstrate the application potential of S3TL through model interpretation analysis and federated learning extension.