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

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.

I. Introduction

Mental health problems, including autism spectrum disorder (ASD) and attention-deficit hyperactivity disorder (ADHD), etc., have become a global concern and troubled over 17% of the adult population [1]. In the past decades, computer-aided diagnosis (CAD) approaches have been developed to analyze the high-resolution medical images, such as functional magnetic resonance imaging (fMRI) [2], for addressing the escalating psychiatrist shortage. fMRI can perform a noninvasive, safe measure of brain activity by detecting tiny changes in blood flow. With the developments of graphics processing units and data-driven intelligence, deep learning models are reported to achieve outstanding performance for CAD [3]. The successful application of deep learning models usually relies on a large number of labeled training samples [4].