Your new experience awaits. Try the new design now and help us make it even better

ORIGINAL RESEARCH article

Front. Med.

Sec. Precision Medicine

Volume 12 - 2025 | doi: 10.3389/fmed.2025.1627617

This article is part of the Research TopicIntegrating AI and Machine Learning in Advancing Patient Care: Bridging Innovations in Mental Health and Cognitive NeuroscienceView all 9 articles

Enhancing Mental Health Diagnostics through Deep Learning-Based Image Classification

Provisionally accepted
  • Guangxi University, Nanning, China

The final, formatted version of the article will be published soon.

The integration of intelligence (AI) and machine learning technologies into healthcare, particularly for enhancing mental health diagnostics, represents a critical frontier in advancing patient care. This paper addresses the persistent challenges outlined within the domain, including data scarcity, model interpretability, robustness under domain shifts, and trustworthy decision-making, which are pivotal concerns in the context of mental health and cognitive neuroscience. Traditional approaches to image-based diagnostics have typically struggled with limited adaptability to heterogeneous data sources and insufficient incorporation of domainspecific clinical knowledge, leading to suboptimal generalization and unreliable uncertainty estimation in clinical practice. To bridge these gaps, we propose a novel deep learning framework, MedIntelligenceNet, enhanced with Clinical-Informed Adaptation. MedIntelligenceNet unifies multi-modal data fusion, probabilistic uncertainty quantification, hierarchical feature abstraction, and adversarial domain adaptation into a cohesive model architecture, designed to handle the complex demands of healthcare data. Our Clinical-Informed Adaptation strategy systematically integrates structured clinical priors, symbolic reasoning, and domain alignment techniques to enhance robustness and interpretability. Empirical evaluations conducted on multi-modal mental health datasets demonstrate significant improvements in diagnostic accuracy, model calibration, and resistance to domain shifts compared to baseline deep learning methods. These advancements align closely with the goals of integrating AI and machine learning for better mental healthcare outcomes, contributing towards more personalized, transparent, and reliable diagnostic systems.

Keywords: mental health diagnostics, deep learning, Multi-modal data fusion, uncertainty quantification, Clinical-Informed Adaptation

Received: 13 May 2025; Accepted: 16 Jul 2025.

Copyright: © 2025 Zeng. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

* Correspondence: Ruotong Zeng, Guangxi University, Nanning, China

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.