Research on the Mechanisms, Interpretability, and Modeling of Mental Health Disorders with AI Assistance

The rising prevalence of mental health disorders has made them a major public health challenge. These conditions exhibit complex pathological mechanisms and high individual heterogeneity. Conventional research methods are limited in mechanistic depth and diagnostic objectivity, impeding precision medicine. Artificial intelligence (AI), including large language models, offers new pathways for investigating disease mechanisms and building diagnostic or predictive models by leveraging its strengths in data mining and pattern recognition. However, the "black-box" nature of AI leads to insufficient interpretability, which severely restricts its clinical translation. There is an urgent need for systematic research on AI-assisted disease mechanism interpretation, optimization of model explainability, and precision modeling. Furthermore, integrating environmental psychology and environmental exposure data into AI research can help reveal how surroundings influence mental health and interact with neural mechanisms.

This research topic aims to address core challenges in AI (e.g., large language models) applications in depression, mania, and related disorders research, with three primary goals: First, leveraging AI technologies to uncover potential disease pathological mechanisms (e.g., parsing neural circuitry and molecular mechanism correlations based on multimodal data), compensating for the dimensional limitations of traditional research and deepening the understanding of disease; second, breaking through the interpretability bottlenecks of AI models by developing explainable frameworks tailored to mental health scenarios (e.g., feature attribution and logic visualization algorithms), enhancing clinical trustworthiness of models; third, constructing high-precision, highly generalizable disease-related models (e.g., early screening and treatment response prediction models) to promote the clinical implementation of AI technologies, providing scientific tools for mental health disorder diagnosis and treatment, while establishing an interdisciplinary exchange platform to foster collaborative development in the field. Last but not least, integrating environmental psychology perspectives and environmental exposure data into AI-driven mental health research aims to elucidate how environmental factors influence psychological well-being and interact with the neural mechanisms underlying mental disorders.

We welcome contributions addressing, but not limited to, the following areas:

1. Research on AI assisted exploration of neurological mechanisms/pathogenesis of depression, bipolar disorder, and other mental health conditions;
2. Application and algorithmic optimization of Explainable AI in disease-assisted diagnosis and prognostic evaluation models;
3. AI-based disease modeling utilizing multimodal data, including neuroimaging, behavioral information, clinical scales, and other relevant biomarkers;
4. Applications of AI models in disease subtype classification, treatment recommendation, and mechanistic validation;
5. Development of datasets, ethical considerations, and pathways for clinical translation in AI-assisted mental health research.
6. Development of AI-driven multimodal modeling frameworks that integrate environmental psychology indicators (e.g., urban form metrics, green space exposure, sensory environment data) with neuroimaging, behavioral, and clinical data to investigate how environmental factors contribute to the onset, progression, and recovery of depression, bipolar disorder, and other mental health conditions.

We invite submissions from authors with demonstrated expertise in mental health disorder research or AI technology development. Manuscripts should closely align with the scope of this Research Topic, showcase innovation and scientific rigor, and represent original, unpublished work. Interdisciplinary contributions—especially those integrating medicine, environmental psychology, mathematics, economics, and computer science, and supported by experimental or clinical validation—will be given priority.