Brain-Gut Health Initiative supports AI-assisted diagnosis of psychiatric disorders

Psychiatric disorders affect millions worldwide, but their diagnosis still relies on clinical observation instead of standard biological tests. In an attempt to identify reliable biomarkers, scientists in China have now launched the Brain-Gut Health Initiative, a large-scale longitudinal study combining neuroimaging, electrophysiology, microbiome sequencing, blood biomarkers, and clinical assessments across major psychiatric disorders. The findings revealed coordinated changes linking gut microbes, brain networks and symptoms—supporting the development of AI-assisted diagnosis and personalized therapies.

Psychiatric disorders such as schizophrenia, depression, and bipolar disorder affect roughly one in seven people worldwide. While these disorders pose a major and growing global health challenge, their underlying biological mechanisms remain poorly understood. Current diagnostic practices rely heavily on the evaluation of clinical symptoms rather than underlying causes, highlighting an urgent need for identifying reliable biomarkers that can guide clinical decision-making for improved treatment outcomes.

To fill this gap, a research team comprising Professors Fengchun Wu and Yuanyuan Huang from the Department of Psychiatry at The Affiliated Brain Hospital of Guangzhou Medical University, together with Professor Kai Wu from the South China University of Technology, China, and other researchers, initiated the Brain-Gut Health Initiative (BIGHI). BIGHI is an ongoing long-term clinical study designed to investigate how interactions between the brain and gut microbiome contribute to psychiatric disorders. The findings of their study were published in Volume 9 of Research on January 1, 2026 and made available online on March 3, 2026.

Highlighting the novelty of their study, Prof. Wu, Prof. Huang, and Prof. Kai Wu say, “To the best of our knowledge, BIGHI is the first prospective cohort in China dedicated to investigating the MGBA in psychiatric disorders.”

“Currently, BIGHI includes more than 1,200 participants aged between 18 and 45, diagnosed with psychiatric disorders along with healthy controls,” note the corresponding authors, Prof. Wu, Prof. Huang, and Prof. Kai Wu. “The participants undergo multiple assessments, including clinical evaluations, neurocognitive testing, resting-state electroencephalography, structural and functional magnetic resonance imaging (MRI), blood-based inflammatory and metabolic profiling, fecal genomic sequencing, and a detailed lifestyle and dietary survey, to uncover potential biological markers.”

Early findings of the study suggest that certain features observed in electroencephalography may serve as non-invasive biomarkers indicating the severity of the disease and possible treatment response. For example, the alterations in neural microstates (patterns of the brain’s electrical activity) are linked to improvement in schizophrenia symptoms post-neuromodulation therapy. Similarly, patients with depression often showed reduced alpha-band brain activity, indicating a reduction in relaxed wakefulness.

Neuroimaging investigations also revealed widespread alterations in the brain network structure across different psychiatric conditions. When trained on the MRI data, machine learning models demonstrated high accuracy in distinguishing schizophrenia patients from healthy individuals and identified distinct connectivity patterns that were associated with suicidal thoughts in bipolar disorder and the impact of childhood trauma in depression.

“We also observed distinct changes in gut bacteria within the cohort,” add the corresponding authors. “Patients with psychiatric disorders showed a decrease in beneficial short-chain fatty acid-producing bacteria and an increase in pro-inflammatory microbes. Notably, these microbial shifts were linked to the severity of the symptoms, oxidative stress, and cognitive performance, highlighting the relevance of microbiome alterations in psychiatry.”

One of the most significant contributions of the study lies in the integration of the brain and gut data sets, which helped uncover the underlying mechanisms of various disorders. When the patients were grouped using combined brain and gut data, the brain-derived profiles were more closely related to symptom severity, while gut-based profiles showed stronger links to cognitive performance. Researchers found that differences in gut bacteria were linked to changes in brain functions. The combined analysis of neuroimaging, microbiome, and blood biomarkers also revealed accelerated biological aging in patients with schizophrenia, supporting the growing view that psychiatric disorders can affect multiple body systems rather than only the brain.

While the cohort is currently only based at a single research center and longitudinal follow-up is ongoing, BIGHI represents one of the most comprehensive efforts to characterize psychiatric disorders using integrated multi-omics approaches. The study underscored that psychiatric disorders are highly complex and heterogeneous conditions with distinct pathological features emerging across different systems, such as the gut microbiome, neuroimaging, EEG signals, and blood biomarkers.

The researchers believe that expanding the BIGHI initiative may enable the development of reliable diagnostic tools, microbiome-based therapies, neuromodulation strategies, and AI-driven strategies for managing psychiatric disorders. By providing compelling insights into the microbiota-gut-brain axis in psychiatric disorders, the initiative supports advances in biomarker-driven diagnosis and personalized treatment strategies—paving the way for a better mental healthcare.