NIMHANS Opens CALM-Brain, India's First Psychiatric Data Repository

NIMHANS, the National Institute of Mental Health and Neurosciences in Bengaluru, has launched CALM-Brain, described as India's first repository of clinical, neuroimaging, behavioural, genetic and other datasets across five psychiatric disorders: addiction, bipolar disorder, dementia, OCD and schizophrenia. Reported in March 2026, the resource already draws on more than 2,000 participants and nearly 900 families, with data spanning fMRI, near-infrared spectroscopy, EEG, eye-tracking and genetic blood testing, linked to a stem-cell biorepository. It was built by the Rohini Nilekani Centre for Brain and Mind, a NIMHANS partnership with NCBS and TIFR, and funded by the Department of Biotechnology and the Pratiksha Trust.

Why a repository is the bottleneck

It is tempting to want the AI model first, but the harder and more valuable thing is the data underneath it. Psychiatric AI, models that might one day help diagnose or stratify mental illness, depends on large, well-curated, multimodal datasets that link brain imaging to genetics to behaviour to clinical outcomes. Those datasets are scarce, expensive to assemble, and overwhelmingly drawn from Western populations, which limits how well any model trained on them transfers to Indian patients.

CALM-Brain is infrastructure, not a product. By assembling India-specific, multimodal psychiatric data in one governed place, it creates the substrate on which AI-driven diagnosis and personalised treatment can later be built. The combination of modalities is the point: a genetic marker means far more when it can be tied to an imaging phenotype and a documented clinical course, and that linkage is exactly what a purpose-built repository provides and scattered hospital records do not.

From data to diagnosis

The repository slots into a wider Indian brain-data effort. IISc's Centre for Brain Research runs the long-running SANSCOG and TLSA aging cohorts that feed India-specific dementia data; the national tele-mental-health programme Tele MANAS, anchored by NIMHANS, has added an AI chatbot and handled millions of calls; and the 2026 Union Budget announced a second NIMHANS to expand the country's mental-health infrastructure. Together these point to mental health being treated as a data and AI problem, not only a clinical-capacity one.

The caveats are real. A repository's value is only as good as the access researchers are granted, the governance that protects intensely sensitive psychiatric data, and the quality and breadth of what is collected; no diagnostic model or clinical outcome has yet come out of it. But assembling the data foundation first, deliberately and with consent and governance built in, is the right order of operations, and it is the unglamorous groundwork on which precision psychiatry in India will eventually stand.