India has introduced MANAS 1 AI, a new artificial intelligence model trained on brainwave data that aims to help doctors detect neurological and psychiatric disorders before symptoms become severe. The development marks a significant step in applying AI to neuroscience and early disease detection.

The brain language foundation model is built on 60,000 hours of brainwave recordings from more than 25,000 patients and is designed to interpret electrical activity in the brain, potentially transforming early diagnosis and treatment.

MANAS 1 AI Trained on Brainwave Data

MANAS 1 AI was developed by Intellihealth (NeuroDx), led by neurologist Dr. Puneet Agarwal, former Professor at the All India Institute of Medical Sciences (AIIMS), along with his research team. The model was launched during an AI summit and released as open source on Hugging Face, with computational support from the Indian AI Mission under the Ministry of Electronics and Information Technology.

Unlike traditional AI models, MANAS 1 AI focuses on interpreting electroencephalogram (EEG) signals — the electrical activity generated by the brain. Built with 400 million parameters, the system is designed as a foundational platform capable of supporting disease-specific AI tools.

According to Dr. Agarwal, the model is built to “understand the basic language of the brain.” He explained that it works similarly to large language models like ChatGPT but is trained on large-scale EEG data to decode neural signals that conventional tests such as MRI scans cannot fully interpret.

Early Detection of Neurological and Psychiatric Disorders

The primary goal of MANAS 1 AI is early disease detection. Researchers say the platform could eventually support diagnostic tools for conditions such as epilepsy, dementia, and other neurological disorders.

If validated at scale, the system could reduce the time between symptom onset and diagnosis — a critical factor in managing brain-related diseases. Earlier detection may help improve treatment outcomes while reducing long-term disability and healthcare costs.

The model also enables researchers to study aspects of brain function that remain poorly understood, potentially advancing scientific understanding of neurological conditions.

Addressing Healthcare Access Challenges in India

A key motivation behind the development of MANAS 1 AI is improving healthcare access. India faces a shortage of neurologists and psychiatrists, particularly in rural and underserved regions, which often leads to delayed diagnosis of brain disorders.

Developers say AI tools built on the platform could assist doctors in primary healthcare facilities, including Ayushman Arogya Mandirs, community health centres, and district hospitals. These tools could support preliminary screening and enable faster referrals to specialists.

However, any disease-specific AI applications derived from MANAS 1 AI would require regulatory approval before clinical use.

Future Developments and Research Potential

The research team has already announced plans for a next-generation model, MANAS 2, expected to be released in the coming weeks. The new version is anticipated to expand capabilities and improve performance in interpreting brain signals.

As artificial intelligence continues to evolve, MANAS 1 AI represents a shift from analyzing human language to decoding the brain’s electrical language. The technology could reshape medical research, diagnosis, and patient care by enabling more proactive healthcare systems.