India has unveiled MANAS 1, a Brain Language Foundation Model built using artificial intelligence and trained on over 60000 hours of brainwave recordings from more than 25000 patients. The initiative aims to enable earlier identification of neurological and psychiatric disorders by analysing complex brainwave data through advanced machine learning techniques.
The model has been developed by Intellihealth NeuroDx, led by neurologist Puneet Agarwal. Dr Agarwal previously served as a professor at the All India Institute of Medical Sciences and has been involved in neurological research and clinical practice for several years. The team stated that MANAS 1 represents a step forward in integrating artificial intelligence with clinical neuroscience.
According to the developers, the model functions as a Brain Language Foundation Model, meaning it has been trained to interpret patterns in electroencephalogram recordings in a way similar to how language models interpret text. By analysing large volumes of brainwave data, the system is designed to detect subtle abnormalities that may not be immediately visible through conventional diagnostic methods.
The project received computational backing under the Indian AI Mission, an initiative of the Ministry of Electronics and Information Technology aimed at strengthening India’s artificial intelligence ecosystem. The support enabled large scale data processing and model training required for handling thousands of hours of clinical recordings.
MANAS 1 was launched during an artificial intelligence summit and has been released as open source on Hugging Face. By making the model openly accessible, the developers intend to encourage collaboration among researchers, clinicians, and technology experts. Open source availability allows independent validation, adaptation, and further development of the model for diverse clinical settings.
Medical experts note that early detection of neurological and psychiatric conditions can significantly improve treatment outcomes. Disorders such as epilepsy, dementia, depression, and other brain related conditions often involve subtle changes in brainwave patterns before visible symptoms develop. An AI system trained on extensive datasets may assist clinicians by providing additional insights during diagnosis.
The team behind MANAS 1 emphasised that the tool is designed to support, not replace, medical professionals. Clinical decisions would continue to rely on comprehensive assessment, including patient history, imaging studies, and laboratory tests. The AI model aims to function as a decision support system, enhancing diagnostic precision.
As artificial intelligence becomes increasingly integrated into healthcare, initiatives such as MANAS 1 highlight India’s growing role in AI driven medical research. Further clinical validation studies and regulatory evaluations are expected before large scale implementation in hospitals.
