From decades to years - AI could speed search for brain drugs hiding in plain sight

Artificial intelligence is being deployed to analyze patient data and identify existing drugs that could be repurposed to treat neurological diseases like motor neurone disease in years rather than decades [1]. Researchers at the UK Dementia Research Institute in Edinburgh use AI algorithms to detect disease patterns in data like voice recordings and eye scans, predicting which of approximately 1,500 already-approved drugs might be effective for brain conditions [1]. Drugs flagged by the AI can then enter clinical trials, such as the MND-SMART trial involving participant Steven Barrett [1]. Barrett, diagnosed with MND 10 years ago, describes the disease as stripping a person of their identity but sees the research as a "bright light" of hope [1]. The approach aims to bypass the typical 10-year-plus timeline for new drug discovery by leveraging compounds with established safety profiles [1]. This computational drug-repurposing strategy builds on a broader trend of using AI in biomedical research. Institutions like the Broad Institute of MIT and Harvard represent long-standing collaborations focused on advancing biomedical and genomic research [8]. Furthermore, the Harvard–MIT Program in Health Sciences and Technology, founded in 1970, is one of the oldest and largest biomedical engineering training programs, underscoring the institutional history of interdisciplinary medical innovation [6]. While AI offers promise, the field has faced setbacks; a recent review of Alzheimer's drugs lecanemab and donanemab found their slowing of disease progression was not significant enough to make a meaningful difference to patients [1]. Despite such challenges, researchers believe the convergence of AI and new technologies marks a tipping point for neurological treatment discovery [1].