AI Vocal Biomarkers to be Studied for MS Diagnosis
Canary Speech and Intermountain Ventures have launched a new study to identify multiple sclerosis using vocal biomarkers. The research will utilize AI-driven technology to analyze speech patterns, aiming to develop a non-invasive method for early detection of the disease.
- Current diagnostic methods for MS are often a lengthy process involving MRI scans, detailed clinical histories, and invasive tests like lumbar punctures (spinal taps). - Canary Speech has previously applied its AI-driven vocal analysis to identify other neurodegenerative diseases, including Huntington's, Alzheimer's, and Parkinson's. - The technology functions by analyzing thousands of vocal features—such as pitch, rhythm, and articulatory coordination—from a voice sample as short as 40 seconds to find subtle patterns indicative of neurological changes. - The IRB-approved study is being led by Dr. Timothy West, a neurologist at Intermountain Health's Salt Lake Clinic, who will collect and analyze voice samples from local patients. - While this study focuses on voice, other applications of AI in MS diagnosis have centered on analyzing medical imaging, with machine learning models achieving high accuracy in detecting MS lesions on MRI scans. - The research aims to address a significant challenge, as MS affects more than 2.9 million people globally, and earlier detection can help slow damage to the central nervous system. - The study is part of a broader field of research into vocal biomarkers for health; for example, MIT Lincoln Laboratory has also researched how speech patterns can help screen for conditions like traumatic brain injury and depression. - Canary Speech's platform is built as an API-first, HIPAA-compliant solution, designed for integration into healthcare systems and applications like telehealth platforms.
