April 16, 2021 (Los Altos, CA) – EverythingALS, a California nonprofit dedicated to bringing technological innovations and data science to support people with ALS, has been selected to present its scientific research at the American Academy of Neurology (AAN) Virtual Annual Meeting, April 17-22, 2021.
EverythingALS team members, Dr. Aria Anvar and CEO Indu Navar, will be presenting their research, “Towards a Large-Scale Audio-Visual Corpus for Research on Amyotrophic Lateral Sclerosis,” at the world’s top neurology meeting. Basically, the team will present the genesis and steps toward an innovative and scalable audio-visual database for ALS research.
Their research is based on the creation of a large, open data platform comprising speech and video recordings of patients diagnosed with amyotrophic lateral sclerosis (ALS) and healthy controls. Each of the subjects in the collection is interviewed by a virtual agent emulating the role of a neurologist or speech pathologist walking them through speaking exercises.
The collected data is made available to the academic and research community to foster acceleration of the development of biomarkers, diagnostics, therapies, and fundamental scientific understanding of ALS.
Recruitment of patients is done through EverythingALS advocacy with over 2,000 subscribers. Each patient is paired with one of the program’s student ambassadors, who are often pre-medical students rendering support to subjects. Each subject engages in weekly conversations with a virtual avatar, Nina. Interview sessions are done once a week, for about ten minutes per session, and produce video recordings of the subject’s face as well as full-duplex audio recordings, audio and facial measures, measurement of progression using self-reported ALSFRS-R and ROADS, and demographic information.
Results
Within six months, the collection was designed, IRB approved, recruitment launched, and more than 100 subjects participated in regular sessions. EverythingALS is now aiming to include 1,000 subjects to produce the largest and most comprehensive audio-visual ALS database ever made openly available to the research community.
