J.T.M.F Foundation Speech Biomarkers for Cognitive Decline Project

J.T.M.F Foundation Speech Biomarkers for Cognitive Decline Project J.T.M.F Foundation Speech Biomarkers for Cognitive Decline Project One of the main research priorities in Alzheimers disease is the design of scalable assessments of cognitive impairment that can be administered at home, in a clinic, or in community settings. Speech-based elicitations for assessing cognition are easy to collect due to the proliferation of mobile devices. Furthermore, speech production is a cognitively taxing task that requires activation of a distributed neuronal network in the brain; therefore, the hypothesis is that any disturbances to this network due to the presence of a disease will manifest as a change in the speech signal. One of the promises of AI in this context is the potential for using the speech signal to detect an underlying neurological disease by training a classification model to predict a clinical diagnosis. However, there is a gap between the promise of AI and the reality of AI in this context. Our aim in this work is to bridge this gap by validating fully automated acoustic and language metrics relative to gold-standard clinical endpoints and biomarkers.