Project: Automated short-term longitudinal analysis of speech and language in dementia

Brief Abstract:

Many major symptoms of Alzheimer’s disease (AD) and other dementia are manifested in an individual’s speech. Particularly, patients with mild cognitive impairment, MCI (i.e. preclinical stage of dementia) and early-stage AD display subtle, speech-based symptoms like word-finding difficulties. Unfortunately, current gold standard cognitive assessments are not sensitive enough for detecting these subtle changes in speech that are indicative of cognitive impairment. Winterlight Labs (WLL) has created a speech-based cognitive assessment app that can detect mild changes in speech. WLL collaborated on a research study with VHA Home HealthCare to test their technology as a cognitive assessment for seniors diagnosed with MCI, AD, or other dementia. The study suggests that WLL’s assessment is easy to do in the elderly with dementia, reliable, more engaging, and faster than gold standards. In addition, our algorithms can correctly distinguish between residents with AD and dementia, using their speech profiles alone. Furthermore, WLL’s cognitive assessment scores agree with the gold standard Mini Mental State Exam. Using results from their various home-care studies, WLL has recently developed a cognitive assessment service that they are piloting at senior care settings across North America.

Contact Information:

Sonia Nizzer MSW., RSW
Research Associate
VHA Home HealthCare

Affiliated Organizations: