ORCATECH Council

When
September 6, 2019
9 a.m. to 12 p.m.
Where
5221 SW Corbett Ave
Portland
,
OR
97239
Contact Information

45.4973002, -122.6860744

RSVP: https://www.surveymonkey.com/r/H8WBGCC for lunch by Tuesday, 9/3.

After introductions and updates we will have two guest speakers.

Presentation Title: MyCogHealth Mobile Assessment Platform to Improve Detection of Cognitive Change

 Jonathan Rush, PhD, Institute on Aging and Lifelong Health, University of Victoria

Presentation Title: Automatic Assessment of Cognitive Tests for Differentiating Mild Cognitive Impairment

Meysam Asgari, Assistant Professor, Center for Spoken Language Understanding, Oregon Health & Science University

Assessing the cognitive changes for early detection of signs indicating the transition from normal cognitive aging to Mild Cognitive Impairment (MCI) is highly valuable, particularly for preventing the transition to more severe stages of dementia such as Alzheimers disease (AD). Current conventional cognitive assessments are limited in their efficiency and sensitivity, often relying on a single score such as the total items correct. Typically, multiple features of response, for example,  the timing of responses or the characteristics of the speech and language contained within each response go uncaptured. The use of digital technologies along with the recent advances in speech and language processing technologies allow for these response characteristics to be readily captured and automatically analyzed thus increasing the potential power of assessments.  In this talk, I will present a computational framework,  composed of an automatic speech recognition (ASR) system and machine learning algorithms, that overcomes some of the limitations of the conventional scoring, and automatically extracts a new set of speech and language “markers” from the audio recordings of test  responses. As a proof of concept, I will ultimately present our recent experimental results on automatically distinguishing participants with MCI from those with intact cognition based on these new speech and language “markers”.