Predicting Dementia Risk Using Paralinguistic and Memory Test Features with Machine Learning Models

Yilun You, Beena Ahmed, Polly Barr, Kirrie Ballard, Michael Valenzuela

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Cognitive reserve exposures are a major class of dementia risk predictors, but a biomarker has proven elusive. Here, we show that paralinguistic features extracted from audio recordings of older participants completing the LOGOS episodic memory test can be used to identify participants with high and low estimable cognitive reserve, and hence low and high dementia risk, respectively. We present a parallel classification system consisting of an ensemble of a k-NN model and SVM model that discriminates between participants at high risk and low risk of dementia with an accuracy of 94.7% when trained with paralinguistic features only and 97.2% when trained with paralinguistic and episodic memory features.

Original languageEnglish
Title of host publication2019 IEEE Healthcare Innovations and Point of Care Technologies, HI-POCT 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages56-59
Number of pages4
ISBN (Electronic)9781728138121
DOIs
Publication statusPublished - Nov 2019
Event2019 IEEE Healthcare Innovations and Point of Care Technologies, HI-POCT 2019 - Bethesda, United States
Duration: 20 Nov 201922 Nov 2019

Publication series

Name2019 IEEE Healthcare Innovations and Point of Care Technologies, HI-POCT 2019

Conference

Conference2019 IEEE Healthcare Innovations and Point of Care Technologies, HI-POCT 2019
CountryUnited States
CityBethesda
Period20/11/1922/11/19

    Fingerprint

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Biomedical Engineering
  • Health Informatics
  • Instrumentation
  • Health(social science)

Cite this

You, Y., Ahmed, B., Barr, P., Ballard, K., & Valenzuela, M. (2019). Predicting Dementia Risk Using Paralinguistic and Memory Test Features with Machine Learning Models. In 2019 IEEE Healthcare Innovations and Point of Care Technologies, HI-POCT 2019 (pp. 56-59). [8962887] (2019 IEEE Healthcare Innovations and Point of Care Technologies, HI-POCT 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/HI-POCT45284.2019.8962887