Prediction of cognitive impairment using higher order item response theory and machine learning models

Authors

Lihua Yao, Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, United States.
Yusuke Shono, School of Community and Global Health, Claremont Graduate University, Claremont, CA, United States.
Cindy Nowinski, Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, United States.
Elizabeth M. Dworak, Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, United States.
Aaron Kaat, Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, United States.
Shirley Chen, Advocate Health - MidwestFollow
Rebecca Lovett, Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, United States.
Emily Ho, Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, United States.
Laura Curtis, Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, United States.
Michael Wolf, Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, United States.
Richard Gershon, Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, United States.
Julia Yoshino Benavente, Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, United States.

Affiliations

Aurora St. Luke's Medical Center

Abstract

Timely detection of cognitive impairment (CI) is critical for the wellbeing of elderly individuals. The MyCog assessment employs two validated iPad-based measures from the NIH Toolbox® for Assessment of Neurological and Behavioral Function (NIH Toolbox). These measures assess pivotal cognitive domains: Picture Sequence Memory (PSM) for episodic memory and Dimensional Change Card Sort Test (DCCS) for cognitive flexibility. The study involved 86 patients and explored diverse machine learning models to enhance CI prediction. This encompassed traditional classifiers and neural-network-based methods. After 100 bootstrap replications, the Random Forest model stood out, delivering compelling results: precision at 0.803, recall at 0.758, accuracy at 0.902, F1 at 0.742, and specificity at 0.951. Notably, the model incorporated a composite score derived from a 2-parameter higher order item response theory (HOIRT) model that integrated DCCS and PSM assessments. The study's pivotal finding underscores the inadequacy of relying solely on a fixed composite score cutoff point. Instead, it advocates for machine learning models that incorporate HOIRT-derived scores and encompass relevant features such as age. Such an approach promises more effective predictive models for CI, thus advancing early detection and intervention among the elderly.

Type

Article

PubMed ID

38495777


 

Share

COinS