Date of Award

12-2024

Document Type

Open Access Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

School Psychology

First Advisor

Boaz Levy

Second Advisor

Amy Cook

Third Advisor

Paul Nestor

Abstract

As Alzheimer’s disease (AD) presents a growing public health challenge, accessible early detection tools are urgently needed. This study used data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (N = 1,536) to evaluate the ability of machine learning models to predict a conversion from normal cognition or mild cognitive impairment (MCI) to AD. A brief neuropsychological battery, including the Alzheimer’s Disease Assessment Scale–Cognitive Subscale (ADAS13), Mini-Mental State Examination (MMSE), Rey Auditory Verbal Learning Test (RAVLT), and Functional Activities Questionnaire (FAQ), was employed. The Gradient Boosted Trees model demonstrated superior performance, achieving an F1 score of 0.88, recall of 0.93, precision of 0.84, accuracy of 0.91, and specificity of 0.91. The model predicted AD onset as early as 10.71 years before clinical diagnosis, with an average lead time of 1.11 years (SD = 2.23 years). AD was accurately predicted on-time (51.13%) or early (35.34%) in 86.47% of cases. Missed diagnoses were associated with younger age, longer time to diagnosis, and the absence of the APOE4 allele. This study underscores the critical role of psychologists in administering brief assessments and providing ongoing care to AD patients. Future research should prioritize improving model generalizability to diverse populations and incorporating multimodal data to enable more personalized diagnostic approaches.

Comments

Free and open access to this Campus Access Thesis is made available to the UMass Boston community by ScholarWorks at UMass Boston. Those not on campus and those without a UMass Boston campus username and password may gain access to this thesis through Interlibrary Loan. If you have a UMass Boston campus username and password and would like to download this work from off-campus, click on the

Download

Share

COinS