Back

Wearable and Interview-based Assessment of Psychological Risk in Alzheimers Caregivers: Machine Learning vs. Large Language Models

Xiao, J.; Zhao, Z.; King, Z. D.; Khalid, M.; Davies, S.; Zanna, K.; Argueta, D. L.; Brice, K. N.; Wu-Chung, E. L.; Lai, V. D.; Paoletti-Hatcher, J.; Denny, B. T.; Henry, S.; Schulz, P. E.; Fagundes, C. P.; Sano, A.

2026-05-27 psychiatry and clinical psychology
10.64898/2026.05.24.26353993 medRxiv
Show abstract

Spousal caregivers of individuals with Alzheimers disease and related dementias frequently experience elevated perceived stress, caregiver burden, and loneliness, which are associated with adverse health outcomes. Early identification is therefore critical for timely intervention. Existing approaches commonly rely on wearable sensor data and standardized psychological questionnaires, while recent multimodal methods aim to improve prediction by integrating behavioral and linguistic information. In this study, we explored three modality configurations, wearable-derived features, interview-based text, and their combination, to classify caregiver psychological risk using the Perceived Stress Scale (PSS), Zarit Burden Interview, and UCLA Loneliness Scale. We compared traditional machine learning models and large language models (LLMs) (Gemini 2.0, Llama 4, and GPT-4o) under psychometrician-centered and caregiver-centered prompting strategies. Traditional machine learning models performed better under multimodal settings, while LLMs achieved stronger performance with Interview-Only input. We further demonstrate that PSS was the most predictable construct and prompting strategies substantially influenced LLM performance.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

1
npj Digital Medicine
97 papers in training set
Top 0.5%
10.1%
2
Journal of Medical Internet Research
85 papers in training set
Top 0.5%
8.4%
3
Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring
38 papers in training set
Top 0.2%
7.2%
4
Alzheimer's Research & Therapy
52 papers in training set
Top 0.2%
6.8%
5
IEEE Journal of Biomedical and Health Informatics
34 papers in training set
Top 0.3%
4.3%
6
Frontiers in Aging Neuroscience
67 papers in training set
Top 0.8%
4.3%
7
Scientific Reports
3102 papers in training set
Top 27%
4.3%
8
Journal of Alzheimer's Disease
43 papers in training set
Top 0.4%
4.2%
9
Frontiers in Digital Health
20 papers in training set
Top 0.2%
3.6%
50% of probability mass above
10
PLOS ONE
4510 papers in training set
Top 42%
3.1%
11
Frontiers in Psychiatry
83 papers in training set
Top 1%
2.6%
12
Frontiers in Artificial Intelligence
18 papers in training set
Top 0.2%
2.1%
13
IEEE Access
31 papers in training set
Top 0.3%
1.9%
14
Alzheimer's & Dementia: Translational Research & Clinical Interventions
16 papers in training set
Top 0.3%
1.9%
15
NeuroImage
813 papers in training set
Top 4%
1.7%
16
Diagnostics
48 papers in training set
Top 1%
1.7%
17
Computers in Biology and Medicine
120 papers in training set
Top 2%
1.7%
18
Nature Communications
4913 papers in training set
Top 54%
1.5%
19
Nature Medicine
117 papers in training set
Top 3%
1.5%
20
Journal of Alzheimer’s Disease
39 papers in training set
Top 0.7%
1.3%
21
JMIR mHealth and uHealth
10 papers in training set
Top 0.3%
1.2%
22
PLOS Computational Biology
1633 papers in training set
Top 20%
1.2%
23
BMC Medical Informatics and Decision Making
39 papers in training set
Top 2%
1.1%
24
Sensors
39 papers in training set
Top 1%
1.0%
25
Bioengineering
24 papers in training set
Top 1%
0.9%
26
Frontiers in Neuroimaging
11 papers in training set
Top 0.3%
0.8%
27
Human Brain Mapping
295 papers in training set
Top 4%
0.8%
28
Translational Psychiatry
219 papers in training set
Top 4%
0.7%
29
Journal of Biomedical Informatics
45 papers in training set
Top 1%
0.7%
30
NeuroImage: Clinical
132 papers in training set
Top 4%
0.7%