Back

Small-sized Reasoning Language Models for Linguistic Screening of Alzheimer's Disease

Addepalli, V. r.; Abdalnabi, N.; Kummerfeld, E.; Hembroff, G.; Kiselica, A. M.; Rao, P.; Lee, K.

2026-01-01 health informatics
10.64898/2025.12.24.25342972
Show abstract

Alzheimers disease (AD) is increasing in prevalence, and early detection is essential for timely care. Clinical services face growing demand, leading to delays in diagnostic appointments and increasing the risk of disease progression before evaluation. This work examines artificial intelligence (AI) methods for assessing cognitive status from linguistic features. The proposed architecture uses small language models (SLMs) to analyze speech patterns, and its compact design allows deployment on mobile devices. Recent reasoning-focused models, including Deepseek-R1 and Llama, were evaluated for dementia classification. Multiple fine-tuning strategies were compared, and the best model achieved 91% accuracy and an F1 score. The findings show that AI systems built on SLMs can achieve performance comparable to large language models, indicating their potential as efficient tools that may support health care providers through accessible pre-clinical screening for AD.

Matching journals

The top 10 journals account for 50% of the predicted probability mass.

1
Scientific Reports
based on 701 papers
Top 24%
7.5%
2
PLOS ONE
based on 1737 papers
Top 63%
6.3%
3
Computers in Biology and Medicine
based on 39 papers
Top 0.7%
5.8%
4
Journal of Medical Internet Research
based on 81 papers
Top 3%
5.3%
5
Frontiers in Digital Health
based on 18 papers
Top 0.2%
5.3%
6
IEEE Journal of Biomedical and Health Informatics
based on 14 papers
Top 0.1%
5.0%
7
Frontiers in Aging Neuroscience
based on 22 papers
Top 0.4%
5.0%
8
Alzheimer's Research & Therapy
based on 31 papers
Top 2%
4.7%
9
Journal of Alzheimer's Disease
based on 31 papers
Top 2%
4.5%
10
npj Digital Medicine
based on 85 papers
Top 6%
2.9%
50% of probability mass above
11
Journal of the American Medical Informatics Association
based on 53 papers
Top 3%
2.9%
12
BMC Medical Informatics and Decision Making
based on 36 papers
Top 4%
2.8%
13
Journal of Alzheimer’s Disease
based on 23 papers
Top 2%
2.4%
14
PLOS Digital Health
based on 88 papers
Top 6%
2.4%
15
Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring
based on 28 papers
Top 2%
2.4%
16
Journal of Biomedical Informatics
based on 37 papers
Top 3%
2.3%
17
JAMIA Open
based on 35 papers
Top 4%
2.3%
18
JMIR Medical Informatics
based on 16 papers
Top 3%
2.3%
19
International Journal of Medical Informatics
based on 25 papers
Top 3%
2.2%
20
Frontiers in Artificial Intelligence
based on 11 papers
Top 1%
1.6%
21
JMIR Formative Research
based on 31 papers
Top 3%
1.6%
22
Frontiers in Public Health
based on 135 papers
Top 20%
1.3%
23
Patterns
based on 15 papers
Top 2%
1.3%
24
JMIR Public Health and Surveillance
based on 45 papers
Top 8%
1.3%
25
GeroScience
based on 22 papers
Top 2%
1.2%
26
Biology Methods and Protocols
based on 19 papers
Top 2%
0.8%
27
Informatics in Medicine Unlocked
based on 11 papers
Top 2%
0.8%
28
NeuroImage: Clinical
based on 77 papers
Top 7%
0.8%
29
Frontiers in Neuroscience
based on 29 papers
Top 4%
0.8%
30
Healthcare
based on 14 papers
Top 4%
0.7%