Development and Validation of a Machine Learning-Based Dementia Screening Tool: The Six-Question Dementia Screening...

Area of Science:

• Neurology
• Gerontology
• Artificial Intelligence in Medicine

Background:

• Timely dementia detection is critical for mitigating health and societal impacts.
• Existing screening tools like Mini-Mental State Examination (MMSE) and Cognitive Abilities Screening Instrument (CASI) face limitations due to time and resource constraints.

Purpose of the Study:

• To develop and validate a machine learning-based dementia screening tool.
• To assess the efficacy of the Six-Question Dementia Screening Test (6Q-DS) as a rapid screening method.

Main Methods:

• Development of a machine learning model using eXtreme Gradient Boosting.
• Utilized data from 533 older adults at a neurology clinic in Taiwan.
• The Six-Question Dementia Screening Test (6Q-DS), a six-item interview, was employed for data collection.

Main Results:

• The 6Q-DS demonstrated high performance in distinguishing dementia from non-dementia, achieving an AUC of 0.936, sensitivity of 0.879, specificity of 0.951, and accuracy of 0.907.
• For identifying very mild dementia, the 6Q-DS achieved an AUC of 0.874, with sensitivity of 0.818, specificity of 0.805, and accuracy of 0.810.
• Performance was comparable to established tools like MMSE and CASI.

Conclusions:

• The 6Q-DS is a practical, rapid, and user-friendly tool for dementia screening.
• Machine learning application enhances the efficiency of dementia detection.
• The 6Q-DS shows promise as an alternative to traditional screening methods.