Applications of Machine Learning for Cognitive Health in Older Individuals With HIV: Rapid Systematic Review

Area of Science:

• Artificial Intelligence in Medicine
• Neuroscience Research
• Geriatric Medicine

Background:

• Over half of individuals with HIV are now over 50, facing a significantly higher dementia risk.
• Artificial intelligence (AI), especially machine learning (ML), offers new tools for dementia prediction and management in this population.

Purpose of the Study:

• To systematically review ML applications in dementia research for older adults with HIV.
• To identify gaps and future research directions in this field.

Main Methods:

• Searched PubMed, CINAHL, and Embase for studies within the last 10 years.
• Included original research on people with HIV using ML techniques for dementia-related outcomes.

Main Results:

• 26 out of 721 articles met inclusion criteria; most were retrospective US-based studies.
• Supervised ML techniques were common and showed strong predictive performance for HIV-associated neurocognitive disorders.
• Methodological limitations included small sample sizes, lack of external validation, and limited diversity.

Conclusions:

• Current ML research focuses mainly on HIV-associated neurocognitive disorders, neglecting other dementias like Alzheimer's.
• Future research requires longitudinal, multimodal data, advanced ML, and broader dementia scope for improved detection and management.