Selecting the most important self-assessed features for predicting conversion to mild cognitive impairment with...

Area of Science:

• Gerontology
• Neuroscience
• Public Health

Background:

• Alzheimer's Disease (AD) is a complex condition where early detection is challenging due to asymptomatic progression.
• Mild Cognitive Impairment (MCI) represents a transitional stage between normal aging and dementia.
• Identifying risk factors for MCI is essential for timely intervention and patient empowerment.

Purpose of the Study:

• To identify significant self-reported features predicting future conversion to MCI.
• To explore the utility of personal information in early MCI detection.
• To inform personalized intervention strategies for healthy aging.

Main Methods:

• Utilized a large-scale, longitudinal study on healthy aging in Spain.
• Employed machine learning (random forest) and permutation-based methods for feature selection.
• Analyzed self-reported personal data including demographics, lifestyle, and cognitive perceptions.

Main Results:

• Subjective cognitive decline emerged as the most critical predictor for MCI conversion.
• Other significant features included educational level, working experience, social engagement, and dietary habits.
• Machine learning models effectively identified key self-reported variables associated with MCI development.

Conclusions:

• Self-reported data, particularly subjective cognitive decline, are vital for predicting MCI.
• These findings support the development of personalized strategies for early MCI detection and management.
• Empowering individuals with knowledge of their risk factors can promote proactive brain health.