Smartphone-Based Hand Function Assessment: Systematic Review.

Area of Science:

• Biomedical Engineering
• Digital Health
• Rehabilitation Technology

Background:

• Traditional hand function assessments face challenges in validity, reliability, and data management.
• Smartphones present a cost-effective and accessible solution for objective hand function evaluation.
• Utilizing built-in smartphone sensors can overcome limitations of conventional assessment methods.

Purpose of the Study:

• To systematically review and evaluate existing research on smartphone-based hand function assessment.
• To identify common hand dysfunctions and assessment methods utilizing smartphone technology.

Main Methods:

• A comprehensive literature search was conducted across 8 databases.
• Studies were screened and appraised using the Mixed Methods Appraisal Tool.
• Data extraction focused on study characteristics, sensors, and analytical methods (statistical and machine learning).

Main Results:

• 46 studies were included, identifying 11 types of hand dysfunctions and 6 specific functional impairments.
• Accelerometers and cameras were the most frequently used smartphone sensors.
• Machine learning algorithms showed promise for disease detection, severity evaluation, and prediction.

Conclusions:

• Smartphone-based assessment is a viable and promising approach for evaluating hand function.
• Machine learning is effective for classifying hand dysfunction levels using smartphone data.
• Future research should focus on establishing a gold standard, leveraging multiple sensors, and developing real-time ML applications for rehabilitation.