Mobile Personal Health Monitoring for Automated Classification of Electrocardiogram Signals in Elderly

Area of Science:

• Biomedical Engineering
• Digital Health
• Cardiology

Background:

• Mobile electrocardiogram (ECG) monitoring shows promise but lacks real-world validation in low- and middle-income countries, especially for the elderly.
• Cardiovascular diseases remain a leading cause of mortality, necessitating accessible and cost-effective monitoring solutions.

Purpose of the Study:

• To develop and validate a wearable ECG monitor with a smartphone application for automated cardiac abnormality detection in older adults.
• To assess the system's accuracy, sensitivity, and specificity in a real-life setting.

Main Methods:

• Development of a wearable ECG monitor with an integrated wireless sensor for signal acquisition.
• Utilization of a custom smartphone application employing machine learning for automated ECG beat classification.
• Testing the system on 100 older adults to evaluate its diagnostic performance.

Main Results:

• The system achieved high accuracy (97%), sensitivity (100%), and specificity (96.6%) in discriminating normal and abnormal ECG signals.
• Demonstrated feasibility of real-life validation for elderly individuals in diverse settings.

Conclusions:

• The wearable ECG monitoring system shows significant potential for detecting cardiac abnormalities in home environments.
• This technology can aid in the prevention, early diagnosis, and effective treatment of cardiovascular diseases, improving healthcare access for older persons.
• The system offers a cost-effective solution for enhanced cardiac care in underserved populations.