AI in the Social and Business World: A Comprehensive Approach

Advancements in Remote Heart Monitoring: Wearable Technology and AI-based Approaches for Cardiovascular Disease Detection

Author(s):

Pp: 102-117 (16)

DOI: 10.2174/9789815256864124010006

* (Excluding Mailing and Handling)

Abstract

In the era of precision medicine and individualized approaches, remote monitoring and control of heart function have emerged as critical components of patient evaluation and management. The integration of consumer-grade software and hardware devices for health monitoring has gained popularity as technological advancements become increasingly integrated into daily life. The cardiology community must adapt to the demands of distant and decentralized care, as highlighted during the COVID-19 pandemic. Wearable technology, such as vital sign monitors, holds significant potential for monitoring heart disease and associated risk factors. This book chapter explores the expanding applications of wearable technology in cardiology, focusing on examples such as Holter-event recording and electrocardiogram (ECG) patches. Textile-based sensors and wristbands are implemented across various patient groups, emphasizing real-time deployment and the evolving role of wearables in arrhythmia, cardiovascular disorders, and associated risk factors. The importance of conducting clinical trials and using proper terminology for clinical validation is also highlighted. To enhance the accuracy and efficiency of ECG signal analysis, this chapter proposes a novel approach that combines AI-based unsupervised Long Short-Term Memory (LSTM) with a recursive-based Ensemble Neural Network (ENN). The LSTM component effectively denoises raw ECG signals and enables faster convergence. The ENN, with its built-in deep layers, provides an improved classification of cardiovascular diseases (CVD) present in the input ECG data. The recursive approach employed by the ENN efficiently utilizes the available parameters, even in the presence of noisy labels. The proposed method demonstrates enhanced prediction and classification of CVD, with high precision, recall, and F1 score. The objective is to derive a checkpoint between clinical and research potentials, identify gaps, and address potential risks associated with CVD detection using ECG measurements. By leveraging wearable technology and advanced AI techniques, clinicians and researchers can benefit from improved diagnostic accuracy, remote patient monitoring, and personalized care. The insights gained from this chapter will contribute to the ongoing advancements in remote heart monitoring and facilitate the adoption of innovative approaches in cardiovascular disease management.

© 2024 Bentham Science Publishers | Privacy Policy