Muhammad Aamir Nashrullah, Pradini Puspitaningayu, Hapsari Peni Agustin Tjahyaningtijas, Nobuo Funabiki, Muhamad Bagus Fikril Alan, I. Made Suartana, Afif Rusdiawan, Donny Ardy Kusuma, Lucy Widya Fathir, Agus Wiyono, Aries Dwi Indriyanti
The rapid advancement of machine learning (ML) and deep learning (DL) technologies has revolutionized various domains, particularly healthcare, by providing innovative solutions to challenges such as fatigue prediction. Fatigue, a multifaceted physiological and psychological condition, significantly impacts performance in sports, transportation, and healthcare. This systematic review investigates the role of artificial intelligence (AI) in predicting fatigue, focusing on the use of heart rate (HR) data as a primary physiological parameter. The study categorizes AI approaches into ML and DL, comparing their accuracy, efficiency, and applicability across various contexts. Parameters such as heart rate variability (HRV), electroencephalography (EEG), electromyography (EMG), eye state, and facial behavior are analyzed for their relevance in predicting fatigue. The integration of wearable devices enables real-time data collection and personalized interventions, facilitating the early detection and management of fatigue. However, challenges such as data variability, model generalization, and ethical considerations like data privacy and security are highlighted as critical areas for future research. ML models such as Support Vector Machine (SVM), Random Forest (RF), and K-Nearest Neighbor (KNN) are noted for their efficacy with smaller datasets, while DL models like Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) excel in analyzing large and complex datasets, achieving superior accuracy. The review underscores the potential of multimodal approaches that integrate multiple data sources to improve predictive capabilities. This comprehensive analysis demonstrates the transformative potential of AI-driven systems in fatigue prediction, enabling advancements in human performance and safety across diverse applications, from sports and transportation to clinical and industrial environments. © 2025, University of Bahrain. All rights reserved.
Faculty of Engineering, Universitas Negeri Surabaya, Surabaya, Indonesia; Dept. of Electrical and Communication Engineering, Okayama University, Okayama, Japan; Faculty of Sport Science, Universitas Negeri Surabaya, Surabaya, Indonesia
