Abstract

This paper presents a framework for real-time cognitive fatigue detection among shift workers using an integrated approach that combines photoplethysmography (PPG) data and reaction time analysis with advanced deep learning models, including Long Short-Term Memory (LSTM) networks and Feedforward Neural Networks (FNNs). The system leverages heart rate variability (HRV) and reaction time data to identify fatigue indicators. The results demonstrate significant performance, with the first FNN model achieving a test accuracy of 98.94% and a loss of 0.2928, while the second FNN model achieved the same accuracy with a slightly higher loss of 0.3089. The LSTM model, designed for sequential data processing, exhibited a test accuracy of 88.25% with a loss of 0.4218. These outcomes underline the effectiveness of combining physiological signal processing and deep learning for fatigue prediction. The proposed system offers a non-invasive, real-time monitoring solution, promising to enhance safety and productivity in industries such as transportation, healthcare, and mining. Future work includes further model optimization and integration of additional sensors to improve prediction accuracy and robustness.

Department(s)

Electrical and Computer Engineering

Second Department

Engineering Management and Systems Engineering

Keywords and Phrases

CNN; cognitive fatigue; heart rate variability; LSTM; photoplethysmography; predictive accuracy; real-time monitoring; safety enhancement; shift work

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2025 Institute of Electrical and Electronics Engineers, All rights reserved.

Publication Date

01 Jan 2025

Download

Full Text Link

Share

 
COinS