Deep learning-Based Fit Level Detection for Industrial Respirators with Embedded Breath Sensors

Abstract

This study introduces a deep learning approach for real-time monitoring of respirator fit and breathing conditions using an embedded breath sensor system. Data from controlled experiments, including breathing frequency monitoring and fit test under OSHA protocols, were analyzed to estimate fit quality under different operating conditions. 21 healthy subjects were measured using the proposed embedded breath sensor system on industrial respirators under conditions of proper fitted- and no- sealed mask. Results show that the developed deep learning model reach up to 80% accuracy to detect the fit level of industrial respirators in real-time, enhancing occupational safety and health by enabling continuous, non-intrusive monitoring of working conditions on industrial environments.Clinical Relevance-Proposed system will support to detect impairing of occupational safety & health conditions and related illness.