Optical-based Measurement of Protection Factors for Powered Air-Purifying Respirators
XINYI NIU, Chandrashekhar Choudhary, Michael Yermakov, Tao Li, Sergey A. Grinshpun, Jun Wang, University of Cincinnati
Abstract Number: 87
Working Group: Health-Related Aerosols
Abstract
Powered air-purifying respirators (PAPRs) are widely used by healthcare workers to protect against inhalation hazards. However, their efficacy can be compromised by a poor initial fit which may progressively deteriorate during work. The traditional respirator fit testing relies on specialized equipment such as TSI PortaCount® that is not designed for a real-time performance monitoring while the wearer is working. Developing a field portable device to conduct real-time performance monitoring of PAPRs is critical since the respirator fit can substantially change during use. A novel device named Exposure Protection Integrated Communicator (EPIC) was prototyped for direct-reading and real-time monitoring of respirators. EPIC employed a pair of optical-based sensors to monitor concentrations inside and outside of a respirator, quantitatively evaluate the protection factor, correct readings with algorithms, and alert wearers if protection is compromised.
The prototype EPIC was evaluated utilizing a manikin headform connected to a breathing simulation system to simulate a sinusoidal breathing pattern of human. The PortaCount® served as the reference measurement device. The outcome was expressed as a protection factor (PF), a ratio of the particle concentrations outside versus inside of the respirator. Flow rates of the tested PAPR were adjusted to achieve PF values of 10, 50, and 100 as reference points. The correlation coefficients (CC) between PFEPIC and PFPortaCount at different ambient concentrations were calculated. A high CC of 0.99 was achieved at the ambient concentration of 50,000 ~ 60,000 particles/L. The high correlation is appealing since EPIC and PortaCount® were based on different measuring principles and particle size ranges. EPIC did have a relatively low CC in extreme situations, and a supervised machine-learning approach will be used to enhance its performance. The results demonstrate the EPIC’s ability to quantitatively monitor the respirator faceseal leakage; it is wearable and does not interfere with job functions.