Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Real-Time Automated donning and doffing detection of PPE based on Yolov4-tiny

Anusha Verma, Ghazal Ghajari, K M Tawsik Jawad, Hugh P. Salehi, Fathi Amsaad

TL;DR

Healthcare workers' donning and doffing of PPE is error-prone, risking self-contamination. The authors develop a real-time PPE monitoring system based on YOLOv4-tiny that runs on edge hardware to detect PPE items and enforce correct donning/doffing sequences per WHO guidelines. The work comprises Phase 1 dataset construction with CPPE-5, Phase 2 PPE item detection with a 38-layer YOLOv4-tiny model, and Phase 3 a threshold-driven sequencing algorithm providing immediate feedback. They demonstrate feasibility on a Raspberry Pi 4 with low latency, supporting deployment in offline or remote clinical settings.

Abstract

Maintaining patient safety and the safety of healthcare workers (HCWs) in hospitals and clinics highly depends on following the proper protocol for donning and taking off personal protective equipment (PPE). HCWs can benefit from a feedback system during the putting on and removal process because the process is cognitively demanding and errors are common. Centers for Disease Control and Prevention (CDC) provided guidelines for correct PPE use which should be followed. A real time object detection along with a unique sequencing algorithms are used to identify and determine the donning and doffing process in real time. The purpose of this technical research is two-fold: The user gets real time alert to the step they missed in the sequence if they don't follow the proper procedure during donning or doffing. Secondly, the use of tiny machine learning (yolov4-tiny) in embedded system architecture makes it feasible and cost-effective to deploy in different healthcare settings.

Real-Time Automated donning and doffing detection of PPE based on Yolov4-tiny

TL;DR

Healthcare workers' donning and doffing of PPE is error-prone, risking self-contamination. The authors develop a real-time PPE monitoring system based on YOLOv4-tiny that runs on edge hardware to detect PPE items and enforce correct donning/doffing sequences per WHO guidelines. The work comprises Phase 1 dataset construction with CPPE-5, Phase 2 PPE item detection with a 38-layer YOLOv4-tiny model, and Phase 3 a threshold-driven sequencing algorithm providing immediate feedback. They demonstrate feasibility on a Raspberry Pi 4 with low latency, supporting deployment in offline or remote clinical settings.

Abstract

Maintaining patient safety and the safety of healthcare workers (HCWs) in hospitals and clinics highly depends on following the proper protocol for donning and taking off personal protective equipment (PPE). HCWs can benefit from a feedback system during the putting on and removal process because the process is cognitively demanding and errors are common. Centers for Disease Control and Prevention (CDC) provided guidelines for correct PPE use which should be followed. A real time object detection along with a unique sequencing algorithms are used to identify and determine the donning and doffing process in real time. The purpose of this technical research is two-fold: The user gets real time alert to the step they missed in the sequence if they don't follow the proper procedure during donning or doffing. Secondly, the use of tiny machine learning (yolov4-tiny) in embedded system architecture makes it feasible and cost-effective to deploy in different healthcare settings.
Paper Structure (8 sections, 5 figures, 1 table, 1 algorithm)

This paper contains 8 sections, 5 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Works
  3. Architecture
  4. Phase 1: Dataset and Training Object Detection Model
  5. Phase 2: Object Detection Model for PPE equipment
  6. Phase 3: Doffing/Donning Sequence Analysis
  7. Experimental Results
  8. Conclusion

Figures (5)

  • Figure 1: Architecture of the proposed system
  • Figure 2: Donning(wearing) Sequence Gown, Mask, Googles/Face shield, Gloves
  • Figure 3: Doffing(removing) Sequence Gloves, Googles/Face shield, Gown, Mask- Unfastening Ties Strategy
  • Figure 4: Loss Graph for the Yolov4-tiny model
  • Figure 5: output video snapshots.