Journal article
Development of an Automated and Scalable Virtual Assistant to Aid in PPE Adherence: A Study with Implications for Applications within Anesthesiology
Journal of medical systems, v 48(1), pp 7-7
29 Dec 2023
PMID: 38157145
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
Virtual assistants (VAs) are conversational agents that are able to provide cognitive aid. We developed a VA device for donning and doffing personal protective equipment (PPE) procedures and compared it to live human coaching to explore the feasibility of using VAs in the anesthesiology setting. An automated, scalable, voice-enabled VA was built using the Amazon Alexa device and Alexa Skills application. The device utilized voice-recognition technology to allow a touch-free interactive user experience. Audio and video step-by-step instructions for proper donning and doffing of PPE were programmed and displayed on an Echo Show device. The effectiveness of VA in aiding adherence to PPE protocols was compared to traditional human coaching in a randomized, controlled, single-blinded crossover design. 70 anesthesiologists, anesthesia assistants, respiratory therapists, and operating room nurses performed both donning and doffing procedures, once under step-by-step VA instructional guidance and once with human coaching. Performance was assessed using objective performance evaluation donning and doffing checklists. More participants in the VA group correctly performed the step of "Wash hands for 20 seconds" during both donning and doffing tests. Fewer participants in the VA group correctly performed the steps of "Put cap on and ensure covers hair and ears" and "Tie gown on back and around neck". The mean doffing total score was higher in the VA group; however, the donning score was similar in both groups. Our study demonstrates that it is feasible to use commercially available technology to create a voice-enabled VA that provides effective step-by-step instructions to healthcare professionals.
Metrics
4 Record Views
Details
- Title
- Development of an Automated and Scalable Virtual Assistant to Aid in PPE Adherence: A Study with Implications for Applications within Anesthesiology
- Creators
- Eric Plitman - Toronto Western HospitalEdward Kim - Drexel University, Computer Science (Computing)Rajesh Patel - University of TorontoSeema Kohout - Toronto Western HospitalRongyu Jin - Toronto Western HospitalVincent Chan - University of TorontoMichael Dinsmore - University of Toronto
- Publication Details
- Journal of medical systems, v 48(1), pp 7-7
- Publisher
- Springer Nature
- Number of pages
- 8
- Grant note
- AHSC AFP Innovation Fund
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- Computer Science (Computing)
- Web of Science ID
- WOS:001131905000001
- Scopus ID
- 2-s2.0-85180861965
- Other Identifier
- 991021861297004721
UN Sustainable Development Goals (SDGs)
This publication has contributed to the advancement of the following goals:
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Collaboration types
- Domestic collaboration
- International collaboration
- Web of Science research areas
- Health Care Sciences & Services
- Medical Informatics