Intelligent Hand Hygiene Support
Overview
We are designing an intelligent computer vision system for remote monitoring, assessment, and support of hand hygiene in hospital environments. By improving hand-wishing quality and compliance, we hope to reduce the rate of hospital-acquired infections due to contaminated equipment, bed linens, or improper patient handling.
We are investigating the use of multiple sensors for the detection, measurement, and evaluation of hand hygiene in controlled laboratory environments, hospital corridors, and patient bedroom units. Our goal is to automatically detect missed hand hygiene events and intervene in real-time to prevent potentially contaminating events. This can include physical contact with patients, handling of biologically hazardous materials, or insufficient hand washing quality.
Our sensors are deployed at two major healthcare partners: Intermountain Healthcare and Lucile Packard Children's Hospital (LPCH) at Stanford. There is continuous active research exploring computer vision technologies and clinical outcome improvement. Our findings have been published in both medical and machine learning venues.
If your hospital organization would like to join this groundbreaking collaboration, we welcome any questions and are happy to facilitate discussion. Contact information is below.
Intelligent Hand Hygiene at Work
Our Computer Vision Approach
Dispenser Usage Detection
With the help of artificial neural networks, our method uses deep learning to automatically detect usage of an alcohol-based sanitizer dispenser from challenging ceiling-mounted top views.
Physical Space Analytics
Intuitive, qualitative results analyze human movement patterns and conduct spatial analytics which convey our method's interpretability. Red regions denote high traffic areas while blue denotes low traffic regions.
Privacy Safe Assessment
To comply with privacy regulations, we use de-identified depth images instead of color photos to track and analyze hand hygiene compliance. Our method can track multiple clinicians throughout a hospital ward.
Pilot Partnerships
We have partnered with Intermountain's Healthcare Transformation Lab where we have deployed 3D depth sensors in several patient rooms. With the help of Intermountain, we are using live data streams to teach our computer vision algorithms to discern events of clinical relevance such as hand hygiene events and patient interaction.
In collaboration with Lucile Packard Children's Hospital, we have installed state-of-the-art sensors in over 10 patient rooms and multiple corridors for hand hygiene activities. Our machine learning algorithms learn routine movement patterns by staff and individual hand hygiene behaviors by guests.
People
Publications
Automatic Detection of Hand Hygiene Using Computer Vision Technology
Amit Singh*, Albert Haque*, Alexandre Alahi, Serena Yeung, Michelle Guo, Jill R Glassman, William Beninati, Terry Platchek, Li Fei-Fei, Arnold Milstein
Towards Vision-Based Smart Hospitals: A System for Tracking and Monitoring Hand Hygiene Compliance
Albert Haque, Michelle Guo, Alexandre Alahi, Serena Yeung, Zelun Luo, Alisha Rege, Amit Singh, Jeffrey Jopling,
Lance Downing, William Beninati, Terry Platchek, Arnold Milstein, Li Fei-Fei
Vision-Based Hand Hygiene Monitoring in Hospitals
Serena Yeung, Alexandre Alahi, Zelun Luo, Boya Peng, Albert Haque, Amit Singh, Terry Platchek, Arnold Milstein, Li Fei-Fei
American Medical Informatics Association (AMIA) Annual Symposium, November 2016
Towards Viewpoint Invariant 3D Human Pose Estimation
Albert Haque, Boya Peng, Zelun Luo, Alexandre Alahi, Serena Yeung, Li Fei-Fei
Recurrent Attention Models for Person Identification
Albert Haque, Alexandre Alahi, Li Fei-Fei
Conference on Computer Vision and Pattern Recognition (CVPR)
June 2016
Vision-Based Hand Hygiene Monitoring in Hospitals
Serena Yeung, Alexandre Alahi, Zelun Luo, Boya Peng, Albert Haque, Amit Singh, Terry Platchek,
Arnold Milstein, Li Fei-Fei