System and method for monitoring clinical activities
Inventors
Ranasinghe, Nadeesha O. • Al-Ali, Ammar • Kiani, Massi Joe E. • Usman, Mohammad • Zhao, Bowen • de Malliard, Pierre • Min, Peter Sunghwan
Assignees
Publication Number
US-12347202-B2
Publication Date
2025-07-01
Expiration Date
2041-02-12
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
A monitoring system can be configured to monitor activities or actions occurring in clinical settings, such as hospitals. The monitoring system can improve patient safety. The system can use visual and/or other tracking methods. The system can detect and/or identify people in a clinical setting. The system can also track activities of the people, for example, to improve adherence to hygiene protocols.
Core Innovation
The system addresses the problem of low hand hygiene compliance in clinical environments, which leads to healthcare-associated infections (HAI) with significant patient morbidity, mortality, and increased healthcare costs. Conventional monitoring methods are insufficient for consistently ensuring protocol adherence. By integrating video monitoring, audio monitoring, tagging, and tracking modalities, the system provides comprehensive, automatic, and real-time clinical activity monitoring to improve patient safety, reduce HAIs, and facilitate compliance with hygiene protocols.
Claims Coverage
The patent includes one independent claim focusing on a system for monitoring hand hygiene compliance in a clinical room involving multiple cameras and processors using deep learning for detection and tracking.
Deep learning-based hand hygiene activity detection
A first hardware processor receives sequential image frames from a first camera and uses a first deep learning structure to extract features related to handwashing or hand sanitizing. These features are fed into a second deep learning structure that combines outputs across the image frames to determine if hand hygiene has occurred.
Dual viewpoint monitoring with separate hardware processors
The system includes a second camera capturing images from a different viewpoint and a second hardware processor that detects and tracks persons using bounding boxes without storing images, enabling interaction between the first and second processors to determine whether the detected hand hygiene activity corresponds to the tracked person.
Updating hand hygiene status based on detected activities
The first or second hardware processor updates the hand hygiene status of the detected person based on the detected hand hygiene activity, facilitating compliance monitoring.
Motion tracking without image storage
The second hardware processor tracks movements of the person by tracking coordinates of unique bounding boxes without storing any images.
The claims cover a hand hygiene compliance monitoring system in clinical rooms that employs multiple cameras and processors implementing deep learning to detect handwashing activities and identify individuals through bounding box tracking without image storage, facilitating real-time status updates and ensuring hygiene protocol adherence efficiently and privately.
Stated Advantages
Improves patient safety by monitoring adherence to hand hygiene protocols in clinical settings.
Protects privacy by processing images locally and transmitting only non-image data, avoiding storage or transmission of raw images.
Provides real-time detection and alerts of non-compliance with hand hygiene, enabling immediate feedback.
Reduces bandwidth and processing loads by distributing processing across multiple cameras.
Improves accuracy by combining multiple sensor modalities and resolving occlusions through multi-camera coordination.
Documented Applications
Monitoring hand hygiene compliance in hospital or clinical rooms to reduce healthcare-associated infections.
Tracking movements and interactions of clinicians, patients, and visitors within enclosed clinical settings.
Providing real-time alerts for non-compliance events, such as contaminated individuals entering patient zones.
Integrating with patient monitoring systems to display alerts at the nearest bedside display.
Interested in licensing this patent?