System and method for monitoring clinical activities
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.
1. A method for tracking a status of a person in a clinical room, wherein more than one person is present at the clinical room, the method comprising:
receiving first image data captured by a first imager from a first viewpoint in the clinical room;
detecting and identifying the person in an image frame from the first image data;
assigning a unique boundary box to the person;
tracking movement of the unique boundary box in subsequent image frames from the first image data by assigning the person to a track nearest to the unique boundary box;
outputting the track to a server to store the track in a database located at server;
receiving second image data captured by a second imager from a second viewpoint in the clinical room, the second imager being oriented with a downward tilt and at a higher location than the first imager;
assigning the unique boundary box to the same person detected in an image frame from the second image data by comparing the first image data and the second image data;
detecting an activity of the same person based on the second image data;
updating a tracker status of the person based on the activity and/or location of the person.
2. The method of claim 1 , further comprising associating boundary boxes of the subsequent image frames that overlap by a predetermined threshold to a same track.
3. The method of claim 1 , further comprising creating a new track for a new person detected at or near an entrance of the clinical room.
4. The method of claim 1 , further comprising deleting an existing track for a previously detected person at or near an entrance of the clinical room.
5. The method of claim 1 , further comprising automatically assigning a contaminated status to a newly detected person in the clinical room.
6. The method of claim 5 , wherein updating the tracker status comprises changing the contaminated status to a clean status in response to detecting a hand hygiene activity by the newly detected person.
7. The method of claim 6 , further comprising ignoring a movement of the newly detected person toward a patient's bed when the tracker status of the newly detected person is a clean status.
8. The method of claim 5 , further comprising outputting an alert on a device in the clinical room in response to the newly detected person with the contaminated status being within a predetermined distance of a patient's bed.
9. The method of claim 8 , wherein the alert comprises an audio alert and/or an alert displayed on the device.
10. The method of claim 8 , further comprising locating a device nearest to the newly detected person for outputting the alert.
11. The method of claim 10 , further comprising retrieving configuration information of all display devices in the clinical room from the database.
12. The method of claim 6 , further comprising training detection of the hand hygiene activity using a dataset of handwashing or hand sanitizing demonstration videos.
13. The method of claim 1 , wherein no image in the first image data or the second image data is stored.
14. The method of claim 1 , wherein only non-image data is transmitted to the server.
15. The method of claim 1 , wherein the first and second imagers each comprise a depth camera.
16. The method of claim 15 , wherein a depth information of the first imager translates to information in a direction of the second imager that is orthogonal to Z axis of the first imager.
17. The method of claim 1 , further comprising detecting occlusion of the person by comprising the first image data and the second image data.
18. The method of claim 17 , wherein occlusion of the detected person is detected based on a different number of boundary boxes in the image frame from the first image data compared to a corresponding image frame from the second image data.
19. The method of claim 1 , wherein tracking the movement of the unique boundary box comprises using a Kalman filter.
20. The method of claim 1 , further comprising receiving third image data captured by a third imager from a third viewpoint in the clinical room.