System and method for predicting patient falls

• US 9,311,540 B2
• Filed: 05/06/2008
• Issued: 04/12/2016
• Est. Priority Date: 12/12/2003
• Status: Active Grant

• Current Assignee: Careview Communications Incorporated
• Original Assignee: Careview Communications Incorporated
• Inventors: Ecker, Stephen, Johnson, Kyle Brook, Johnson, Steve Gail
• Primary Examiner(s): Nguyen, Phuoc
• Assistant Examiner(s): Belani, Kishin G
• Application Number:
  US12/151,452
• Publication Number:
  US 20090278934A1
• Time in Patent Office: 2,898 Days
• Field of Search: 348/152
• US Class Current: 1/1
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G06V 40/25 Recognition of walking or running movements, e.g. gait recognition

Abstract

A patient fall prediction system receives video image frames from a surveillance camera positioned in a patient'"'"'s room and analyzes the video image frames for movement that may be a precursor to a patient fall. In set up phase, the viewpoint of the camera is directed at a risk area associated with patient falls, beds, chairs, wheelchairs, etc. A risk area is defined graphically in the viewport. The patient fall prediction system generates a plurality of concurrent motion detection zones that are situated proximate to the graphic markings of the risk areas. These motion detection zones are monitored for changes between video image frames that indicate a movement. The pattern of detections is recorded and compared to a fall movement detection signature. One fall movement detection signature is a sequential detection order from the motion detection zone closest to the risk area in the frames associated with patient falls, to the motion detection zone farthest away from the risk area. The patient fall prediction system continually monitors the motion detection zones for changes between image frames and compiles detections lists that are compared to known movement detection signatures, such as a fall movement detection signature. Once a match is identified, the patient fall prediction system issues a fall warning to a healthcare provider.

Claims

1. A method for predicting a condition of elevated risk of a fall comprising:

receiving a surveillance viewport of an area from a surveillance camera;

displaying the surveillance viewport on a screen, the surveillance viewport showing a bed on the screen;

receiving a user designation indicating a left risk line and a right risk line of the bed as displayed on the screen, the left risk line and the right risk line associated with an elevated risk of patient falls;

creating a left set of concurrent motion detection zones and a right set of concurrent motion detection zones, the zones of the left set created to be separate from the zones of the right set, each of the left set and the right set created to project laterally outward from the left risk line and the right risk line respectively, each of the left set and the right set respectively comprising at least an innermost zone and an outermost zone, the innermost zone of the left set extending along and adjacent to the left risk line, the innermost zone of the left set positioned between the left risk line and the outermost zone of the left set, the innermost zone of the right set extending along and adjacent to the right risk line, the innermost zone of the right set positioned between the right risk line and the outermost zone of the right set, each of the left set and the right set being automatically created by a computing system based on the user designation indicating the left risk line and the right risk line, wherein each of the left set and the right set respectively comprise at least one detection zone between the innermost detection zone and the outermost detection zone;

monitoring for motion in each zone of each of the left set and the right set based on a plurality of frames sequentially generated by the surveillance camera while viewing the bed;

detecting a plurality of movement signatures of movement traversing one or both of the left set and the right set, the plurality of movement signature comprising an outward movement signature associated with high patient fall risk from the bed and an inward movement signature that is not associated with high patient fall risk from the bed, wherein the outward movement signature is detected based on motion being first detected within the innermost zone of either of the left set or the right set before motion is detected in any other zones of the left set or the right set, and wherein the inward movement signature is detected based on motion being first detected within the outermost zone of either of the left set or the right set before motion is detected in any other zones of the left set or the right set, the plurality of movement signatures detected by the computing system; and

issuing a fall alert with the computing system based on the detection of the outward movement signature.

2. The method of claim 1, wherein the user designation is received from a user input device.

3. The method of claim 1, wherein the user designation comprises placement of a bounding box over the bed as displayed on the screen.

4. The method of claim 3, wherein the user designation comprises manipulation of the bounding box as displayed on the screen to correspond to the orientation of a bed.

5. The method of claim 1, wherein the user designation indicates risk lines which extend along lateral sides of the bed.

6. The method of claim 1, wherein the innermost zone of the left set is coextensive with the left risk line and the innermost zone of the right set is coextensive with the right risk line.

7. The method of claim 1, wherein the left risk line and the right risk line correspond to virtual bedrails of the bed.

8. The method of claim 1, wherein monitoring for motion comprises comparing grayscale or RGB values of the pixels of each zone with grayscale or RGB values of the corresponding zone of a previous frame and calculating a change value.

9. The method of claim 1, wherein the outward movement signature is detected from a chronological order of motion detections that begin with motion in the innermost detection zone of one of the left set or the right set and ends with motion in the outermost detection zone of the left set or the right set.

10. The method of claim 9, wherein the inward movement signature is detected from a chronological order of motion detections that begin with motion in the outermost detection zone of one of the left set or the right set and ends with motion in the innermost detection zone of the left set or the right set.

11. The method of claim 1, wherein the outward detection signature is detected across either of the left set or the right set based on motion being first detected within the innermost detection zone and then being sequentially detected across the at least one detection zone and then being detected within the outermost detection zone.

12. The method of claim 11, wherein the inward detection signature is detected across either of the left set or the right set based on motion being first detected within the outermost detection zone and then being sequentially detected across the at least one detection zone and then being detected within the innermost detection zone.

13. A surveillance system for detecting an elevated fall risk condition, the system comprising:

a surveillance camera configured to generate a plurality of frames showing a surveillance viewport of an area including a bed;

a user interface, the user interface comprising a user input and a display, the display configured to display the surveillance viewport showing the bed, the user input configured to generate a user designation indicating a left risk line and a right risk line of the bed as displayed on the display, the left risk line and right risk line associated with an elevated risk of patient falls; and

a computer comprising memory and logic circuitry configured to;

receive the user designation indicating the left risk line and the right risk line of the bed;

create a left set of concurrent motion detection zones and a right set of concurrent motion detection zones based on the user designation indicating the left risk line and the right risk line of the bed, the zones of the left set created to be separate from the zones of the right set, each of the left set and the right set created to project laterally outward from the left risk line and the right risk line respectively, each of the left set and the right set respectively comprising at least an innermost zone and an outermost zone, the innermost zone of the left set extending along and adjacent to the left risk line, the innermost zone of the left set positioned between the left risk line and the outermost zone of the left set, the innermost zone of the right set extending along and adjacent to the right risk line, the innermost zone of the right set positioned between the right risk line and the outermost zone of the right set, wherein each of the left set and the right set respectively comprise at least one detection zone between the innermost detection zone and the outermost detection zone;

monitor for motion in each zone of each of the left set and the right set based on the plurality of frames;

detect a plurality of movement signatures of movement traversing one or both of the left set and the right set based on the monitoring for motion, the plurality of movement signatures comprising an outward movement signature associated with high patient fall risk from the bed and an inward movement signature that is not associated with high patient fall risk from the bed, wherein the outward movement signature is detected based on motion being detected within the innermost zone of either of the left set or the right set before motion is detected in any other zones of the left set or the right set, and wherein the inward movement signature is detected based on motion being detected within the outermost zone of either of the left set or the right set before motion is detected in any other zones of the left set or the right set; and

issue a fall alert with the user interface based on the detection of the outward movement signature.

14. The system of claim 13, wherein the user designation comprises placement of a bounding box on the bed as displayed on the display.

15. The system of claim 14, wherein the user designation further comprises manipulation of the bounding box on the surveillance viewport to correspond to the orientation of the bed.

16. The system of claim 13, wherein the user designation indicates risk lines which extend along the lateral sides of the bed.

17. The system of claim 13, wherein the computer is configured to create the innermost zone of the left set coextensive with the left risk line and the innermost zone of the right set coextensive with the right risk line.

18. The system of claim 13, wherein the left risk line and the right risk line correspond to virtual bedrails of the bed.

19. The system of claim 13, wherein the computer is configured to monitor for motion by comparing grayscale or RGB values of the pixels of each zone with grayscale or RGB values of the corresponding zone of a previous frame of the plurality of frames and calculate a change value.

20. The system of claim 13, wherein the computer is configured to detect the outward movement signature based on a chronological order of motion detections that begin with motion in the innermost detection zone of one of the left set or the right set and ends with motion in the outermost detection zone of the left set or the right set.

21. The system of claim 20, wherein the computer is configured to detect the inward movement signature based on a chronological order of motion detections that begin with motion in the outermost detection zone of one of the left set or the right set and ends with motion in the innermost detection zone of the left set or the right set.

22. The system of claim 13, wherein the computer is configured to detect the outward detection signature within either of the left set or the right set based on motion being first detected within the innermost detection zone and then being sequentially detected across the at least one detection zone and then being detected within the outermost detection zone.

23. The system of claim 22, wherein the computer is configured to detect the inward detection signature within either of the left set or the right set based on motion being first detected within the outermost detection zone and then being sequentially detected across the at least one detection zone and then being detected within the innermost detection zone.