System and method for using a video monitoring system to prevent and manage decubitus ulcers in patients

• US 8,675,059 B2
• Filed: 07/29/2010
• Issued: 03/18/2014
• Est. Priority Date: 07/29/2010
• Status: Active Grant

• Current Assignee: Careview Communications Incorporated
• Original Assignee: Careview Communications Incorporated
• Inventors: Johnson, Steve Gail, Clark, Matthew Cameron, Ecker, Stephen, Johnson, Kyle Brook, Kuether, Richard Craig
• Primary Examiner(s): Czekaj, Dave
• Assistant Examiner(s): BRUMFIELD, SHANIKA M
• Application Number:
  US12/804,774
• Publication Number:
  US 20120026308A1
• Time in Patent Office: 1,328 Days
• Field of Search: 340/541, 340/573.1, 340/573.4, 348/143, 348/150, 348/169, 600/587, 600/595, 382/118, 382/159, 382/181, 382/209, 705/2, 705/3, 375/240.16
• US Class Current: 348/77
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 2207/30201 Face
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 7/73 using feature-based methods
• CPC Class Codes: G06V 20/40 in video content extracting overlay text G06V20/62; video retrieval G06F16/70; processing of video elementary streams in video servers H04N21/234; processing of video elementary streams in video clients H04N21/44
• CPC Class Codes: G06V 20/44 Event detection
• 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 40/103 Static body considered as a whole, e.g. static pedestrian or occupant recognition
• CPC Class Codes: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Abstract

A video monitoring system captures image frames of a patient in various positions. The captured image frames are analyzed by the system for changes in a patient'"'"'s position or movement, frames in which the system detects one or both of patient movement and repositioning are retained. The system analyzes an area of interest within each image frame that corresponds to an area in the camera'"'"'s view field with the patient. Sequential image frames are compared for motion, only frames without motion, where the patient is still, are analyzed. Analysis techniques include selecting a plurality of vertical positions in the area of interest, each vertical position corresponding to a part of the patient body indicative of the patient'"'"'s position, horizontally scanning values along the horizontal pixel row at each vertical position in the area of interest, finding an average value for each vertical position and identifying a patient interval of the horizontal pixel row having pixel values different from the average and a representative horizontal position of the patient interval for each vertical interval. Then, comparing those median horizontal positions from a current image frame to the representative horizontal positions from a previous image frame to determine a change in the patient'"'"'s position.

Claims

1. A method for using a video monitoring system to prevent and manage decubitus ulcers in patients, comprising:

receiving a first image frame containing the patient on a support structure from a video camera;

describing a first patient position with regard to a pixel attribute at each of a plurality of vertical positions within the first image frame, the plurality of vertical positions corresponding to different parts of the patient;

receiving a second image frame containing the patient from the video camera, the second image generated by the video camera after the first image;

describing a second patient position with regard to the pixel attribute at each of the plurality of vertical positions within the second image frame;

detecting a patient repositioning event by respectively comparing the pixel attribute at each of the plurality of vertical positions of the second patient position with the pixel attribute at each of the plurality of vertical positions of the first patient position; and

issuing an alert for the patient repositioning event.

2. The method recited in claim 1, further comprises:

identifying the plurality of vertical positions within the first image frame.

3. The method recited in claim 1, wherein the patient is recumbent or seated on the support structure in each of the first image frame and the second image frame.

4. The method recited in claim 1, further compromises:

receiving a description of an area of interest within an image frame at the video monitoring system, said area of interest within the image frame corresponding to a portion of a camera view field containing a patient.

5. The method recited in claim 3, further compromises:

receiving a description of an area of interest within an image frame at the video monitoring system, said area of interest within the image frame corresponding to a portion of a camera view field containing a patient.

6. The method recited in claim 1, wherein describing the first patient position with regard to the pixel attribute at each of the plurality of vertical positions within the first image frame further comprises:

describing the first patient position within the first image frame as a first plurality of horizontal positions within the first image frame, for each of the plurality of vertical positions finding a single horizontal position representative of a position corresponding to the respective body part, comprising;

horizontally scanning pixel values for the pixel attribute along the horizontal pixel row in the first image frame;

finding a representative pixel value for the horizontal pixel row;

finding an interval of pixels for the horizontal pixel row having pixel values different from the representative pixel value;

identifying the interval of pixels as the patient; and

finding a single horizontal position within the interval of pixels representative of the patient position at the vertical position.

7. The method recited in claim 6, wherein detecting the patient repositioning event further comprises:

comparing each of the first plurality of horizontal positions describing the first patient position from the first image frame with a respective each of a second plurality of horizontal positions describing the second patient position from the second image frame.

8. The method recited in claim 5, wherein describing the first patient position with regard to the pixel attribute at each of the plurality of vertical positions within the first image frame further comprises:

describing the first patient position within the area of interest of the first image frame as a first plurality of horizontal positions within the area of interest of the first image frame, for each of the plurality of vertical positions within the area of interest finding a single horizontal position representative of a position corresponding to the respective body part, comprising;

horizontally scanning pixel values for the pixel attribute along the horizontal pixel row within the area of interest of the first image frame;

finding a representative pixel value for the horizontal pixel row;

finding an interval of pixels for the horizontal pixel row having pixel values different from the representative pixel value;

identifying the interval of pixels as the patient; and

finding a single horizontal position within the interval of pixels representative of the patient position at the vertical position.

9. The method recited in claim 1, wherein detecting the patient repositioning event further comprises:

deriving a first plurality of angular relationships describing the first patient position by finding an angular relationship between pairs of horizontal positions at the plurality of vertical positions; and

comparing each of the first plurality of angular relationships describing the first patient position from the first image frame with each of a second plurality of angular relationships describing the second patient position from the second image frame.

10. The method recited in claim 1, wherein detecting the patient repositioning event further comprises:

deriving a first plurality of angular relationships describing the first patient position by finding an angular relationship between pairs of horizontal positions at the plurality of vertical positions;

comparing each of the first plurality of angular relationships describing the first patient position with a library of unique body positions, each of the unique body position described by a unique set of angular relationships between pairs of horizontal positions at the plurality of vertical positions;

identifying a first unique patient body position from the library of unique body positions; and

comparing the first unique patient body position describing the first patient position from the first image frame with a second unique patient body position describing a second patient position from the second image frame.

11. The method recited in claim 10, wherein the pairs of horizontal positions are between adjacent vertical positions.

12. The method recited in claim 5, wherein detecting the patient repositioning event further comprises:

deriving a first plurality of angular relationships describing the first patient position by finding an angular relationship between pairs of horizontal positions within the area of interest at the plurality of vertical positions;

comparing each of the first plurality of angular relationships describing the first patient position with a library of unique body positions, each of the unique body positions described by a unique set of angular relationships between pairs of horizontal positions at the plurality of vertical positions;

identifying a first unique patient body position from the library of unique body positions; and

comparing the first unique patient body position describing the first patient position from the first image frame with a second unique patient body position describing the second patient position from the second image frame.

13. The method recited in claim 12, wherein the pairs of horizontal positions within the area of interest are between adjacent vertical positions.

14. The method recited in claim 1, wherein issuing an alert for the patient repositioning event further comprises:

transmitting the alert for the patient repositioning event to a remote location from the video camera; and

generating one of an audible alarm, a visual alarm, an audible message and a visual message at the remote location from the video camera.

15. The method recited in claim 14, further comprises:

escalating the alert for the patient repositioning event to a second remote location from the video camera by generating one of an audible alarm, a visual alarm, an audible message, a visual message, an email message and a text message invoked at the second remote location.

16. The method recited in claim 1, wherein issuing an alert for the patient repositioning event further comprises:

generating one of an audible alarm, a visual alarm, an audible message and a visual message locally to the video camera.

17. The method recited in claim 6, wherein detecting the patient repositioning event further comprises:

comparing each of the first plurality of horizontal positions describing the first patient position from the first image frame with each of a third plurality of horizontal positions describing a third patient position from a third image frame;

detecting a change between the first plurality of horizontal positions and the third plurality of horizontal positions at the respective vertical positions;

receiving a fourth image frame based on detecting the change between the first plurality of horizontal positions and the third plurality of horizontal positions at the respective vertical positions;

describing a fourth patient position within the fourth image frame as a fourth plurality of horizontal positions within the fourth image frame, for each of the plurality of vertical positions finding a single horizontal position representative of a position corresponding to the respective body part, comprising;

horizontally scanning pixel values for the pixel attribute along the horizontal pixel row in the fourth image frame;

finding a representative pixel value for the horizontal pixel row;

finding an interval of pixels for the horizontal pixel row having pixel values different from the representative pixel value;

identifying the interval of pixels as the patient; and

finding a single horizontal position within the interval of pixels representative of the patient position at the vertical position; and

detecting a second patient repositioning event between the second patient position and the fourth patient position; and

issuing an alert for the second patient repositioning event.

18. The method recited in claim 1, further comprising:

associating information related to a plurality of patient repositioning events with a respective plurality of image frames;

retaining the plurality of image frames; and

creating a video history of the patient'"'"'s positions from the plurality of image frames.

19. The method recited in claim 1, further comprising:

receiving a patient schedule of compliant patient body positions, said patient schedule of compliant patient body positions comprising a plurality of compliant patient body positions and a plurality time intervals that respectively correspond to the plurality of compliant patient body positions; and

comparing the second body position to the patient schedule of compliant patient body positions, wherein the alert is issued based on the second body position not complying with the patient schedule of compliant patient body positions.

20. The method recited in claim 1, wherein the pixel attribute is one of chrominance and luminance.

21. The method recited in claim 1, wherein the pixel attribute is luminance.

22. The method recited in claim 1, wherein the vertical positions are horizontal pixel rows and describing a first patient position with regard to a pixel attribute at each of the plurality of vertical positions within the first image frame further comprises:

scanning pixel values for the pixel attribute in the first image frame;

recognizing a regularly repeating pattern from the pixel values for the pixel attribute;

determining an orientation of the regularly repeating pattern; and

assessing the first patient position from the orientation of the regularly repeating pattern.

23. The method recited in claim 4, wherein the vertical positions are horizontal pixel rows and describing the first patient position with regard to the pixel attribute at each of the plurality of vertical positions within the first image frame further comprises:

subdividing the area of interest into a plurality of equal-sized pixel zones, each of the plurality of equal-sized pixel zones containing an equal number of pixels;

scanning pixel values for the pixel attribute in the area of interest of the first image frame;

assigning at least one patient pixel value;

determining a zone weight for each plurality of equal-sized pixel zones based on pixels contained in a respective zone having the at least one patient pixel value; and

summing the zone weights of the plurality of equal-sized pixel zones within the area of interest of the first image frame.

24. The method recited in claim 23, wherein assigning at least one patient pixel value further comprising:

deriving boundaries between areas of like pixel values;

grouping areas of like pixel values;

selecting one area as the patient, pixels in said one having the at least one patient pixel value.

25. The method recited in claim 24, wherein deriving boundaries between areas of like pixel values further comprising:

identifying elevated contrasts between adjacent pixels based on dissimilar pixel values between the adjacent pixels; and

forming contrast contours bounding areas of like pixel values.

26. The method recited in claim 23, wherein the second patient position is a second sum of zone weights of the plurality of equal-sized pixel zones within the area of interest of the first image frame and detecting a patient repositioning further comprising:

finding a cumulative zone weight difference between the second sum of zone weights and the first sum of zone weights; and

comparing the cumulative zone weight difference to a threshold patient repositioning zone weight.

27. The method recited in claim 23, wherein the pixel attribute is one of chrominance and luminance.

28. The method recited in claim 4, wherein the vertical positions are horizontal pixel rows and describing the first patient position with regard to the pixel attribute at each of the plurality of vertical positions within the first image frame further comprises:

horizontally scanning pixel values for the pixel attribute in the area of interest of the first image frame;

deriving boundaries between areas of like pixel values;

grouping areas of like pixel values;

selecting a patient area having at least one patient pixel value;

identifying a head-end of the patient area;

commencing at the head-end, scanning laterally across the patient area;

identifying at least one of a patient head, a patient neck and a patient shoulder within the patient area;

identifying features of the at least one of a patient head, a patient neck, and a patient shoulder within the patient area; and

assessing the first patient position from the features of the at least one of the patient head, the patient neck, and the patient shoulder.

29. The method recited in claim 28, wherein a patient head is identified and identifying features of the at least one of a patient head, a patient neck, and a patient shoulder within the patient area further comprises:

scanning pixel values for the pixel attribute at the patient head within the area interest of the first image frame;

identifying elevated contrasts between adjacent pixels based on dissimilar pixel values between the adjacent pixels;

forming contrast contours bounding areas of like pixel values; and

identifying one of a patient eye, a pair of patient eyes, a patient ear, a pair of patient ears and an orientation of a pair of patient shoulders.

30. The method recited in claim 28, wherein the pixel attribute is one of chrominance and luminance.

31. The method recited in claim 1, further comprising:

transmitting the first image frame to a remote location from the video camera; and

displaying the first image at the remote location.

32. The method recited in claim 1, wherein the patient is recumbent in one of a patient bed, gurney and medical apparatus in the first image frame.

33. The method recited in claim 1, wherein the patient is seated at a chair, wheelchair and medical apparatus in the first image frame.

34. The method recited in claim 1, wherein each of receiving the first image frame, describing the first patient position, receiving the second image frame, describing the second patient position, detecting, and issuing are performed by the video monitoring system.

35. A method for using a video monitoring system to prevent and manage decubitus ulcers in patients, comprising:

receiving a decubitus ulcer patient repositioning schedule at the video monitoring system, said patient repositioning schedule comprises a plurality of patient positions, each of the plurality of patient positions corresponding to a patient position time interval of a plurality of patient position time intervals, the plurality of patient positions respectively corresponding to a plurality of different body orientations on a support surface;

receiving a plurality of image frames from a surveillance camera at the video monitoring system, each of the plurality of image frames containing the patient on a support structure;

determining a first patient position of the patient with the video monitoring system based on a plurality of vertical positions corresponding to different parts of the patient in a first image frame of the plurality of image frames, the first patient position corresponding to one of the plurality of different body orientations and to one of the plurality of patient position time intervals;

monitoring the time in which the patient remains in the first patient position with the video monitoring system based on the plurality of vertical positions of the patient in one or more image frames of the plurality of image frames received after the first image frame is received;

issuing a first alert with the video monitoring system based on the time in which the patient remains in the first patient position exceeding the patient position time interval that corresponds to the first patient position;

detecting a patient repositioning event with the video monitoring system based on the plurality of vertical positions of the patient in at least one image frame of the plurality of image frames received after the first frame is received and indicating that the patient has moved;

determining whether the patient repositioning event is in compliance with the patient repositioning schedule with the video monitoring system; and

issuing a second alert with the video monitoring system based on the patient repositioning event not being in compliance with the patient repositioning schedule.

36. The method recited in claim 35, further compromises:

receiving a description of an area of interest within an image frame at the video monitoring system, said area of interest within the image frame corresponding to a portion of a camera view field containing a patient.

37. The method recited in claim 35, wherein issuing the first and second alerts for the patient repositioning event further comprises:

transmitting each of the first alert and the second alert to a remote location from the video camera; and

generating one of an audible alarm, a visual alarm, an audible message and a visual message at the remote location from the video camera.

38. The method recited in claim 37, further comprises:

escalating the second alert for the patient repositioning event to a second remote location from the video camera by generating one of an audible alarm, a visual alarm, an audible message, a visual message, an email message and a text message invoked at the second remote location.

39. The method recited in claim 35, wherein issuing the second alert for the patient repositioning event further comprises:

generating one of an audible alarm, a visual alarm, an audible message and a visual message locally to the video camera.

40. The method recited in claim 35, further comprising:

transmitting the first image frame to a remote location from the video camera; and

displaying the first image at the remote location.

41. The method recited in claim 35, wherein the decubitus ulcer patient repositioning schedule is one or more of entered, updated, or verified at the video monitoring system by a user.

42. The method recited in claim 35, further comprising:

associating information related to the patient repositioning event with the at least one image frame; and

retaining the at least one image frame and the associated information.

43. A system for video monitoring to prevent and manage decubitus ulcers in patients, the system comprising:

a video camera configured to generate a plurality of image frames of a patient on a support structure;

an interface; and

a processor connected with memory and configured to;

receive and store a decubitus ulcer patient repositioning schedule, the patient repositioning schedule comprising a plurality of patient positions, each of the plurality of patient positions corresponding to a patient position time interval of a plurality of patient position time intervals, the plurality of patient positions respectively corresponding to a plurality of different body orientations on a support surface;

receive the plurality of image frames from the video camera;

determine a first patient position of the patient on the supporting structure based on a plurality of vertical positions corresponding to different parts of the patient in a first image frame of the plurality of image frames, the first patient position corresponding to one of the plurality of different body orientations and to one of the plurality of patient position time intervals;

monitor the time in which the patient remains in the first patient position based on the plurality of vertical positions of the patient in one or more image frames of the plurality of image frames received after the first frame is received;

issue a first alert with the interface based on the time in which the patient remains in the first patient position exceeding the patient position time interval that corresponds to the first patient position;

detect a patient repositioning event based on the plurality of vertical positions of the patient in at least one image frame of the plurality of image frames received after the first frame is received and indicating that the patient has moved;

determine whether the patient repositioning event is in compliance with the patient repositioning schedule; and

issue a second alert with the interface based on the patient repositioning event not being in compliance with the patient repositioning schedule.