DETECTING ACCESS DISCONNECT BY PATTERN RECOGNITION

• US 20090080757A1
• Filed: 09/24/2007
• Published: 03/26/2009
• Est. Priority Date: 09/24/2007
• Status: Active Grant

• Current Assignee: Baxter Healthcare SA (Baxter International Inc.), Baxter International Inc.
• Original Assignee: Baxter Healthcare SA (Baxter International Inc.), Baxter International Inc.
• Inventors: Yardimci, Atif, Roger, Rodolfo G.
• Granted Patent: US 7,995,816 B2
• Time in Patent Office: Days
• US Class Current: 382/134
• CPC Class Codes: A61M 1/3653 Interfaces between patient blood circulation and extra-corporal blood circuit
• CPC Class Codes: A61M 1/3656 Monitoring patency or flow at connection sites; Detecting disconnections
• CPC Class Codes: A61M 1/3659 Cannulae pertaining to extracorporeal circulation
• CPC Class Codes: A61M 2005/1588 having means for monitoring, controlling or visual inspection, e.g. for patency check, avoiding extravasation
• CPC Class Codes: A61M 5/16836 by sensing tissue properties at the infusion site, e.g. for detecting infiltration detecting tissue temperature for diagnostic purposes A61M39/0247

Abstract

A blood detection system is used to detect blood in a patient undergoing an extracorporeal blood therapy, such as hemodialysis or apheresis. The blood detection system uses a digital video device, which may be a camera, such as a standard digital video camera, a low light camera, an infrared camera, or a night vision camera. The digital image is sent to a computer with image processing and pattern recognition software. The software is used to detect whether a pool of blood is forming at the access site. The system is able to detect a pool of blood atop the access site, such as when a leak or a dislodgement of a needle occurs. The system is also able to detect pooling of blood under the skin, as a result, for instance, infiltration of an arteriovenous fistula.

Claims

1. A method of detecting blood leakage, the method comprising:

providing a digital video camera;

mounting the camera to observe an extracorporeal blood processing access site;

taking a baseline image of the access site on a patient;

processing the baseline image with image processing software and saving data of the baseline image;

monitoring the access site by taking additional images of the access site during an extracorporeal blood processing therapy and processing the additional images with the pattern recognition software; and

sending a signal if a pattern of blood is detected.

2. The method of claim 1, wherein the camera is selected from the group consisting of an infrared camera and a night vision camera.

3. The method of claim 1, wherein the signal is sent to a local output device to alert a person or to sound an alarm.

4. The method of claim 1, further comprising placing a contrasting material atop the access site.

5. The method of claim 1, further comprising processing the baseline image for color recognition of the blood and processing the additional images of the access site for color change.

6. The method of claim 1, wherein the step of monitoring includes processing the additional images for pattern recognition, shape analysis or blob analysis.

7. The method of claim 1, further comprising filtering light reaching the camera to selectively pass light with a wavelength of reflected light.

8. A method for detecting blood leakage at an access site, the method comprising:

providing a digital video device;

mounting the device in viewing range of the access site on a patient;

taking a baseline image of the access site;

processing the baseline image with pattern recognition software;

saving data of the baseline image;

monitoring the access site during a therapy procedure by taking additional images and processing the additional images with the pattern recognition software; and

sending a signal if a pattern indicative of a blood leak or infiltration is detected.

9. The method of claim 8, wherein the software uses at least one of machine vision, pattern recognition, blob analysis, and feature shape to determine blood leakage.

10. The method of claim 8, wherein the software interprets a pattern with growing edges and a temperature or color change as blood leakage or infiltration.

11. The method of claim 8, wherein the software processing includes at least one technique selected from the group consisting of machine vision, pattern recognition, blob analysis, feature shape, feature moment, neural networks, object segmentation, object recognition, fuzzy logic, K-nearest neighbor, quadratic classification, a polyphase filter, and linear classification.

12. The method of claim 8, further comprising filtering images taken by the camera to enhance detecting blood.

13. The method of claim 8, wherein the baseline image includes data on colors of the image and wherein the step of monitoring includes color analysis of the additional images.

14. A system for detecting blood leakage, the system comprising:

a digital video device suitable for mounting near an access site for extracorporeal therapeutic blood processing;

signal processing circuitry operably connected to the digital video device;

image processing software on a computer readable medium operably connected to the signal processing circuitry, the image processing software suitable for recognizing a pattern;

a microcontroller in communication with the digital video device and the circuitry; and

a communications circuit in communication with the microcontroller, wherein the digital video device is configured for monitoring the access site during an extracorporeal blood processing therapy and for sending images of the site to the signal processing circuitry, and wherein the image processing software is configured to detect a leakage of blood or infiltration of blood by recognizing a pattern or shape during the therapy.

15. The system for detecting blood leakage according to claim 14, further comprising an output device for raising an alarm or ceasing dialysis if blood is detected.

16. The system for detecting blood leakage according to claim 14, further comprising an optical filter.

17. The system for detecting blood leakage according to claim 14, wherein the software is configured to interpret infrared images as a pattern of temperatures.

18. The system for detecting blood leakage according to claim 14, wherein the image processing software is configured for recognizing patterns and shapes

19. The system for detecting blood leakage according to claim 14, wherein the system is configured for using at least one of machine vision, pattern recognition, blob analysis, and feature shape to determine blood leakage or infiltration.

20. The system for detecting blood leakage according to claim 14, further comprising a hemodialysis machine operably connected to the system for detecting blood leakage.

21. A hemodialysis machine, comprising:

a machine for administering extracorporeal hemodialysis of blood of a patient;

a system for detecting blood leakage or infiltration during hemodialysis, the system operably connected to the machine, the system further comprising;

a digital video camera for monitoring an access site of the patient;

signal processing circuitry for receiving signal data from the camera;

image processing software on a computer readable medium operably connected to the signal processing circuitry;

a microcontroller in communication with the digital video device and the circuitry; and

a communications circuit in communication with the microcontroller, wherein the digital video device is configured for monitoring the access site during an extracorporeal blood processing therapy and for sending images of the site to the signal processing circuitry, and wherein the image processing software is configured to detect a leakage of blood or infiltration of blood that is growing during the therapy.

22. The hemodialysis machine according to claim 21, further comprising a movable mount for the digital video camera.

23. The hemodialysis machine according to claim 21, wherein the software further comprises logic for at least one technique selected from the group consisting of machine vision, pattern recognition, blob analysis, feature shape, feature moment, neural networks, object segmentation, object recognition, fuzzy logic, K-nearest neighbor, quadratic classification, a polyphase filter, and linear classification.

24. The hemodialysis machine according to claim 21, wherein the image processing software is configured for recognizing patterns and shapes.