IV monitoring by video and image processing

• US 20120013735A1
• Filed: 07/15/2010
• Published: 01/19/2012
• Est. Priority Date: 07/15/2010
• Status: Active Grant

• Current Assignee: Kai Tao
• Original Assignee: Kai Tao
• Inventors: Tao, Kai
• Granted Patent: US 8,531,517 B2
• Time in Patent Office: Days
• US Class Current: 348/135
• CPC Class Codes: H04N 7/188 Capturing isolated or intermittent images triggered by the occurrence of a predetermined event, e.g. an object reaching a predetermined position signal generation from motion picture films H04N5/253

Abstract

We describe a method and apparatus for monitoring the dripping speed in IV process, as well as detecting the its end. The apparatus includes a camera and a processing unit to analyze the information from the acquired video. Features from the image sequence are extracted and the dripping speed is computed accordingly by discrete Fourier transform. The apparatus is also capable of detecting the end of the dripping process by finding the location of liquid surface in the drip chamber. We also describe the use of barcode to provide information to the monitoring device and program.

Claims

1. An apparatus for IV dripping speed monitoring, comprising

a) a containing box that holds the drip chamber

b) an image capturing device, such as a camera, to capture video of the dripping process

c) a processing unit to perform analysis based on the captured video

and a method of dripping speed monitoring based on

a) any measurement from the image sequence that exhibits the same periodicity as the dripping process

b) the use of Fourier analysis methods to determine the dripping speed.

18. A method and apparatus of claim 1, in which the determination of dripping speed using Fourier analysis is done byCapturing a certain number (say, N) of frames per second for a number of seconds (say, L).Extracting from each of the frame a number that is characteristic of a measurement exhibiting the same periodicity as the dripping process, storing the numbers into a vector.Performing discrete Fourier transform on the vector of 2), indexing the total NL spectrum components from 0 to NL−

1.

4) Excluding component 0 which corresponds to the average (constant) part, find the component of the largest complex magnitude among spectrum components

2. (canceled)

3. (canceled)

4. (canceled)

5. (canceled)

6. (canceled)

7. (canceled)

8. (canceled)

9. (canceled)

25. A label or any other printed material of claim 9 that uses barcode to provide information on the type of the drip chamber to the monitoring device and program.

26. A label or any other printed material of claim 9 that uses barcode to provide information on physical measurements of the drip chamber to the monitoring device and program.

10. (canceled)

19. A method and apparatus of claim 18 in which the periodic measurement of claim 18-2 is the vertical position of the drip during its forming and/or falling process, and is obtained by:

1) Selecting an area within the image in which the phenomena is most pronounced, including but not limited to, the area containing the mouth that the drips come out of, or any area in which adequate variation due to drips'"'"' vertical position change can be detected.

2) Enhancing the acquired image in the selected area using spatial domain methods, including but not limited to, sharpening, which in turn including but not limited to, different variations of the gradient method;

or alternatively, enhancing the image using frequency domain methods. Spatial domain and frequency domain methods are equivalent in functionality here.

3) Converting the image, either sharpened by

     2) or not, into binary image by using any possible method including, but not limited to, thresholding. The threshold can be selected using unsupervised method such as Otsu'"'"'s method, or other methods including using a fixed predetermined value.

4) Preserving the larger parts within the binary result of

     3) while eliminating the relatively smaller parts, using morphological erosion, either followed by dilation or not, with any suitable structuring element, or any other method serving the same purpose such as simply dropping the smaller parts;

Or alternatively, in enhanced or unenhanced grayscale image, using the grayscale extension of morphological methods, or any other method such as simply dropping the smaller connected parts (which require the grayscale version of the definition of connectivity).

5) Identifying the location of the drip from binary image obtained either from

     3), or

     3) followed by

     4), by finding the largest connected component, using either 4, 8 or m-connectivity or other stricter or looser connectivity criteria, and computing a number that is characteristic of its vertical position, including but not limited to, the average of its pixels'"'"' vertical coordinates;

or alternatively, finding the largest connected component using the definition of connectivity for grayscale image from

     1) or

     2) or

     4), and computing a number that is characteristic of its vertical position, including but not limited to, the average of its pixels'"'"' vertical coordinates.

6) Any combination, either inclusive or not, of methods

     1)-5), or their equivalent or similar alternatives, in any order comprising logical processing stages, for the purpose of yielding a number characteristic of the drip'"'"'s vertical position.

20. A method and apparatus of claim 18 in which the periodic measurement of claim 18-2 is the size of the drip during its forming and/or falling process, and is obtained by:

1) Selecting an area within the image in which the change of drip size is most pronounced, including but not limited to, the area containing the mouth that the drips come out of, or any area in which adequate variation due to change in drip size can be detected.

2) Enhancing the acquired image in the selected area using spatial domain methods, including but not limited to, sharpening, which in turn including but not limited to, different variations of the gradient method;

or alternatively, enhancing the image using frequency domain methods. Spatial domain and frequency domain methods are equivalent in functionality here.

3) Converting the image, either sharpened by

     2) or not, into binary image by using any possible method including, but not limited to, thresholding. The threshold can be selected using unsupervised method such as Otsu'"'"'s method, or other methods including using a fixed predetermined value.

4) Preserving the larger parts within the binary result of

     3) while eliminating the relatively smaller parts, using morphological erosion, either followed by dilation or not, with any suitable structuring element, or any other method serving the same purpose such as simply dropping the smaller parts;

Or alternatively, in enhanced or unenhanced grayscale image, using the grayscale extension of morphological methods, or any other method such as simply dropping the smaller connected parts (which require the grayscale version of the definition of connectivity).

5) Identifying the location of the drip from binary image obtained either from

     3), or

     3) followed by

     4), by finding the largest connected component, using either 4, 8 or m-connectivity or other stricter or looser connectivity criteria, and computing a number that is characteristic of its size, including but not limited to, the total number of pixels in the largest connected component;

or alternatively, finding the largest connected component using the definition of connectivity for grayscale image from

     1) or

     2) or

     4), and computing a number that is characteristic of its size, including but not limited to, the total number of pixels in the largest connected component.

6) Any combination, either inclusive or not, of methods

     1)-5), or their equivalent or similar alternatives, in any order comprising logical processing stages, for the purpose of yielding a number characteristic of the drip'"'"'s size.

21. A method and apparatus of claim 18 in which the periodic measurement of claim 18-2 is the change of pixels'"'"' average gray level within a certain area, and is obtained by:

1) Selecting an area within the image in which change of pixels'"'"' average gray level due to drips'"'"' forming and/or falling process can be detected, including but not limited to, the area containing the mouth that the drips come out of, or an area that the falling path of the drip crosses, or an area containing part of the liquid surface in which image content variation due to the periodic agitation caused by drips'"'"' falling can be detected, or an area containing a part below the liquid surface in which image content change due to drip'"'"'s entering and disappearing can be detected.

2) Simply compute the average gray level of all pixels in the area.

22. A method and apparatus of claim 18 in which the periodic measurement of claim 18-2 is extracted by:

1) Selecting an area within the image in which image content variation due to the periodic dripping process can be detected, including but not limited to, the area containing the mouth that the drips come out of, or an area that the falling path of the drip crosses, or an area containing part of the liquid surface in which image content variation due to the periodic agitation caused by drips'"'"' falling can be detected, or an area containing a part below the liquid surface in which image content change due to drip'"'"'s entering and disappearing can be detected.

2) For each of area selected in

     1), as the specification has shown, there might be more than one feature that can be extracted to capture the periodicity. For any of the subsequently chosen feature, depending on the quality and characteristics of image, stage(s) of image processing, either for grayscale or binary image, either in spatial or frequency domain, using either standard or modified methods, can be applied. A number characteristic of the dripping process shall in the end be appropriately extracted.

11. (canceled)

12. (canceled)

13. (canceled)

14. (canceled)

15. (canceled)

16. (canceled)

17. (canceled)

23. An apparatus for detecting the end of the dripping process, comprising

a) a containing box that holds the drip chamber

b) an image capturing device, such as a camera, to capture video of the dripping process

c) a processing unit to perform analysis based on the captured video

and a method of liquid surface detection based on

Averaging a sequence of images to eliminate the effect of agitation of the liquid surface due to falling drips.Enhancing the acquired image in the selected area using spatial domain methods, including but not limited to, sharpening, which in turn including but not limited to, different variations of the gradient method, in particular, vertical version of these gradient methods;

or alternatively, enhancing the image using frequency domain methods. Spatial domain and frequency domain methods are equivalent in functionality here.

3) Converting the image, either sharpened by

     2) or not, into binary image by using any possible method including, but not limited to, thresholding. The threshold can be selected using unsupervised method such as Otsu'"'"'s method, or other methods including using a fixed predetermined value.

4) Counting within each row of the binary image obtained from

     3) the maximum number of consecutive pixels, or simply the total number of pixels, and storing the counting result into a vector.

5) Within the whole, or a range, of the vector generated in

     4), find the maximum element which corresponds to the location of the liquid surface. The aforementioned range of the vector could either be determined by a predetermined fixed value, user input, or information acquired from barcode.

And a method of detecting end of the dripping process by comparing the index (row) of the maximum vector element of

     5) with the row index corresponding to the bottom the drip chamber, which triggers the alarm when their difference becomes sufficiently small.

24. A label or any other printed material containing barcode which can read by a camera to provide information to the drip monitoring device and program.