Early coagulation monitoring system of arteriovenous chamber based on image processing

Abstract

The present invention relates to the field of image analysis technology, and specifically to an early coagulation monitoring system for an arteriovenous pot based on image processing. The system obtains an arterial pot image and a venous pot image in real time; obtains the early coagulation coefficient of the arteriovenous pot at the current moment based on the blood hanging on the wall in the arterial pot image and the floating blood in the venous pot image in the current time period; obtains the abnormality of the dialysis pipeline at the current moment based on the bubble distribution in the arterial pot image at the current moment and the change in the liquid level in the arterial pot image and the venous pot image in the current time period; corrects the early coagulation coefficient according to the abnormality of the dialysis pipeline, obtains the corrected early coagulation coefficient at the current moment, and determines whether there is an early coagulation phenomenon at the current moment. The present invention accurately obtains the corrected early coagulation coefficient in real time, which is conducive to timely and accurate detection of early coagulation phenomena and timely treatment, thereby effectively improving the stability of hemodialysis.

Description

Technical Field

[0001] The present invention relates to the technical field of image analysis, and in particular to an early coagulation monitoring system for an arteriovenous chamber based on image processing. Background Art

[0002] During hemodialysis, the arterial and venous cisterns are two key components of the dialysis circuit. These cisterns prevent air from entering the patient's blood vessels through multi-stage bubble interception. Medical staff also assess early coagulation by observing blood clinging to the cisterns and floccules in the venous cisterns. However, in practice, abnormal pressure changes caused by dialysis tubing can affect these clinging conditions, making it impossible to accurately monitor early coagulation based solely on these conditions. Summary of the Invention

[0003] In order to solve the technical problem that early coagulation cannot be accurately monitored directly based on the blood sticking to the wall of the arterial bottle and the floating blood in the venous bottle, the purpose of the present invention is to provide an early coagulation monitoring system for the arteriovenous bottle based on image processing. The technical solution adopted is as follows:

[0004] An embodiment of the present invention provides an early coagulation monitoring system for an arteriovenous cisternae based on image processing, the system comprising the following steps:

[0005] An image acquisition module is used to acquire an arterial pot image and a venous pot image at each moment in real time;

[0006] An early coagulation coefficient acquisition module is used to obtain the early coagulation coefficient of the arteriovenous pot at the current moment based on the blood sticking to the wall in each arterial pot image and the floating blood in each venous pot image in the current time period;

[0007] The dialysis tubing abnormality degree acquisition module is used to acquire the abnormality degree of the dialysis tubing at the current moment based on the distribution of bubbles in the arterial pot image at the current moment and the changes in the liquid level in the arterial pot image and the venous pot image during the current time period;

[0008] The early coagulation monitoring module is used to correct the early coagulation coefficient according to the abnormality of the dialysis tube, obtain the corrected early coagulation coefficient of the arteriovenous pot at the current moment, and determine whether there is early coagulation phenomenon at the current moment.

[0009] Furthermore, the method for obtaining the early coagulation coefficient is:

[0010] Obtain the blood hanging coefficient of each arterial bottle image in the current time period according to the liquid level height, the color of the blood hanging area and the change trend of the blood hanging height in the arterial bottle image at each moment in the current time period;

[0011] According to the color significance and size of the floating catkins area in the venous pot image at each moment in the current time period, the floating catkins coefficient of each venous pot image in the current time period is obtained;

[0012] According to the correlation between the wall hanging coefficient and the floating coefficient in the current time period, as well as the wall hanging coefficient and the floating coefficient at the current moment, the early coagulation coefficient of the arteriovenous kettle at the current moment is obtained.

[0013] Furthermore, the wall hanging coefficient is obtained by:

[0014] For the artery pot image at any moment in the current time period, a blood region in the artery pot image is obtained as a first blood region through HSV color space segmentation and morphological optimization;

[0015] A straight line on the artery wall without the catheter in the artery image is obtained by Hough transform detection as a reference straight line, all line segments parallel to the reference straight line are extracted in the first blood region as reference line segments, and the average length of all reference line segments is used as the liquid level in the artery image;

[0016] When the liquid level is equal to the preset liquid level, the wall hanging coefficient of the arterial pot image defaults to 0;

[0017] When the liquid level is lower than the preset liquid level, a connected region with a grayscale value variance within a preset variance range within a specified range above the first blood region is regarded as a blood wall region;

[0018] Obtaining the color conspicuity of the blood wall area according to the difference in R channel values between the blood wall area and the first blood area;

[0019] The length of the longest line segment parallel to the reference line in the blood wall area is used as the wall height, and the degree of change of the wall height of the arterial pot image is obtained based on the difference between the wall heights of the arterial pot image and the arterial pot images in its preset neighborhood.

[0020] The product of the color obviousness and the degree of change in the wall height is normalized to obtain a result as the wall coefficient of the artery pot image.

[0021] Furthermore, the method for obtaining the color obviousness is:

[0022] Obtain the mean R channel value of the pixel points in the blood wall area as the first value;

[0023] Obtaining an average of the R channel values of the pixels in the first blood area as the second value;

[0024] The ratio of the first value to the second value is used as the color obviousness of the blood wall area.

[0025] Furthermore, the method for obtaining the degree of change in the wall hanging height is:

[0026] using a preset number of arterial pot images that are closest to the arterial pot image in time sequence as preset neighboring arterial pot images of the arterial pot image;

[0027] Arrange the wall heights of the artery pot image and its preset neighboring artery pot images according to the time sequence of the corresponding artery pot images to obtain a wall height sequence;

[0028] The slope of the straight line fitted by the wall height in the wall height sequence is obtained as the degree of change of the wall height of the arterial pot image.

[0029] Furthermore, the method for obtaining the floating coefficient is:

[0030] For the venous pot image at any moment in the current time period, the blood area in the venous pot image is extracted as the second blood area through HSV color space segmentation. The second blood area is segmented using the Otsu threshold segmentation algorithm, and the high grayscale area is regarded as the floating catkin area.

[0031] The area except the floating area in the second blood area is used as the reference blood area;

[0032] Obtaining a difference between a mean grayscale value of pixels in the floating catkin area and a mean grayscale value of pixels in the reference blood area as a color significance of the floating catkin area;

[0033] The product of the color prominence and the area of the floating area is normalized to obtain a floating coefficient of the venous pot image.

[0034] Furthermore, the method for obtaining the early coagulation coefficient is:

[0035] Arrange the wall coefficients in the current time period according to the time sequence of the corresponding arterial pot images to obtain a wall coefficient sequence;

[0036] Arrange the floating coefficients in the current time period according to the time sequence of the corresponding venous pot images to obtain a floating coefficient sequence;

[0037] The Pearson correlation coefficient between the wall hanging coefficient series and the floating coefficient series is used as the first eigenvalue;

[0038] The average of the wall hanging coefficient and the floating coefficient at the current moment is used as the coagulation analysis value at the current moment;

[0039] The result of normalizing the product of the coagulation analysis value and the first eigenvalue is used as the early coagulation coefficient of the arteriovenous pump at the current moment.

[0040] Furthermore, the method for obtaining the abnormality degree of the dialysis tubing is:

[0041] The connected domains in the first blood region in the arterial pot image at the current moment are all regarded as bubbles, and the area of each bubble is obtained by the Hough circle detection algorithm, and is regarded as the first area;

[0042] The sum of all first areas is used as the first abnormal value of the dialysis tubing;

[0043] Obtain the difference between the liquid level in the arterial pot image at each moment in the current time period and the next adjacent moment, and use them as height analysis change values;

[0044] The variance of the height analysis change value is taken as the second abnormal value of the dialysis pipeline;

[0045] For the intravenous pot image at any moment in the current time period, a straight line without the catheter side wall in the intravenous pot image is obtained through Hough transform detection as a target straight line. All line segments parallel to the target straight line in the second blood region of the intravenous pot image are extracted as target line segments, and the average length of all target line segments is used as the liquid level in the intravenous pot image;

[0046] Obtain the variance of the liquid level in all venous bottle images in the current time period as the third abnormal value of the dialysis pipeline;

[0047] The result of normalizing the product of the first abnormal value of the dialysis tubing, the second abnormal value of the dialysis tubing, and the third abnormal value of the dialysis tubing is used as the abnormality degree of the dialysis tubing at the current moment.

[0048] Furthermore, the method for obtaining the corrected early coagulation coefficient is:

[0049] The result of negative correlation of the abnormality degree of the dialysis tubing is used as a correction weight;

[0050] The product of the correction weight and the early coagulation coefficient is used as the corrected early coagulation coefficient of the arteriovenous pump at the current moment.

[0051] Furthermore, the method for determining whether there is early coagulation phenomenon at the current moment is:

[0052] When the corrected early coagulation coefficient is greater than the preset early coagulation coefficient threshold, it is determined that early coagulation occurs at the current moment;

[0053] When the corrected early coagulation coefficient is less than or equal to the preset early coagulation coefficient threshold, it is determined that no early coagulation phenomenon exists at the current moment.

[0054] The present invention has the following beneficial effects:

[0055] The present invention first obtains the early coagulation coefficient of the arteriovenous pot at the current moment based on the blood hanging on the wall in each arterial pot image and the floating blood in each venous pot image in the current time period, which preliminarily reflects the possibility of early coagulation at the current moment; in order to reduce the interference of dialysis pipeline abnormality on the analysis of early coagulation, the abnormality degree of the dialysis pipeline at the current moment is obtained according to the distribution of bubbles in the arterial pot image at the current moment and the change of liquid level in the arterial pot image and the venous pot image in the current time period, accurately reflecting the abnormality of the dialysis pipeline at the current moment, which is conducive to the subsequent accurate correction of the early coagulation coefficient and improving the accuracy of early coagulation monitoring; therefore, the early coagulation coefficient is corrected by the abnormality degree of the dialysis pipeline, and the corrected early coagulation coefficient of the arteriovenous pot at the current moment is accurately obtained, which accurately reflects the possibility of early coagulation at the current moment; and then based on the corrected early coagulation coefficient, it is accurately judged whether there is early coagulation phenomenon at the current moment, effectively improving the accuracy of early coagulation monitoring, and being conducive to timely and accurate discovery of early coagulation phenomenon and timely treatment, thereby improving the stability of hemodialysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0057] Figure 1 This is a structural block diagram of an early coagulation monitoring system for an arteriovenous cuff based on image processing provided by one embodiment of the present invention;

[0058] Figure 2 A flow chart of a method for obtaining an early coagulation coefficient provided by one embodiment of the present invention;

[0059] Figure 3 A schematic diagram of a computer device provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0060] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effectiveness of an early coagulation monitoring system for an arteriovenous cuff based on image processing proposed by the present invention. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0061] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0062] The specific scheme of the early coagulation monitoring system of the arteriovenous cuff based on image processing provided by the present invention is described in detail below with reference to the accompanying drawings.

[0063] Example 1:

[0064] This invention proposes an early coagulation monitoring system for arteriovenous blood clots based on image processing. Figure 1 , which shows a structural block diagram of an early coagulation monitoring system for an arteriovenous cuvette based on image processing provided by an embodiment of the present invention. The system includes: an image acquisition module 10, an early coagulation coefficient acquisition module 20, a dialysis tubing abnormality degree acquisition module 30 and an early coagulation monitoring module 40.

[0065] The image acquisition module 10 is used to acquire the arterial pot image and the venous pot image at each moment in real time.

[0066] Specifically, the hemodialysis process involves a blood pump drawing blood from the patient's body through an arterial chamber into the dialyzer. The blood then passes through the dialyzer for toxin removal before entering a venous chamber, where it is returned to the patient. The arteriovenous chamber traps air bubbles in the blood to prevent air from entering the patient's body. It also monitors blood pressure to ensure smooth, unobstructed blood flow and prevent severe clotting.

[0067] To promptly and accurately detect early coagulation in the arteriovenous cuff, allowing medical staff to intervene promptly and effectively avoid the effects of coagulation, this embodiment uses a medical camera to capture real-time images of the arterial and venous cuffs at each moment. It should be noted that during the acquisition process, the medical camera is mounted on a fixed, anti-shake bracket to prevent blurring of the captured images. A small fill light is installed, the surfaces of the arterial and venous cuffs are disinfected with alcohol, and the positions and camera angles of the cuffs are adjusted to ensure a clear, unobstructed, and non-reflective viewing position. This ensures that the captured images of the arterial and venous cuffs are free of reflections and surface contamination. This embodiment analyzes a complete hemodialysis process as an example. The time interval between two adjacent moments is set to 1 second in this embodiment. Implementers can adjust the time interval based on actual circumstances and are not limited here.

[0068] As is known, this embodiment uses real-time analysis of the arteriovenous clot to promptly detect early coagulation. Therefore, to improve analysis efficiency, this embodiment uses a semantic segmentation network to obtain an arterial clot image containing only the arterial clot region and a venous clot image containing only the venous clot region. The semantic segmentation network of this embodiment uses a U-net neural network. The input is a denoised arterial clot image and a venous clot image; the output is an arterial clot image containing only the arterial clot region and a venous clot image containing only the venous clot region. The U-net neural network is trained and labeled by marking the areas to be detected, namely the arterial clot region and the venous clot region, as 1, and other areas as 0. The loss function of the U-net neural network is the cross-entropy loss function. U-net neural networks are well known and will not be described in detail.

[0069] It should be noted that the subsequent arterial pot images are images containing only the arterial pot and the venous pot images are images containing only the venous pot.

[0070] The early coagulation coefficient acquisition module 20 is used to acquire the early coagulation coefficient of the arteriovenous pot at the current moment according to the blood sticking to the wall in each arterial pot image and the floating blood in each venous pot image in the current time period.

[0071] Specifically, during hemodialysis, when coagulation occurs, it can cause blockage in the hemodialysis tubing, leading to a slight increase in pressure in the arteriovenous chamber. Due to the different locations of the arterial and venous chambers, the arterial and venous chambers exhibit different behaviors during coagulation. When pressure from early coagulation is present in the arterial chamber, the blood level in the chamber will slightly decrease. Furthermore, due to the poor anticoagulability of the blood, blood may cling to the wall during this decrease. When pressure from early coagulation is present in the venous chamber, protein may precipitate from the blood in the chamber, resulting in a phenomenon called "floating" in the chamber. During coagulation, the blood clinging to the wall and the flocculent clinging in the chamber should occur simultaneously; that is, the more severe the blood clinging in the arterial chamber, the more severe the flocculent clinging in the venous chamber. Furthermore, it is known that coagulation is a continuous process that persists without human intervention. Therefore, this embodiment obtains the early coagulation coefficient of the arteriovenous chamber at the current moment based on the blood clinging to the wall and the flocculent clinging in each arterial chamber image during the current time period. The larger the early coagulation coefficient, the more likely early coagulation is present at the current moment. In this embodiment, the current time period is set to 20 seconds. The user can adjust the current time period based on actual circumstances and this is not a limitation. It should be noted that the end time of the current time period is always the current moment, and hemodialysis must have begun for at least 20 seconds.

[0072] Preferably, in one possible implementation of this embodiment, the method for obtaining the early coagulation coefficient is as follows: Figure 2 , which shows a flow chart of a method for obtaining an early coagulation coefficient provided by this embodiment, the method comprising the following steps:

[0073] Step S201: Obtain the blood hanging coefficient of each arterial amputee image in the current time period according to the liquid level, the color of the blood hanging area, and the change trend of the blood hanging height in the arterial amputee image at each moment in the current time period.

[0074] For an arterial blood clot image taken at any moment in the current time period, if the blood level in the arterial blood clot image does not drop, it indicates that there is no blood clot in the arterial blood clot image. If the blood level in the arterial blood clot image does drop, it indicates that there is blood clot in the arterial blood clot image. It is known that as coagulation worsens, the color of the blood clot area becomes increasingly distinct, more consistent with the color of the blood, and the degree of adhesion of the blood clot area increases. The degree of adhesion of the blood clot area can be analyzed by the changing trend of the blood clot height. Specifically, when the clot height changes, it indicates that the blood clot is becoming more severe, indirectly reflecting the severity of the coagulation. Therefore, this embodiment obtains the clot coefficient for each arterial blood clot image in the current time period based on the blood level height, the color of the blood clot area, and the changing trend of the clot height in the arterial blood clot image at each moment in the current time period. The larger the clot coefficient, the more severe the coagulation condition at the corresponding arterial blood clot image.

[0075] In one possible implementation of this embodiment, the wall hanging coefficient is obtained by first segmenting the arterial ampoule image at any moment in the current time period using HSV color space segmentation to obtain the blood region in the arterial ampoule image. To avoid interference from bubbles and blood hanging regions, morphological optimization is then performed to accurately obtain the blood region in the arterial ampoule image as the first blood region. HSV color space segmentation and morphology are both well-known techniques and are not further described. To accurately analyze whether the liquid level in the arterial ampoule image has dropped, this embodiment uses Hough transform detection to obtain a straight line on the arterial ampoule wall in the arterial ampoule image without the catheter as a reference line. All line segments parallel to the reference line are then extracted from the first blood region as reference line segments. The endpoints of the reference line segments are edge pixels on the outermost edge line of the first blood region. Finally, the average length of all reference line segments is used as the liquid level in the arterial ampoule image. Hough transform detection is well-known and is not further described. When the liquid level is equal to the preset liquid level, it indicates that the liquid level in the arterial pot image has not dropped; when the liquid level is less than the preset liquid level, it indicates that the liquid level in the arterial pot image has dropped. The preset liquid level is the liquid level in the arterial pot manually set during the hemodialysis process. The implementer can set it according to actual conditions and is not limited here.

[0076] When the liquid level is equal to the preset liquid level, the wall hanging coefficient of the arterial pot image is set to 0 by default. When the liquid level is less than the preset liquid level, the connected domain within the specified range above the first blood region, whose grayscale value variance is within the preset variance range, is defined as the blood wall hanging region. Because the blood wall hanging region is adjacent to the liquid surface and the color distribution within the blood wall hanging region is consistent, this embodiment sets the specified range to 2px to 30px and the preset variance range to 0 to 0.3. The implementer can set the specified range and preset variance range according to actual conditions and is not limited here. Thus, the blood wall hanging region in the arterial pot image is accurately determined.

[0077] In order to analyze the blood clinging to the arterial blood vessel image, the color conspicuity of the blood clinging area is obtained based on the difference in R channel values between the blood clinging area and the first blood area. The greater the color conspicuity, the more serious the blood clinging to the arterial blood vessel image. The color conspicuity is obtained by first obtaining the average R channel values of the pixels in the blood clinging area as the first value; then obtaining the average R channel values of the pixels in the first blood area as the second value; the more equal the first value and the second value are, the more obvious the color of the blood clinging area is. The ratio of the first value to the second value is then used as the color conspicuity of the blood clinging area.

[0078] To more accurately analyze the blood clinging to the arterial pothole image, the length of the longest line segment parallel to the reference line in the blood clinging region is further used as the blood clinging height. The degree of blood clinging height variation of the arterial pothole image is then determined based on the blood clinging heights of the arterial pothole image and its neighboring arterial pothole images. A greater degree of blood clinging height variation indicates a greater tendency for blood clinging to increase in the arterial pothole image, indirectly reflecting a more severe blood clinging condition. The degree of blood clinging height variation is determined by taking a preset number of arterial pothole images that are temporally closest to the arterial pothole image as its neighboring arterial pothole images. In this embodiment, the preset number is set to 10; however, the number can be adjusted based on actual circumstances and is not limited herein. The blood clinging heights of the arterial pothole image and its neighboring arterial pothole images are arranged in time sequence from the beginning to the end of the corresponding arterial pothole images to obtain a blood clinging height sequence. The slope of the line fitted to the blood clinging heights in the blood clinging height sequence is then determined as the degree of blood clinging height variation of the arterial pothole image. The method of fitting a straight line is a well-known technique and will not be described in detail.

[0079] It is known that greater color visibility and greater variation in the blood hanging on the wall height indicate more severe blood hanging on the wall in the arterial pot image. Therefore, in this embodiment, the product of color visibility and variation in the blood hanging on the wall height is normalized to obtain the blood hanging coefficient of the arterial pot image. The calculation formula for the blood hanging coefficient is: Where, is the wall hanging coefficient of the artery pot image at the ath moment in the current time period; is the first value corresponding to the blood wall area in the artery image at the ath moment in the current time period; is the second value corresponding to the first blood area in the arterial pot image at the ath moment in the current time period; The color of the blood clinging area in the arterial image at the ath moment in the current time period is obvious; is the degree of change in the wall height of the artery pot image at the ath moment in the current time period; norm is the normalization function.

[0080] At this point, the wall hanging coefficient of each artery pot image in the current time period is obtained.

[0081] Step S202: obtaining the floating coefficient of each venous pot image in the current time period according to the color prominence and size of the floating area in the venous pot image at each moment in the current time period.

[0082] For a vein bottle image taken at any moment in the current time period, the more pronounced and larger the fluff area in the vein bottle image, the more severe the coagulation condition at that moment. Furthermore, this embodiment obtains a fluff coefficient for each vein bottle image taken during the current time period based on the color prominence and size of the fluff area in the vein bottle image at each moment in the current time period. The larger the fluff coefficient, the more severe the coagulation condition at that moment in the corresponding vein bottle image.

[0083] In one possible implementation of this embodiment, the method for obtaining the fluff coefficient is as follows: for a vein pot image taken at any moment in the current time period, extract the blood region in the vein pot image as the second blood region using HSV color space segmentation. It is known that fluff is significantly brighter than blood in the blood. The second blood region is then segmented using the Otsu threshold segmentation algorithm, with the high-grayscale region being the fluff region. The Otsu threshold segmentation algorithm is a well-known technique and will not be described in detail here. The region of the second blood region excluding the fluff region is used as the reference blood region. The absolute difference between the mean grayscale value of the pixels in the fluff region and the mean grayscale value of the pixels in the reference blood region is obtained as the color significance of the fluff region. A greater color significance indicates a more pronounced fluff region and a more severe fluff phenomenon in the vein pot image. Furthermore, a larger area of the fluff region indicates a more severe fluff phenomenon in the vein pot image. The method for obtaining the area of the fluff region is a well-known technique and will not be described in detail here. In order to accurately represent the floating catkins in the vein pot image, the product of the color prominence and the area of the floating catkins region is normalized to obtain the floating catkins coefficient of the vein pot image. In this embodiment, the product of the color prominence and the area of the floating catkins region is normalized using the norm normalization function.

[0084] At this point, the floating coefficient of each venous pot image in the current time period is obtained.

[0085] Step S203: obtaining the early coagulation coefficient of the arteriovenous cannula at the current moment according to the correlation between the wall hanging coefficient and the floating coefficient in the current time period, as well as the wall hanging coefficient and the floating coefficient at the current moment.

[0086] Knowing that coagulation conditions are synchronized with the conditions generated by the arteriovenous cannula, this embodiment analyzes the reliability of early coagulation of the arteriovenous cannula at the current moment based on the correlation between the wall cling coefficient and the wadding coefficient during the current time period. The greater the wall cling coefficient and the wadding coefficient at the current moment, the more likely coagulation is present. This embodiment determines the early coagulation coefficient of the arteriovenous cannula at the current moment based on the correlation between the wall cling coefficient and the wadding coefficient during the current time period, as well as the wall cling coefficient and the wadding coefficient at the current moment.

[0087] In one implementation of this embodiment, the early coagulation coefficient is obtained by: arranging the wall hanging coefficients in the current time period according to the time sequence of the corresponding arterial bottle image from front to back, thereby obtaining a wall hanging coefficient sequence; arranging the floating coefficients in the current time period according to the time sequence of the corresponding venous bottle image from front to back, thereby obtaining a floating coefficient sequence; and using the Pearson correlation coefficient between the wall hanging coefficient sequence and the floating coefficient sequence as a first eigenvalue. A larger first eigenvalue indicates a more accurate early coagulation analysis at the current moment. The method for obtaining the Pearson correlation coefficient is well known and will not be described in detail. When both the wall hanging coefficient and the floating coefficient at the current moment are larger, it indicates a higher likelihood of early coagulation at the current moment. Furthermore, in this embodiment, the average of the wall hanging coefficient and the floating coefficient at the current moment is used as the coagulation analysis value at the current moment. To improve the accuracy of the coagulation analysis at the current moment, the coagulation analysis value is corrected using the first eigenvalue. Therefore, the product of the coagulation analysis value and the first eigenvalue is normalized and used as the early coagulation coefficient of the arteriovenous bottle at the current moment. In this embodiment, the product of the coagulation analysis value and the first eigenvalue is normalized by using the norm normalization function.

[0088] The dialysis tubing abnormality degree acquisition module 30 is used to acquire the abnormality degree of the dialysis tubing at the current moment based on the distribution of bubbles in the arterial pot image at the current moment and the changes in the liquid level in the arterial pot image and the venous pot image during the current time period.

[0089] Specifically, in actual situations, when an abnormality in the dialysis tubing occurs, it can also cause abnormal pressure, which in turn affects the amount of blood clinging to the arteriovenous chamber and the amount of blood floating around. Therefore, the early coagulation coefficient obtained by the early coagulation coefficient acquisition module 20 cannot accurately determine whether early coagulation is present at the current moment. It is known that pressure changes caused by abnormalities in the dialysis tubing can cause bubbles to form in the arterial chamber and a sudden drop in the liquid level, while the drop in the arterial chamber caused by early coagulation is slight and steady. On the other hand, pressure changes caused by abnormalities in the dialysis tubing can cause significant fluctuations in the liquid level in the venous chamber. Therefore, this embodiment determines the degree of abnormality in the dialysis tubing at the current moment based on the distribution of bubbles in the arterial chamber image at the current moment and the changes in the liquid level in the arterial and venous chamber images over the current time period. The greater the degree of abnormality in the dialysis tubing, the greater the likelihood of abnormality in the dialysis tubing at the current moment, and the more inaccurate the early coagulation coefficient of the arteriovenous chamber at the current moment, indicating a greater need for correction.

[0090] Preferably, in one possible implementation of this embodiment, the method for obtaining the degree of abnormality in the dialysis tubing is as follows: each connected domain in the first blood region in the current arterial image is treated as a bubble, and the area of each bubble is obtained using the Hough circle detection algorithm, which is used as the first area. The Hough circle detection algorithm is a well-known technique and will not be described in detail. The larger the first areas, the more bubbles there are in the current arterial image, which indirectly indicates a greater likelihood of abnormality in the dialysis tubing at the current moment. The sum of all first areas is then used as the first abnormality value of the dialysis tubing. The larger the first abnormality value of the dialysis tubing, the greater the likelihood of abnormality in the current dialysis tubing.

[0091] In order to accurately analyze the abnormality of the dialysis tube at the current moment, the difference in the liquid level in the arterial cuvette image at each moment in the current time period and its adjacent next moment is obtained and used as the height analysis change value. When there is a significant difference in the height analysis change value, the more likely the dialysis tube abnormality is, the greater the possibility of the dialysis tube abnormality at the current moment. The variance of the height analysis change value is then used as the second abnormality value of the dialysis tube. The larger the second abnormality value of the dialysis tube, the greater the possibility of the dialysis tube abnormality at the current moment.

[0092] Furthermore, for the intravenous pot image at any moment in the current time period, a straight line without the catheter side pot wall in the intravenous pot image is obtained through Hough transform detection as a target straight line, and all line segments parallel to the target straight line in the second blood region of the intravenous pot image are extracted as target line segments, wherein the endpoints of the target line segments are edge pixel points on the outermost edge line of the second blood region; the mean of the lengths of all target line segments is used as the liquid level height in the intravenous pot image; the variance of the liquid level height in all intravenous pot images in the current time period is obtained as the third abnormal value of the dialysis tubing; the larger the third abnormal value of the dialysis tubing, the greater the fluctuation of the liquid level in the intravenous pot in the current time period, and the greater the possibility of abnormality in the dialysis tubing at the current moment;

[0093] To accurately represent the current abnormality of the dialysis tubing and enable subsequent accurate correction of the early coagulation coefficient, this embodiment normalizes the product of the first, second, and third abnormal values of the dialysis tubing as the current abnormality level of the dialysis tubing. This embodiment normalizes the product of the first, second, and third abnormal values of the dialysis tubing using the norm normalization function.

[0094] At this point, the abnormality level of the dialysis tubing at the current moment is obtained.

[0095] The early coagulation monitoring module 40 is used to correct the early coagulation coefficient according to the abnormality of the dialysis tube, obtain the corrected early coagulation coefficient of the arteriovenous cannula at the current moment, and determine whether there is early coagulation phenomenon at the current moment.

[0096] Specifically, it is known that the greater the degree of abnormality in the dialysis tubing, the less accurate the early coagulation coefficient at the corresponding moment. Therefore, this embodiment corrects the early coagulation coefficient at the current moment based on the degree of abnormality in the dialysis tubing, obtaining the corrected early coagulation coefficient for the arteriovenous device at the current moment. A larger corrected early coagulation coefficient indicates a higher likelihood of early coagulation at the current moment. Therefore, this embodiment accurately determines whether early coagulation is present based on the corrected early coagulation coefficient.

[0097] Preferably, in one possible implementation of this embodiment, the method for obtaining the corrected early coagulation coefficient is as follows: the result of negative correlation of the abnormality degree of the dialysis tubing at the current moment is used as the correction weight; the product of the correction weight and the early coagulation coefficient at the current moment is used as the corrected early coagulation coefficient of the arteriovenous cannula at the current moment. The calculation formula for the corrected early coagulation coefficient is: Where, is the corrected early coagulation coefficient of the arteriovenous chamber at the current moment; is the early coagulation coefficient of the arteriovenous chamber at the current moment; The abnormality level of the dialysis tube at the current moment; To correct the weight.

[0098] Preferably, in one possible implementation of this embodiment, the method for determining whether early coagulation occurs at the current moment is as follows: This embodiment sets a preset early coagulation coefficient threshold value at 0.5. The implementer may set the preset early coagulation coefficient threshold value based on actual conditions, and this is not limited here. When the corrected early coagulation coefficient is greater than the preset early coagulation coefficient threshold value, it is determined that early coagulation occurs at the current moment, and an early coagulation alarm is issued at the current moment, reminding medical staff to conduct further manual review of the coagulation status in the arteriovenous bottle, so as to timely and accurately control the coagulation risk and ensure the stable progress of hemodialysis.

[0099] When the corrected early coagulation coefficient is less than or equal to the preset early coagulation coefficient threshold, it is judged that there is no early coagulation phenomenon at the current moment. However, if the early coagulation coefficient is greater than the preset early coagulation coefficient threshold at this time, the medical staff will be reminded to check the dialysis pipeline and accurately analyze whether there is any abnormality in the dialysis pipeline at the current moment to avoid the impact of the dialysis pipeline abnormality on hemodialysis; if the early coagulation coefficient is less than or equal to the preset early coagulation coefficient threshold at this time, it means that the hemodialysis is stable at the current moment.

[0100] In summary, this embodiment obtains the arterial and venous pot images in real time; obtains the early coagulation coefficient of the arteriovenous pot at the current moment based on the blood hanging on the wall in the arterial pot image and the floating blood in the venous pot image in the current time period; obtains the abnormality of the dialysis tubing at the current moment based on the bubble distribution in the arterial pot image at the current moment and the change in the liquid level in the arterial and venous pot images in the current time period; corrects the early coagulation coefficient according to the abnormality of the dialysis tubing, obtains the corrected early coagulation coefficient at the current moment and determines whether there is an early coagulation phenomenon at the current moment. The present invention accurately obtains the corrected early coagulation coefficient in real time, which is conducive to timely and accurate detection of early coagulation phenomena and timely treatment, thereby effectively improving the stability of hemodialysis.

[0101] Example 2:

[0102] The present invention also proposes an early coagulation monitoring device for an arteriovenous cisternae based on image processing. The device includes a memory and a processor. The memory stores executable program code, and the processor is configured to call and execute the executable program code to implement the early coagulation monitoring system for an arteriovenous cisternae based on image processing provided in an embodiment of the present application. The device can be a chip, component, or module. The chip may include a connected processor and memory. The memory is configured to store instructions. When the processor calls and executes the instructions, the chip can execute the early coagulation monitoring system for an arteriovenous cisternae based on image processing provided in the above embodiment.

[0103] In addition, the present application also protects a computer device, see Figure 3 The computer device includes a memory 401, a processor 402, and a computer program 403 stored in the memory 401 and running on the processor 402, wherein when the processor 402 executes the computer program 403, the computer device can execute any one of the above-mentioned image processing-based early coagulation monitoring systems for arteriovenous vessels.

[0104] Example 3:

[0105] The present invention also provides a computer-readable storage medium, which stores computer program code. When the computer program code is run on a computer, the computer executes the above-mentioned related method steps to implement an early coagulation monitoring system for an arteriovenous cannula based on image processing provided in the above embodiment.

[0106] Example 4:

[0107] The present invention also provides a computer program product. When the computer program product is run on a computer, it enables the computer to execute the above-mentioned related steps to implement an early coagulation monitoring system for an arteriovenous chamber based on image processing provided in the above embodiment.

[0108] Among them, the device, computer-readable storage medium, computer program product or chip provided in this embodiment are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0109] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0110] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. An early coagulation monitoring system for arteriovenous blood clots based on image processing, characterized in that: The system includes the following steps: An image acquisition module is used to acquire an arterial pot image and a venous pot image at each moment in real time; An early coagulation coefficient acquisition module is used to obtain the early coagulation coefficient of the arteriovenous pot at the current moment based on the blood sticking to the wall in each arterial pot image and the floating blood in each venous pot image in the current time period; The dialysis tubing abnormality degree acquisition module is used to acquire the abnormality degree of the dialysis tubing at the current moment based on the distribution of bubbles in the arterial pot image at the current moment and the changes in the liquid level in the arterial pot image and the venous pot image during the current time period; The early coagulation monitoring module is used to correct the early coagulation coefficient according to the abnormality of the dialysis tube, obtain the corrected early coagulation coefficient of the arteriovenous pot at the current moment, and determine whether there is early coagulation phenomenon at the current moment.

2. The early coagulation monitoring system for arteriovenous blood clots based on image processing according to claim 1, characterized in that: The method for obtaining the early coagulation coefficient is: Obtain the blood hanging coefficient of each arterial bottle image in the current time period according to the liquid level height, the color of the blood hanging area and the change trend of the blood hanging height in the arterial bottle image at each moment in the current time period; According to the color significance and size of the floating catkins area in the venous pot image at each moment in the current time period, the floating catkins coefficient of each venous pot image in the current time period is obtained; According to the correlation between the wall hanging coefficient and the floating coefficient in the current time period, as well as the wall hanging coefficient and the floating coefficient at the current moment, the early coagulation coefficient of the arteriovenous kettle at the current moment is obtained.

3. The early coagulation monitoring system for arteriovenous blood clots based on image processing according to claim 2, characterized in that: The method for obtaining the wall hanging coefficient is: For the artery pot image at any moment in the current time period, a blood region in the artery pot image is obtained as a first blood region through HSV color space segmentation and morphological optimization; A straight line on the artery wall without the catheter in the artery image is obtained by Hough transform detection as a reference straight line, all line segments parallel to the reference straight line are extracted in the first blood region as reference line segments, and the average length of all reference line segments is used as the liquid level in the artery image; When the liquid level is equal to the preset liquid level, the wall hanging coefficient of the arterial pot image defaults to 0; When the liquid level is lower than the preset liquid level, a connected region with a grayscale value variance within a preset variance range within a specified range above the first blood region is regarded as a blood wall region; Obtaining the color conspicuity of the blood wall area according to the difference in R channel values between the blood wall area and the first blood area; The length of the longest line segment parallel to the reference line in the blood wall area is used as the wall height, and the degree of change of the wall height of the arterial pot image is obtained based on the difference between the wall heights of the arterial pot image and the arterial pot images in its preset neighborhood. The product of the color obviousness and the degree of change in the wall height is normalized to obtain a result as the wall coefficient of the artery pot image.

4. The early coagulation monitoring system for arteriovenous blood clots based on image processing according to claim 3, characterized in that: The method for obtaining the color obviousness is: Obtain the mean R channel value of the pixel points in the blood wall area as the first value; Obtaining an average of the R channel values of the pixels in the first blood area as the second value; The ratio of the first value to the second value is used as the color obviousness of the blood wall area.

5. The early coagulation monitoring system for arteriovenous blood clots based on image processing according to claim 3, characterized in that: The method for obtaining the degree of change in the wall height is: using a preset number of arterial pot images that are closest to the arterial pot image in time sequence as preset neighboring arterial pot images of the arterial pot image; Arrange the wall heights of the artery pot image and its preset neighboring artery pot images according to the time sequence of the corresponding artery pot images to obtain a wall height sequence; The slope of the straight line fitted by the wall height in the wall height sequence is obtained as the degree of change of the wall height of the arterial pot image.

6. The early coagulation monitoring system for arteriovenous blood clots based on image processing according to claim 3, characterized in that: The method for obtaining the floating coefficient is: For the venous pot image at any moment in the current time period, the blood area in the venous pot image is extracted as the second blood area through HSV color space segmentation. The second blood area is segmented using the Otsu threshold segmentation algorithm, and the high grayscale area is regarded as the floating catkin area. The area except the floating area in the second blood area is used as the reference blood area; Obtaining a difference between a mean grayscale value of pixels in the floating catkin area and a mean grayscale value of pixels in the reference blood area as a color significance of the floating catkin area; The product of the color prominence and the area of the floating area is normalized to obtain a floating coefficient of the venous pot image.

7. The early coagulation monitoring system for arteriovenous blood clots based on image processing according to claim 2, characterized in that: The method for obtaining the early coagulation coefficient is: Arrange the wall coefficients in the current time period according to the time sequence of the corresponding arterial pot images to obtain a wall coefficient sequence; Arrange the floating coefficients in the current time period according to the time sequence of the corresponding venous pot images to obtain a floating coefficient sequence; The Pearson correlation coefficient between the wall hanging coefficient series and the floating coefficient series is used as the first eigenvalue; The average of the wall hanging coefficient and the floating coefficient at the current moment is used as the coagulation analysis value at the current moment; The result of normalizing the product of the coagulation analysis value and the first eigenvalue is used as the early coagulation coefficient of the arteriovenous pump at the current moment.

8. The early coagulation monitoring system for arteriovenous blood clots based on image processing according to claim 6, characterized in that: The method for obtaining the abnormality degree of the dialysis tubing is: The connected domains in the first blood region in the arterial pot image at the current moment are all regarded as bubbles, and the area of each bubble is obtained by the Hough circle detection algorithm, and is regarded as the first area; The sum of all first areas is used as the first abnormal value of the dialysis tubing; Obtain the difference between the liquid level in the arterial pot image at each moment in the current time period and the next adjacent moment, and use them as height analysis change values; The variance of the height analysis change value is taken as the second abnormal value of the dialysis pipeline; For the intravenous pot image at any moment in the current time period, a straight line without the catheter side wall in the intravenous pot image is obtained through Hough transform detection as a target straight line. All line segments parallel to the target straight line in the second blood region of the intravenous pot image are extracted as target line segments, and the average length of all target line segments is used as the liquid level in the intravenous pot image; Obtain the variance of the liquid level in all venous bottle images in the current time period as the third abnormal value of the dialysis pipeline; The result of normalizing the product of the first abnormal value of the dialysis tubing, the second abnormal value of the dialysis tubing, and the third abnormal value of the dialysis tubing is used as the abnormality degree of the dialysis tubing at the current moment.

9. The early coagulation monitoring system for arteriovenous blood clots based on image processing according to claim 1, characterized in that: The method for obtaining the modified early coagulation coefficient is: The result of negative correlation of the abnormality degree of the dialysis tubing is used as a correction weight; The product of the correction weight and the early coagulation coefficient is used as the corrected early coagulation coefficient of the arteriovenous pump at the current moment.

10. The early coagulation monitoring system for arteriovenous blood clots based on image processing according to claim 1, characterized in that: The method for determining whether there is early coagulation phenomenon at the current moment is: When the corrected early coagulation coefficient is greater than the preset early coagulation coefficient threshold, it is determined that early coagulation occurs at the current moment; When the corrected early coagulation coefficient is less than or equal to the preset early coagulation coefficient threshold, it is determined that no early coagulation phenomenon exists at the current moment.

Images