Spine health state positioning and monitoring method and system

Abstract

The invention relates to the technical field of image recognition, in particular to a spine health state positioning and monitoring method and system. The method comprises the steps of firstly obtaining an infrared grayscale image of a spine area of a patient; screening out segmented spine videos in abnormal spine states; further analyzing from two angles of gray distribution in a gray abnormal region in each frame of image and difference between the inside and the outside of the gray abnormal region, and screening out an optimal positioning frame image; further obtaining the pain possibility of each pixel point in the gray level abnormal region in the optimal positioning frame image; and determining an abnormal region according to the pain possibility of each pixel point. According to the method, the optimal positioning frame image is screened, and the pixel points of the gray level abnormal area are further screened according to the optimal positioning frame image, so that interference of a normal area of a pain edge is reduced, the abnormal area is accurately positioned, and an accurate analysis basis is provided for judgment and research of related personnel.

Description

Technical Field

[0001] The present invention relates to the technical field of image recognition, and in particular to a spinal health status positioning monitoring method and system. Background Art

[0002] Traditional spinal health detection methods mainly include clinical physical examinations, X-rays, CT, MRI and other imaging examinations. Although these methods can provide detailed spinal structure data, there are certain radiation risks. At the same time, the related equipment is relatively expensive, not portable and cannot achieve real-time monitoring.

[0003] The existing use of infrared technology to detect pain in the monitoring area provides a new direction for spinal health status detection. However, since the onset time and location of pain in the spinal pain area vary from person to person, and the pain area will also affect the infrared imaging effect of the adjacent area, the accuracy of abnormal area analysis directly based on infrared imaging is poor, and the detection effect of abnormal area positioning is insufficient, which affects the judgment and research of relevant personnel. Summary of the invention

[0004] In order to solve the technical problem that the existing infrared monitoring technology cannot accurately locate the abnormal area of ​​the spine, the purpose of the present invention is to provide a spinal health status positioning monitoring method and system, and the technical solution adopted is as follows:

[0005] A method for positioning and monitoring the health status of the spine, the method comprising:

[0006] Obtain an infrared grayscale image of the patient's spine area; filter out segmented spine videos of abnormal spine states based on the grayscale value in each frame of the infrared grayscale image;

[0007] In the segmented spine video, according to the grayscale distribution characteristics inside the grayscale abnormal area in each frame of the image, the internal best positioning possibility of each frame of the image is obtained; the grayscale abnormal area in each frame of the image is derived outward to obtain the edge area; according to the grayscale gradient change characteristics of the edge area in each frame of the image, combined with the grayscale difference between the grayscale abnormal area and the edge area, the external best positioning possibility of each frame of the image is obtained; the internal best positioning possibility and the external best positioning possibility are integrated to screen out the best positioning frame image;

[0008] Taking the optimal positioning frame image as the center, a preset number of adjacent frame images are selected to form an image set to be analyzed; based on the grayscale change characteristics of each pixel point in the grayscale abnormal area of ​​the optimal positioning frame image in the image set to be analyzed, the disease possibility of each pixel point is obtained; based on the disease possibility of each pixel point, the abnormal area is determined.

[0009] Furthermore, the method for obtaining the segmented spine video includes:

[0010] When a pixel whose grayscale value exceeds the preset normal range appears in the infrared grayscale image, the corresponding infrared grayscale image is used as the starting frame; when the grayscale values ​​of all pixels in the infrared grayscale image return to within the preset normal range, the corresponding infrared grayscale image is used as the ending frame; the adjacent starting frame and the ending frame and the infrared grayscale image in between constitute a segmented spinal video of an abnormal spinal state.

[0011] Furthermore, the method for obtaining the internal best positioning possibility includes:

[0012] The connected domain formed by the pixels whose grayscale values ​​in each frame of the image exceed the preset normal range is regarded as the grayscale abnormal area;

[0013] According to the violent grayscale fluctuation characteristics of the pixels in the grayscale abnormal area in each frame of the image, combined with the overall grayscale characteristics of the pixels, the internal grayscale balance parameters of each frame of the image are obtained; the overall grayscale characteristics of the pixels are positively correlated with the internal grayscale balance parameters;

[0014] According to the grayscale change rate of the central pixel point of the grayscale abnormal area in each frame of the image in a preset direction, the internal grayscale divergence speed of each frame of the image is obtained;

[0015] The internal grayscale balance parameter and the internal grayscale divergence speed of each frame of image are integrated to obtain the internal optimal positioning possibility of each frame of image; the internal grayscale balance parameter is positively correlated with the internal optimal positioning possibility; the internal grayscale divergence speed is negatively correlated with the internal optimal positioning possibility.

[0016] Furthermore, the method for obtaining the edge area includes:

[0017] The edge points of the grayscale abnormal area in each frame of the image are used to derive a preset number of pixels in a direction away from the central pixel, and the connected area formed by the derived pixels is used as the edge area.

[0018] Furthermore, the method for obtaining the external best positioning possibility includes:

[0019] The gradient value of each pixel point in the edge area of ​​each frame of image is obtained; the external optimal positioning possibility is obtained according to the fluctuation severity of the gradient value of the pixel point in the edge area of ​​each frame of image, combined with the difference between the overall grayscale characteristics in the edge area and the overall grayscale characteristics in the grayscale abnormal area; the fluctuation severity and the difference in the overall grayscale characteristics corresponding to the edge area are positively correlated with the external optimal positioning possibility.

[0020] Furthermore, the method for acquiring the optimal positioning frame image includes:

[0021] The internal best positioning possibility and the external best positioning possibility are integrated to obtain the comprehensive best positioning possibility for each frame of image; the internal best positioning possibility and the external best positioning possibility are both positively correlated with the comprehensive best positioning possibility;

[0022] The infrared grayscale image corresponding to the maximum value of the comprehensive best positioning possibility in the segmented spine video is used as the best positioning frame image.

[0023] Furthermore, the method for obtaining the possibility of illness includes:

[0024] Mark any pixel point in the grayscale abnormal area in the optimal positioning frame image as a target pixel point; select any frame image in the image set to be analyzed and mark it as a target image;

[0025] According to the grayscale value of the target pixel point in the target image and the grayscale change amplitude of the target pixel point between the target image and the adjacent frame image in the negative direction of the time sequence, the pain coefficient of the target pixel point in the target image is obtained; the grayscale value of the target pixel point in the target image is positively correlated with the pain coefficient; the grayscale change amplitude of the target pixel point between the target image and the adjacent frame image in the negative direction of the time sequence is negatively correlated with the pain coefficient;

[0026] The disease possibility of the target pixel is obtained according to the overall concentrated characteristics of all the disease coefficients corresponding to the target pixel.

[0027] Furthermore, the method for obtaining the abnormal area includes:

[0028] In the optimal positioning frame image, clustering is performed according to the disease possibility of each pixel in the grayscale abnormal area, and the area corresponding to the largest cluster is used as the abnormal area.

[0029] Furthermore, Mean Shift algorithm is used for clustering.

[0030] The present invention also proposes a spinal health status positioning monitoring system, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any step of the spinal health status positioning monitoring method.

[0031] The present invention has the following beneficial effects:

[0032] The present invention first obtains an infrared grayscale image of the patient's spine area to provide a data basis for subsequent analysis; further excludes image information when the spine is in a normal state, screens out segmented spine videos of abnormal spine states, reduces interference from irrelevant images, and reduces the amount of calculation; further in the segmented spine video, analyzes from two angles: the internal grayscale distribution of the grayscale abnormal area in each frame image, and the difference between the inside and outside of the grayscale abnormal area, screens out the best positioning frame image, and obtains the most accurate image that can display the grayscale abnormal area, reduces the interference of environmental factors and the wearer's activities on the detection results, and ensures the accuracy and reliability of the detection results; further, according to each pixel point in the grayscale abnormal area in the best positioning frame image, the grayscale change characteristics in the image set to be analyzed are used to obtain the disease possibility of each pixel point, reduce the interference of normal pixels affected by the diseased area, and provide a basis for subsequent accurate positioning of the abnormal area; finally, according to the disease possibility of each pixel point, the abnormal area is determined. The present invention reduces the interference of normal areas at the edge of the diseased area by screening the best positioning frame image and further screening the pixel points in the grayscale abnormal area according to the best positioning frame image, accurately locates the abnormal area, and provides an accurate analysis basis for the judgment and research of relevant personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. 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 creative work.

[0034] Figure 1 A flowchart of a spinal health status positioning monitoring method provided by one embodiment of the present invention;

[0035] Figure 2 An infrared grayscale image of a spinal region provided by an embodiment of the present invention;

[0036] Figure 3 A flowchart of a method for obtaining the best internal positioning possibility provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a spinal health status positioning monitoring method and system proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0038] 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.

[0039] The specific scheme of the spinal health status positioning monitoring method and system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0040] See also Figure 1 , which shows a flow chart of a spinal health status positioning monitoring method provided by an embodiment of the present invention, specifically comprising:

[0041] Step S1: Obtain an infrared grayscale image of the patient's spine area; and filter out segmented spine videos of abnormal spine states according to the grayscale value in each frame of the infrared grayscale image.

[0042] First, an infrared grayscale image of the patient's spine area is obtained to provide a data basis for subsequent analysis.

[0043] In one embodiment of the present invention, a medical infrared imager is used to collect data of the patient's spine region in real time, and each frame of the captured infrared image is converted into a grayscale image. The existing Gaussian filtering technology is used to perform noise reduction on the grayscale image, which effectively reduces the interference that may occur during data transmission and ensures the stability and reliability of the data. Figure 2 , which shows an infrared grayscale image of a spinal region provided by an embodiment of the present invention.

[0044] In order to accurately locate the specific location of the abnormal area of ​​the spine and evaluate the health status of the spine, it is first necessary to exclude the image information when the spine is in a normal state, filter out the segmented spine video when the spine is in an abnormal state, reduce the interference of irrelevant images, and reduce the amount of calculation.

[0045] Preferably, in one embodiment of the present invention, considering that when the spinal region is in a normal state, the temperature of the spinal region is within the normal range, and the corresponding grayscale value is within the normal grayscale interval, when a pixel whose grayscale value exceeds the preset normal range appears in the infrared grayscale image, the corresponding infrared grayscale image is used as the starting frame; when the grayscale values ​​of all pixels in the infrared grayscale image are restored to within the preset normal range, the corresponding infrared grayscale image is used as the ending frame; the adjacent starting frames and ending frames and the infrared grayscale images therebetween constitute a segmented spinal video of an abnormal spinal state.

[0046] As an example, the main monitoring of pain and inflammation in the spinal region is to set the temperature range corresponding to the grayscale distribution range to [30, 42.75], the temperature difference between adjacent grayscale levels is 0.05 degrees Celsius, and the normal temperature range is set to [36, 37.5], corresponding to the preset normal range of [119, 149], where the temperature of 37.5 degrees Celsius corresponds to a grayscale value of 149. When pain or inflammation occurs, the temperature of the spinal region will rise significantly, causing the grayscale value to exceed the preset normal range. When the grayscale value of a pixel exceeds 149 for the first time in the infrared grayscale image, it is considered abnormal. The acquisition frequency is 15 frames / second, and the infrared grayscale image is continuously recorded until there is no abnormality in the image, forming a segmented spinal video.

[0047] In other embodiments of the present invention, a temperature interval may also be set, and the preset normal range may be adjusted accordingly.

[0048] It should be noted that, considering that a patient may have multiple abnormal areas in the spinal region, and the onset time or duration of different diseased areas are different, the abnormal areas without any repeated pixels between the pixels involved in different abnormal areas during the entire process from the occurrence of abnormality to recovery to normal are regarded as different abnormal areas. The start and end time of each segmented spinal video corresponds to the appearance of an abnormal area to its disappearance after recovery. Therefore, it can be regarded as a segmented spinal video containing only one abnormal area, and this abnormal area is analyzed and finally accurately located and monitored.

[0049] For example, area A starts to be abnormal at 13:20:15 and ends at 13:30:20; area B starts to be abnormal at 13:25:30 and ends at 13:28:10. During the entire abnormal process of area A and area B, there are no repeated pixels between area A and area B. Area A and area B are considered to be independent of each other and need to be located and monitored separately.

[0050] In other embodiments of the present invention, it may also be stipulated that the minimum interval between edge pixels of two abnormal regions exceeds a certain distance, such as 20 pixels, and the two abnormal regions are considered to be independent of each other and different abnormal regions.

[0051] Step S2: In the segmented spine video, according to the grayscale distribution characteristics inside the grayscale abnormal area in each frame image, the internal best positioning possibility of each frame image is obtained; the grayscale abnormal area in each frame image is derived outward to obtain the edge area; according to the grayscale gradient change characteristics of the edge area in each frame image, combined with the grayscale difference between the grayscale abnormal area and the edge area, the external best positioning possibility of each frame image is obtained; the internal best positioning possibility and the external best positioning possibility are integrated to screen out the best positioning frame image.

[0052] In the embodiment of the present invention, considering that environmental factors and the activities of the wearer may interfere with the detection results, and the abnormal area may have a radiation effect on the surrounding area, causing the pixels in the normal area to be abnormal, thereby affecting the accurate positioning of the abnormal state of the spine, it is necessary to screen out the best positioning frame image from the continuous segmented spine video, so as to most accurately display the grayscale abnormal area to ensure the accuracy and reliability of the detection results.

[0053] Considering that when pain occurs in the spinal area, the corresponding area will experience a gradual increase in temperature, reach a certain stable level and maintain it for a period of time. When the pain gradually eases, the temperature will drop and return to normal. The grayscale distribution characteristics of the corresponding grayscale abnormal area will also change. Therefore, the internal optimal positioning possibility of each frame of image is obtained based on the grayscale distribution characteristics inside the grayscale abnormal area in each frame of image.

[0054] Preferably, in one embodiment of the present invention, the method for obtaining the internal best positioning possibility includes:

[0055] See also Figure 3 , which shows a flow chart of a method for obtaining the internal best positioning possibility provided by an embodiment of the present invention, specifically comprising:

[0056] Step S301: taking a connected domain formed by pixels whose grayscale values ​​in each frame of image exceed a preset normal range as a grayscale abnormal region.

[0057] Considering that the image area involved in the abnormal area may be constantly changing, the grayscale abnormal area is obtained for each frame of the image.

[0058] Step S302: according to the grayscale fluctuation characteristics of the pixels in the grayscale abnormal area in each frame of the image, combined with the overall grayscale characteristics of the pixels, the internal grayscale balance parameters of each frame of the image are obtained; the overall grayscale characteristics of the pixels are positively correlated with the internal grayscale balance parameters.

[0059] Considering that when the abnormal area appears completely and has a greater impact on the surrounding areas, due to the attenuation of thermal radiation, the grayscale fluctuation of the area affected by the radiation is large, which makes the grayscale fluctuation of the entire abnormal area more intense; when the abnormal area gradually appears, although it has little impact on the surrounding areas, the overall grayscale value is also low, which is not conducive to observing the complete abnormal area; therefore, when the abnormal area appears completely and has little impact on the surrounding areas, the overall grayscale distribution of the abnormal area is more balanced, and the overall grayscale value is higher, which is most beneficial for observing the abnormal area, so as to obtain the internal grayscale balance parameters.

[0060] As an example, the variance of the grayscale values ​​of pixels in the grayscale abnormal area is obtained, and the characteristics of drastic grayscale fluctuations are characterized by the variance of the grayscale values; the average value of the grayscale values ​​of pixels in the grayscale abnormal area is obtained, and the overall grayscale characteristics are characterized by the average value of the grayscale values; the ratio of the average value to the variance corresponding to each frame of the image is used as the internal grayscale balancing parameter.

[0061] It should be noted that, in the embodiments of the present invention, in order to prevent the denominator from being zero, a positive parameter for dividing by zero, such as 0.1, is provided.

[0062] In other embodiments of the present invention, the implementer may also obtain fluctuation parameters such as the standard deviation and range of the grayscale value to indicate the drastic grayscale fluctuation characteristics of the pixel points in the grayscale abnormal area; the mode and median of the grayscale value may also be obtained, and weighted together with the mean to characterize the overall grayscale characteristics; the grayscale values ​​may also be weighted and summed according to the proportion of the pixel points corresponding to each grayscale value occupying the grayscale abnormal area to characterize the overall grayscale characteristics, which will not be repeated here.

[0063] Step S303: Obtaining the internal grayscale divergence speed of each frame of image according to the grayscale change rate of the central pixel point of the grayscale abnormal area in each frame of image toward a preset direction.

[0064] Taking into account the grayscale change characteristics from the center to the edge of the grayscale abnormal area, the diffusion characteristics of the abnormal area can be reflected. At the same time, in order to limit the analysis scope, a preset direction is set to reduce the amount of calculation and speed up the analysis rate.

[0065] As an example, the preset direction includes four directions, left and right in the horizontal direction, and up and down in the vertical direction, wherein the horizontal 0 degree is selected as the left, the central pixel point of the grayscale abnormal area is the starting point, and the edge pixel points pointing to the preset direction present a cross shape; the grayscale sequence corresponding to any direction is used to obtain a change curve of the grayscale value of the pixel point in the grayscale sequence, the horizontal axis is the sequence number of the pixel point in the direction from the center pixel point to the edge pixel point, and the vertical axis is the grayscale value of the pixel point, and the slope of each pixel point is obtained;

[0066] The slopes of all pixels in the grayscale sequence corresponding to the four directions are averaged. Considering that the closer the grayscale abnormal area is to the edge, the lower the temperature and the smaller the grayscale value, the absolute value of the slope average is also needed. The absolute value of the slope average represents the grayscale change rate, which is used as the internal grayscale divergence speed.

[0067] The larger the absolute value of the slope average, the greater the grayscale change rate, the more obvious the radiation divergence characteristics of the abnormal area, and the greater the internal grayscale divergence speed.

[0068] Step S304: The internal grayscale balance parameter and the internal grayscale divergence speed of each frame image are integrated to obtain the internal optimal positioning possibility of each frame image; the internal grayscale balance parameter is positively correlated with the internal optimal positioning possibility; the internal grayscale divergence speed is negatively correlated with the internal optimal positioning possibility.

[0069] After analyzing the internal grayscale balance characteristics and grayscale divergence characteristics of each frame of image, the internal grayscale balance parameters and internal grayscale divergence speed of each frame of image can be integrated to obtain the internal optimal positioning possibility of each frame of image; considering that the larger the internal grayscale balance parameter, the more balanced the grayscale distribution inside the grayscale abnormal area of ​​each frame of image, the smaller the impact on the surrounding area of ​​the disease, and the more conducive to locating the abnormal area; considering that the smaller the internal grayscale divergence speed, the less obvious the grayscale radiation divergence characteristics inside the grayscale abnormal area, the more likely the abnormal area is to be fully presented, and the more conducive to locating the abnormal area, so the internal grayscale balance parameter is positively correlated with the internal optimal positioning possibility; the internal grayscale divergence speed is negatively correlated with the internal optimal positioning possibility.

[0070] As an example, the ratio of the internal grayscale balance parameter to the internal grayscale divergence speed is taken as the internal optimal positioning possibility.

[0071] In other embodiments of the present invention, the implementer may also use other positive correlation mapping functions such as exp(x), or other negative correlation mapping functions such as exp(-x) for correlation mapping, where exp() represents an exponential function with the natural constant e as the base; x represents the independent variable.

[0072] Considering that when pain occurs in the spinal area, the surrounding tissues of the abnormal area will be affected by the temperature radiation of the abnormal area, causing the edges of the abnormal area in the acquired image to become blurred, it is also necessary to analyze the impact of the grayscale abnormal area on the edge area. Here, it is first necessary to limit the edge area of ​​the grayscale abnormal area, and derive the grayscale abnormal area in each frame of the image outward to obtain the edge area.

[0073] Preferably, in one embodiment of the present invention, a preset number of pixels are derived from the edge points of the grayscale abnormal area in each frame of the image in a direction away from the central pixel, and the connected area formed by the derived pixels is used as the edge area.

[0074] It should be noted that the derivation process extends outward from the edge of the grayscale abnormal area.

[0075] As an example, the preset number is 20. Analogously to region growing, the current grayscale abnormal region is grown outward by 20 circles of pixels, and the grown region is used as the edge region.

[0076] Taking into account that the edge area belongs to the normal area, but is affected by the temperature radiation of the abnormal area, the grayscale gradient will change dramatically. At the same time, the temperature of the abnormal area is different from that of the normal area. Therefore, according to the grayscale gradient change characteristics of the edge area in each frame image, combined with the grayscale difference between the grayscale abnormal area and the edge area, the external best positioning possibility of each frame image is obtained, which is supplemented by the internal analysis angle, so as to finally screen out the best positioning frame image and accurately locate the abnormal area.

[0077] Preferably, in one embodiment of the present invention, considering that the more drastic the fluctuation of the gradient value of the pixel points in the edge area is, the stronger the temperature radiation ability of the abnormal area is, the more obvious the abnormal area is, and at the same time, the edge area is not included in the grayscale abnormal area, which means that when the abnormal area is presented, too many pixels with normal edges are not included in the grayscale abnormal area, which is more conducive to accurately locating the abnormal area; considering that the greater the difference between the overall grayscale characteristics in the edge area and the overall grayscale characteristics of the grayscale abnormal area is, the greater the difference between the normal area and the abnormal area is, the more obvious the boundary is, and the more conducive to locating the abnormal area;

[0078] Based on this, the gradient value of each pixel point in the edge area of ​​each frame image is obtained; according to the fluctuation severity of the gradient value of the pixel point in the edge area of ​​each frame image, combined with the difference between the overall grayscale characteristics in the edge area and the overall grayscale characteristics in the grayscale abnormal area, the external optimal positioning possibility is obtained; the fluctuation severity and the difference in the overall grayscale characteristics corresponding to the edge area are positively correlated with the external optimal positioning possibility.

[0079] As an example, the absolute value of the difference between the gradient value of each pixel in the edge area and the average gradient value is obtained, and the sum of the absolute values ​​of the differences corresponding to all pixels in the edge area is taken as the external first possibility to characterize the severity of the fluctuation of the gradient value, and the severity of the fluctuation of the gradient value is measured by the deviation characteristics between the gradient value and the average gradient value; the overall grayscale characteristics of the area are characterized by the average grayscale value, and the difference between the average grayscale value of the grayscale abnormal area and the average grayscale value of the edge area is taken as the external second possibility to measure the difference between the overall grayscale characteristics of the edge area and the overall grayscale characteristics of the grayscale abnormal area; the product of the external first possibility and the external second possibility corresponding to each frame of the image is taken as the external optimal positioning possibility.

[0080] As another example, the implementer may also map the external first possibility or the external second possibility through a positive correlation mapping function such as exp(x), for example, mapping the external first possibility through the exp(x) function to amplify the sensitivity of the external first possibility.

[0081] The internal best positioning possibility reflects the temperature conditions inside the abnormal area; the external best positioning possibility reflects the contrast characteristics between the abnormal area and the surrounding normal area; the internal and external best positioning possibilities respectively reflect the different characteristics of the abnormal area. In order to comprehensively evaluate the abnormal area, it is necessary to fuse the internal best positioning possibility and the external best positioning possibility to screen out the best positioning frame image.

[0082] Preferably, in one embodiment of the present invention, the internal best positioning possibility and the external best positioning possibility are integrated to obtain the comprehensive best positioning possibility of each frame of image; the internal best positioning possibility and the external best positioning possibility are both positively correlated with the comprehensive best positioning possibility;

[0083] The infrared grayscale image corresponding to the maximum value of the comprehensive best positioning possibility in the segmented spine video is used as the best positioning frame image.

[0084] As an example, the sum of the internal best positioning possibility and the external best positioning possibility is taken as the comprehensive best positioning possibility of each frame image.

[0085] As another example, the implementer may also adjust the weights of the internal best positioning possibility and the external best positioning possibility, such as 0.6 and 0.4, and perform a weighted sum; other positive correlation methods such as multiplication may also be used to obtain the comprehensive best positioning possibility.

[0086] Step S3: With the best positioning frame image as the center, select a preset number of adjacent frame images to form an image set to be analyzed; according to the grayscale change characteristics of each pixel point in the grayscale abnormal area in the best positioning frame image in the image set to be analyzed, obtain the disease possibility of each pixel point; according to the disease possibility of each pixel point, determine the abnormal area.

[0087] Through step S2, the monitoring image is analyzed and the best positioning frame image is obtained from the segmented spine video. It is believed that the best positioning frame image is least affected by interference factors and is the clearest, so the pain area is accurately located according to the best positioning frame image; considering that the grayscale abnormal area in the best positioning frame image may contain some normal areas affected by the temperature radiation of the abnormal area, further screening is required; considering that the radiation area affected by the abnormal area has no heat source of its own, the grayscale change of the radiation area is more unstable than that of the abnormal area, so with the best positioning frame image as the center, a preset number of adjacent frame images are selected to form an image set to be analyzed; according to the grayscale change characteristics of each pixel point in the grayscale abnormal area in the best positioning frame image in the image set to be analyzed, the possibility of pain of each pixel point is obtained.

[0088] As an example, the preset number is 30 frames. With the best positioning frame image as the center, a preset number of adjacent frame images are selected in the positive and negative directions of the time sequence, and the best positioning frame image and the selected images constitute an image set to be analyzed.

[0089] In other embodiments of the present invention, implementers may also set other preset numbers, which will not be described in detail here.

[0090] Preferably, in one embodiment of the present invention, firstly, any pixel point in the grayscale abnormal area in the best positioning frame image is marked as a target pixel point; any frame image in the image set to be analyzed is selected and marked as the target image;

[0091] Considering that the abnormal area is the heat radiation source of the radiation area, the grayscale value of the pixel in the abnormal area is higher; and the abnormal area has its own tissue heat source. When the radiation area is disturbed by the abnormal area, it may be affected by multiple factors such as patient activities and environmental factors. The temperature stability of the radiation area is poor and the grayscale changes more drastically;

[0092] Based on this, the pain coefficient of the target pixel in the target image is obtained according to the gray value of the target pixel in the target image and the gray value change amplitude of the target pixel in the target image and the adjacent frame image in the negative direction of the time sequence; the gray value of the target pixel in the target image is positively correlated with the pain coefficient; the gray value change amplitude of the target pixel in the target image and the adjacent frame image in the negative direction of the time sequence is negatively correlated with the pain coefficient;

[0093] According to the overall concentrated characteristics of all pain coefficients corresponding to the target pixel, the pain possibility of the target pixel is obtained.

[0094] As an example, the absolute value of the difference between the grayscale value of the target pixel in the target image and the grayscale value of the adjacent frame image in the negative direction of the time sequence is used to represent the grayscale change amplitude of the target pixel in the target image; the ratio of the grayscale value corresponding to the target pixel in the target image to the grayscale change amplitude is used as the pain coefficient of the target pixel in the target image; the average value of the pain coefficients corresponding to the target pixel in all images in the image set to be analyzed is used as the pain possibility of the target pixel.

[0095] In other embodiments of the present invention, the implementer may also use other positive correlation mapping functions such as exp(x), or other negative correlation mapping functions such as exp(-x) for correlation mapping, which will not be repeated here; wherein exp() represents an exponential function with the natural constant e as the base; x represents the independent variable.

[0096] After analyzing the disease possibility of each pixel point in the grayscale abnormal area in the best positioning frame image, the pixel points in the abnormal area can be determined, thereby determining the abnormal area and accurately locating the abnormal area.

[0097] Preferably, in one embodiment of the present invention, in the best positioning frame image, clustering is performed according to the disease possibility of each pixel in the grayscale abnormal area, and the area corresponding to the largest cluster is taken as the abnormal area.

[0098] As an example, the Mean Shift algorithm is used to cluster the disease possibility of each pixel in the grayscale abnormal area, and the area corresponding to the largest cluster is taken as the abnormal area.

[0099] In other embodiments of the present invention, the implementer may also adopt other clustering algorithms such as the DBSCAN algorithm; the disease possibility of the pixel points may also be normalized, and the pixels in the abnormal area may be screened out by setting a threshold, and the connected domain formed by the pixels in the abnormal area may be used as the abnormal area, for example, the threshold is set to 0.55; wherein the DBSCAN algorithm and the MeanShift algorithm are both existing technologies and will not be described in detail here.

[0100] In one embodiment of the present invention, after the abnormal area is determined, the onset time and duration of the abnormal area are also counted; the data of spinal health status positioning monitoring can also be transmitted to a related server and stored to provide analysis basis for relevant personnel.

[0101] An embodiment of the present invention also provides a spinal health status positioning monitoring system, which includes a memory, a processor and a computer program, wherein the memory is used to store the corresponding computer program, and the processor is used to run the corresponding computer program. When the computer program runs in the processor, it can implement a spinal health status positioning monitoring method described in steps S1-S3.

[0102] In summary, in response to the technical problem that the existing infrared monitoring technology cannot accurately locate the abnormal area of ​​the spine, the present invention proposes a method and system for positioning and monitoring the health status of the spine. The present invention first obtains an infrared grayscale image of the patient's spine area; screens out segmented spine videos of abnormal spine states; further analyzes from two angles: the grayscale distribution inside the grayscale abnormal area in each frame image, and the difference between the inside and outside of the grayscale abnormal area, to screen out the best positioning frame image; further, based on the grayscale change characteristics of each pixel point in the grayscale abnormal area in the best positioning frame image in the time-series adjacent image of the best positioning frame image, the disease possibility of each pixel point is obtained; based on the disease possibility of each pixel point, the abnormal area is determined. The present invention reduces the interference of the normal area at the edge of the disease by screening the best positioning frame image, and further screens the pixel points of the grayscale abnormal area according to the best positioning frame image, accurately locates the abnormal area, and provides an accurate analysis basis for the judgment and research of relevant personnel.

[0103] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0104] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for monitoring spinal health status, characterized in that: The method comprises: Obtain an infrared grayscale image of the patient's spine area; filter out segmented spine videos of abnormal spine states based on the grayscale value in each frame of the infrared grayscale image; In the segmented spine video, according to the grayscale distribution characteristics inside the grayscale abnormal area in each frame of the image, the internal best positioning possibility of each frame of the image is obtained; the grayscale abnormal area in each frame of the image is derived outward to obtain the edge area; according to the grayscale gradient change characteristics of the edge area in each frame of the image, combined with the grayscale difference between the grayscale abnormal area and the edge area, the external best positioning possibility of each frame of the image is obtained; the internal best positioning possibility and the external best positioning possibility are integrated to screen out the best positioning frame image; Taking the optimal positioning frame image as the center, a preset number of adjacent frame images are selected to form an image set to be analyzed; based on the grayscale change characteristics of each pixel point in the grayscale abnormal area of ​​the optimal positioning frame image in the image set to be analyzed, the disease possibility of each pixel point is obtained; based on the disease possibility of each pixel point, the abnormal area is determined.

2. A spinal health status positioning monitoring method according to claim 1, characterized in that: The method for obtaining the segmented spine video comprises: When a pixel whose grayscale value exceeds the preset normal range appears in the infrared grayscale image, the corresponding infrared grayscale image is used as the starting frame; when the grayscale values ​​of all pixels in the infrared grayscale image return to within the preset normal range, the corresponding infrared grayscale image is used as the ending frame; the adjacent starting frame and the ending frame and the infrared grayscale image in between constitute a segmented spinal video of an abnormal spinal state.

3. A spinal health status positioning monitoring method according to claim 1, characterized in that: The method for obtaining the internal best positioning possibility includes: The connected domain formed by the pixels whose grayscale values ​​in each frame of the image exceed the preset normal range is regarded as the grayscale abnormal area; According to the violent grayscale fluctuation characteristics of the pixels in the grayscale abnormal area in each frame of the image, combined with the overall grayscale characteristics of the pixels, the internal grayscale balance parameters of each frame of the image are obtained; the overall grayscale characteristics of the pixels are positively correlated with the internal grayscale balance parameters; According to the grayscale change rate of the central pixel point of the grayscale abnormal area in each frame of the image in a preset direction, the internal grayscale divergence speed of each frame of the image is obtained; The internal grayscale balance parameter and the internal grayscale divergence speed of each frame of image are integrated to obtain the internal optimal positioning possibility of each frame of image; the internal grayscale balance parameter is positively correlated with the internal optimal positioning possibility; the internal grayscale divergence speed is negatively correlated with the internal optimal positioning possibility.

4. A spinal health status positioning monitoring method according to claim 3, characterized in that: The method for obtaining the edge area includes: The edge points of the grayscale abnormal area in each frame of the image are used to derive a preset number of pixels in a direction away from the central pixel, and the connected area formed by the derived pixels is used as the edge area.

5. A spinal health status positioning monitoring method according to claim 4, characterized in that: The method for obtaining the external best positioning possibility includes: The gradient value of each pixel point in the edge area of ​​each frame of image is obtained; the external optimal positioning possibility is obtained according to the fluctuation severity of the gradient value of the pixel point in the edge area of ​​each frame of image, combined with the difference between the overall grayscale characteristics in the edge area and the overall grayscale characteristics in the grayscale abnormal area; the fluctuation severity and the difference in the overall grayscale characteristics corresponding to the edge area are positively correlated with the external optimal positioning possibility.

6. A spinal health status positioning monitoring method according to claim 1, characterized in that: The method for acquiring the optimal positioning frame image includes: The internal best positioning possibility and the external best positioning possibility are integrated to obtain the comprehensive best positioning possibility for each frame of image; the internal best positioning possibility and the external best positioning possibility are both positively correlated with the comprehensive best positioning possibility; The infrared grayscale image corresponding to the maximum value of the comprehensive best positioning possibility in the segmented spine video is used as the best positioning frame image.

7. A spinal health status positioning monitoring method according to claim 3, characterized in that: The method for obtaining the possibility of the disease includes: Mark any pixel point in the grayscale abnormal area in the optimal positioning frame image as a target pixel point; select any frame image in the image set to be analyzed and mark it as a target image; According to the grayscale value of the target pixel point in the target image and the grayscale change amplitude of the target pixel point between the target image and the adjacent frame image in the negative direction of the time sequence, the pain coefficient of the target pixel point in the target image is obtained; the grayscale value of the target pixel point in the target image is positively correlated with the pain coefficient; the grayscale change amplitude of the target pixel point between the target image and the adjacent frame image in the negative direction of the time sequence is negatively correlated with the pain coefficient; The disease possibility of the target pixel is obtained according to the overall concentrated characteristics of all the disease coefficients corresponding to the target pixel.

8. A spinal health status positioning monitoring method according to claim 7, characterized in that: The method for obtaining the abnormal area includes: In the optimal positioning frame image, clustering is performed according to the disease possibility of each pixel in the grayscale abnormal area, and the area corresponding to the largest cluster is used as the abnormal area.

9. A spinal health status positioning monitoring method according to claim 8, characterized in that: The MeanShift algorithm is used for clustering.

10. A spinal health status positioning monitoring system, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the spinal health status positioning monitoring method as claimed in any one of claims 1 to 9 are implemented.

Images