Benign prostatic hyperplasia drug curative effect evaluation method based on ultrasonic image

Abstract

The invention relates to the technical field of edge detection, in particular to a prostatic hyperplasia drug curative effect evaluation method based on an ultrasonic image. The method comprises the following steps: acquiring a prostate ultrasonic image of a prostate patient in each examination; obtaining an abnormal hyperplasia area according to the distribution of edge angular points in the prostate ultrasound image; matching the abnormal hyperplasia areas in the prostate ultrasound images of two adjacent examinations to obtain matching areas of the abnormal hyperplasia areas; according to the distribution difference of pixel values, edge angular points and blood vessel forms of the abnormal hyperplasia region and the matching region thereof, obtaining a hyperplasia detail retention degree to correct a Gaussian filtering standard deviation in Canny edge detection of the prostate ultrasound image, obtaining a corrected Gaussian filtering standard deviation, and determining a real edge of the abnormal hyperplasia region; and performing drug curative effect evaluation based on the real edge of the abnormal hyperplasia region. The accuracy of hyperplasia area positioning can be improved, and the curative effect of the prostatic hyperplasia medicine can be accurately evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of edge detection, and particularly to a method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images. Background Art

[0002] Benign Prostatic Hyperplasia (BPH) is a common disease among middle-aged and elderly men. It will not heal on its own and will not disappear with drug treatment. It is mainly manifested as the hyperplasia of prostate tissue leading to urinary tract obstruction, causing symptoms such as frequent urination, urgency of urination, and difficulty in urination. With the aggravation of population aging, the incidence rate of BPH is increasing year by year. Early diagnosis and accurate evaluation of the efficacy of drugs for benign prostatic hyperplasia are crucial for improving the quality of life of patients.

[0003] In the existing methods, the evaluation of the efficacy of drugs for benign prostatic hyperplasia mainly depends on the changes in the hyperplastic area in the prostate ultrasonic images of patients during the process of taking drugs for benign prostatic hyperplasia, so as to evaluate the efficacy of drugs for benign prostatic hyperplasia. When the hyperplastic area significantly decreases, it indicates that the efficacy of drugs for benign prostatic hyperplasia is effective. When the hyperplastic area increases or the change is not obvious, it indicates that the efficacy of drugs for benign prostatic hyperplasia is not good; thus, it is beneficial to timely adjust the treatment drugs for patients with benign prostatic hyperplasia and improve the treatment effect of benign prostatic hyperplasia. However, in the existing methods, there are interferences of low contrast and resolution anisotropy in the prostate ultrasonic images, resulting in that during the Gaussian filtering process of the Canny edge detection algorithm, the edges of the tiny protrusions or depressions in the hyperplastic area of the prostate ultrasonic images are easily smoothed out, and then the positioning of the hyperplastic area is inaccurate, making it impossible to accurately analyze the changes in the hyperplastic area and affecting the accurate evaluation of the efficacy of drugs for benign prostatic hyperplasia. Summary of the Invention

[0004] In order to solve the technical problem that there are interferences of low contrast and resolution anisotropy in the prostate ultrasonic images, resulting in that during the Gaussian filtering process of the Canny edge detection algorithm, the edges of the tiny protrusions or depressions in the hyperplastic area of the prostate ultrasonic images are easily smoothed out, and then the positioning of the hyperplastic area is inaccurate, the purpose of the present invention is to provide a method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images, and the specific technical solution adopted is as follows: The embodiment of the present invention provides a method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images, and the method includes the following steps: Obtain each prostate ultrasonic image of each prostate patient during each examination; Obtain the abnormal hyperplastic area in each prostate ultrasonic image according to the distribution of edge corner points in each prostate ultrasonic image; Match the abnormal hyperplasia regions in the prostate ultrasound images of each examination with those in the adjacent previous examination to obtain the matching regions of each abnormal hyperplasia region; obtain the degree of retention of hyperplasia details in each prostate ultrasound image based on the distribution differences of pixel values and edge corner points between each abnormal hyperplasia region and its matching region in each prostate ultrasound image, as well as the distribution differences of blood vessel morphology. Based on the degree of retention of hyperplasia details, correct the standard deviation of Gaussian filtering in the Canny edge detection process for each prostate ultrasound image to obtain the corrected standard deviation of Gaussian filtering in the Canny edge detection process for each prostate ultrasound image, and determine the true edges of the abnormal hyperplasia regions in the prostate ultrasound images of each examination. Evaluate the efficacy of prostate hyperplasia drugs based on the true edges of the abnormal hyperplasia regions.

[0005] Furthermore, the method for obtaining the abnormal hyperplasia regions is as follows: For any prostate ultrasound image and any edge corner point in this prostate ultrasound image, take the preset number of other edge corner points closest to this edge corner point in this prostate ultrasound image as the neighborhood corner points of this edge corner point. Obtain the variance of the Euclidean distances between this edge corner point and each of its neighborhood corner points as the distribution characteristic value of this edge corner point. For any two edge corner points in this prostate ultrasound image, take the sum of the difference between the distribution characteristic values of these two edge corner points and the Euclidean distance between these two edge corner points as the distance metric value of these two edge corner points. According to the distance metric values of any two edge corner points in this prostate ultrasound image, cluster the edge corner points in this prostate ultrasound image through the K-means clustering algorithm to obtain the corner point clustering clusters in this prostate ultrasound image. Obtain the degree of abnormality of each corner point clustering cluster according to the distribution of edge corner points in each corner point clustering cluster. When the degree of abnormality is greater than the preset abnormality degree threshold, take the corresponding corner point clustering cluster as an abnormal cluster. For any abnormal cluster, take the complete region formed by connecting the edge corner points in this abnormal cluster through the region growing algorithm as an abnormal hyperplasia region in this prostate ultrasound image.

[0006] Furthermore, the method for obtaining the degree of abnormality is as follows: For any corner point clustering cluster, obtain the Euclidean distance between any two edge corner points in this corner point clustering cluster as the first distance. Take the normalized result of the product of the variance of all the first distances and the reciprocal of the mean of all the first distances as the degree of abnormality of this corner point clustering cluster.

[0007] Further, the method for obtaining the matching region is as follows: Arrange the prostate ultrasound images obtained each time in the order of acquisition to form an image sequence for each examination; wherein, the process of obtaining prostate ultrasound images each time is exactly the same; For any prostate ultrasound image in the image sequence of any examination, use the prostate ultrasound image at the same position in the image sequence of the previous adjacent examination of this examination as the matching image of this prostate ultrasound image; For any abnormal hyperplasia region in this prostate ultrasound image, use the DeepSORT (DeepLearning Simple Online and Realtime Tracking, deep sorting) object tracking algorithm to match this abnormal hyperplasia region with the abnormal hyperplasia region in the matching image, and obtain the matching region of this abnormal hyperplasia region.

[0008] Further, the method for obtaining the degree of retention of hyperplasia details is as follows: According to the distribution differences of pixel values and edge corner points in each abnormal hyperplasia region and its matching region, obtain the hyperplasia quantization factor for each abnormal hyperplasia region; According to the distribution differences of blood vessel morphologies in each abnormal hyperplasia region and its matching region, obtain the blood flow dynamic change value for each abnormal hyperplasia region; According to the correlation between the hyperplasia quantization factor and the blood flow dynamic change value in each prostate ultrasound image, obtain the degree of retention of hyperplasia details for each prostate ultrasound image.

[0009] Further, the method for obtaining the hyperplasia quantization factor is as follows: For any abnormal hyperplasia region, obtain the average value of the gray values of all pixel points in this abnormal hyperplasia region as the first gray value; Obtain the average value of the gray values of all pixel points in the matching region of this abnormal hyperplasia region as the second gray value; Take the difference between the first gray value and the second gray value as the first hyperplasia speed analysis value; Take the difference between the number of edge corner points in this abnormal hyperplasia region and its matching region as the second hyperplasia speed analysis value; Take the sum result of the first hyperplasia speed analysis value and the second hyperplasia speed analysis value as the hyperplasia quantization factor for this abnormal hyperplasia region.

[0010] Further, the method for obtaining the blood flow dynamic change value is as follows: For any abnormal hyperplasia region, divide the color Doppler ultrasound image of the abnormal hyperplasia region into multiple local regions; For any local region, fit a straight line based on the positions of all edge pixel points in the local region by the least squares method as the reference straight line of the local region; Obtain the mean of the goodness of fit between each edge line in the local region and the reference straight line as the first index of the local region; Obtain the ratio of the number of edge pixel points in the local region to the number of all pixel points as the second index of the local region; Take the product of the normalized and negatively correlated result of the first index and the second index as the vascular analysis value of the local region; Take the mean of the vascular analysis values of all local regions as the blood flow dynamic characteristic value of the abnormal hyperplasia region; Take the difference between the blood flow dynamic characteristic value of the abnormal hyperplasia region and its matching region as the blood flow dynamic change value of the abnormal hyperplasia region.

[0011] Further, the method for obtaining the retention degree of hyperplasia details is as follows: For any prostate ultrasound image, sort the hyperplasia quantization factors and blood flow dynamic change values of each abnormal hyperplasia region in the prostate ultrasound image in the same order of abnormal hyperplasia regions, and respectively obtain the hyperplasia quantization factor sequence and the blood flow dynamic change value sequence; Take the normalized result of the Pearson correlation coefficient of the hyperplasia quantization factor sequence and the blood flow dynamic change value sequence as the retention degree of hyperplasia details of the prostate ultrasound image.

[0012] Further, the method for obtaining the corrected Gaussian filter standard deviation is as follows: For any prostate ultrasound image, take the product of the retention degree of hyperplasia details of the prostate ultrasound image and the Gaussian filter standard deviation in the Canny edge detection process of the prostate ultrasound image as the corrected Gaussian filter standard deviation in the Canny edge detection process of the prostate ultrasound image.

[0013] Further, the method for obtaining the edge corner points is as follows: Obtain the edge corner points in each prostate ultrasound image through the Harris corner detection algorithm.

[0014] The present invention has the following beneficial effects: First, according to the distribution of edge corner points in each prostate ultrasound image, the abnormal hyperplasia regions in each prostate ultrasound image are obtained, preparing for accurately and efficiently analyzing the loss of hyperplasia details in each prostate ultrasound image in the subsequent process. To accurately analyze the loss of hyperplasia details, the abnormal hyperplasia regions in the prostate ultrasound images of each examination are then matched with those in the adjacent previous examination, and the matching regions of each abnormal hyperplasia region are obtained to accurately determine the regions corresponding to the same hyperplasia site, which is conducive to a more comprehensive analysis of hyperplasia details. Furthermore, based on the distribution differences of pixel values, edge corner points, and blood vessel morphology between each abnormal hyperplasia region and its matching region in each prostate ultrasound image, the degree of retention of hyperplasia details in each prostate ultrasound image is obtained, accurately reflecting the retention of hyperplasia details in each prostate ultrasound image, which is conducive to accurately adjusting the filtering degree in the process of identifying the edges of abnormal hyperplasia regions through the Canny edge detection algorithm. Then, based on the degree of retention of hyperplasia details, the standard deviation of Gaussian filtering in the Canny edge detection process for each prostate ultrasound image is corrected to obtain the corrected standard deviation of Gaussian filtering in the Canny edge detection process for each prostate ultrasound image, enabling the accurate determination of the true edges of abnormal hyperplasia regions in the prostate ultrasound images of each examination, and thus accurately obtaining the abnormal hyperplasia regions of each examination, facilitating the accurate analysis of the changes in abnormal hyperplasia regions, which is conducive to accurately evaluating the efficacy of prostate hyperplasia drugs. Furthermore, based on the true edges of abnormal hyperplasia regions, the efficacy of prostate hyperplasia drugs is accurately evaluated, enabling doctors to timely adjust the treatment plan for prostate patients with drugs, effectively improving the treatment effect of prostate patients. Description of the Drawings

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0016] Figure 1 It is a schematic flowchart of a method for evaluating the efficacy of prostate hyperplasia drugs based on ultrasound images provided by an embodiment of the present invention; Figure 2 It is a flowchart of a method for obtaining the degree of retention of hyperplasia details provided by an embodiment of the present invention; Figure 3 It is a structural diagram of a system for evaluating the efficacy of prostate hyperplasia drugs based on ultrasound images provided by an embodiment of the present invention; Figure 4 It is a schematic diagram of a computer device provided by an embodiment of the present invention. Detailed implementation manners

[0017] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details the specific implementation manners, structures, features and effects of the method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0019] The following specifically describes the specific solution of the method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images provided by the present invention with reference to the accompanying drawings.

[0020] Example 1: The present invention proposes a method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images. Please refer to Figure 1 , which shows a schematic flowchart of a method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images provided by an embodiment of the present invention. The method includes the following steps: Step S1: Obtain each ultrasonic image of the prostate for each examination of the prostate patient.

[0021] Specifically, this embodiment analyzes with a prostate patient and a drug for benign prostatic hyperplasia as an example. It should be noted that the subsequent prostate patients all refer to this prostate patient, and the drugs for benign prostatic hyperplasia all refer to this kind of drug for benign prostatic hyperplasia. In order to analyze the efficacy of the drug for benign prostatic hyperplasia taken by the prostate patient, this embodiment first takes the examination when the doctor first prescribes the drug for benign prostatic hyperplasia to the prostate patient as the first examination. At this time, the prostate patient has not taken the drug for benign prostatic hyperplasia yet; it is known that the treatment of benign prostatic hyperplasia is a process with a relatively long cycle. In order to reasonably and timely analyze the efficacy of the drug for benign prostatic hyperplasia and enable the doctor to timely adjust the treatment plan for the prostate patient, this embodiment sets that the prostate patient comes to the hospital for examination once every month. During each examination, the ultrasonic image of the prostate of the prostate patient is obtained through a high-resolution ultrasonic device. At the same time, it is set that the efficacy of the drug for benign prostatic hyperplasia can be analyzed after the prostate patient comes to the hospital for two more examinations starting from the first examination. The implementer can set the time interval between two adjacent examinations of the prostate patient and the number of examinations for analyzing the efficacy of the drug for benign prostatic hyperplasia according to the actual situation, which is not limited herein.

[0022] In order to accurately examine the prostate hyperplasia of prostate patients, in this embodiment, during each examination, the prostate area of prostate patients is automatically scanned by ultrasound at specified length intervals. Therefore, multiple prostate ultrasound images are obtained during each examination, making the analysis of prostate hyperplasia more accurate. In this embodiment, the specified length is set to 0.5 mm, and the implementer can set the size of the specified length according to the actual situation, which is not limited here.

[0023] In order to avoid interference and accurately and efficiently identify the hyperplastic area in the prostate area, in this embodiment, the Otsu threshold segmentation algorithm is used to automatically distinguish the prostate area from the background area in the prostate ultrasound image. Among them, the Otsu threshold segmentation algorithm is a well-known technology and will not be elaborated here. It should be noted that subsequent analysis is only performed on the prostate area in the prostate ultrasound image.

[0024] Step S2: According to the distribution of edge corner points in each prostate ultrasound image, obtain the abnormal hyperplastic area in each prostate ultrasound image.

[0025] It is known that prostate hyperplasia is a local hyperplasia of prostate tissue, usually manifested as local protrusion or depression of the glandular contour. In the prostate ultrasound image, the morphological changes in the hyperplastic area will cause the edge to be more irregular than the edge of normal tissue. Therefore, in this embodiment, the edge line in each prostate ultrasound image is first obtained by the Canny edge detection algorithm, and then the Harris corner detection algorithm is used to extract the edge corner points on the edge line. When the distribution of edge corner points on a certain edge line is denser and more irregular, it indicates that the area corresponding to this edge line is more likely to be an abnormal hyperplastic area. Therefore, in this embodiment, according to the distribution of edge corner points in each prostate ultrasound image, the abnormal hyperplastic area in each prostate ultrasound image is obtained. Among them, the Canny edge detection algorithm and the Harris corner detection algorithm are both well-known technologies and will not be elaborated here.

[0026] Preferably, in a feasible implementation manner of this embodiment, the method for obtaining the abnormal hyperplastic area is as follows: for any prostate ultrasound image and any edge corner point in this prostate ultrasound image, the preset number of other edge corner points closest to this edge corner point in this prostate ultrasound image are used as the neighborhood corner points of this edge corner point; in this embodiment, the preset number is set to 8, and the implementer can set the size of the preset number according to the actual situation, which is not limited here. The variance of the Euclidean distances between this edge corner point and each of its neighborhood corner points is obtained as the distribution characteristic value of this edge corner point; the larger the distribution characteristic value, the more irregular the distribution of this edge corner point, indirectly indicating that this edge corner point is more likely to be the edge pixel point corresponding to the hyperplastic area. Among them, the method for obtaining the Euclidean distance is a well-known technology and will not be elaborated here; In order to screen out the hyperplastic regions based on the edge corner points, it is necessary to first divide the edge corner points. It is known that the distances between the edge corner points corresponding to the same region are similar, and the distribution characteristic values are also similar. Therefore, in this embodiment, for any two edge corner points in the prostate ultrasound image, the sum of the absolute value of the difference between the distribution characteristic values of the two edge corner points and the Euclidean distance between the two edge corner points is used as the distance metric value of the two edge corner points; the smaller the distance metric value, the more likely the two edge corner points are the edge pixel points corresponding to the same region; therefore, in this embodiment, according to the distance metric values of any two edge corner points in the prostate ultrasound image, the edge corner points in the prostate ultrasound image are clustered by the K-means clustering algorithm to obtain the corner point clustering clusters in the prostate ultrasound image; it should be noted that in this embodiment, the elbow method is used to obtain the K value in the K-means clustering algorithm. Among them, the K-means clustering algorithm and the elbow method are both well-known technologies and will not be elaborated here; It is known that the distribution characteristics of the edge corner points in the hyperplastic region are irregular and dense. Therefore, in this embodiment, according to the distribution of the edge corner points in each corner point clustering cluster, the degree of abnormality of each corner point clustering cluster is obtained; the greater the degree of abnormality, the more likely the region corresponding to the corner point clustering cluster is the hyperplastic region; among them, the method for obtaining the degree of abnormality is: for any corner point clustering cluster, the Euclidean distance between any two edge corner points in the corner point clustering cluster is obtained as the first distance; when the variance of all the first distances is greater, it indicates that the distribution of the edge corner points in the corner point clustering cluster is more irregular, and the region corresponding to the corner point clustering cluster is more likely to be the hyperplastic region; at the same time, when the mean value of all the first distances is smaller, it indicates that the distribution of the edge corner points in the corner point clustering cluster is denser, further indicating that the region corresponding to the corner point clustering cluster is more likely to be the hyperplastic region; therefore, in this embodiment, the normalized result of the product of the variance of all the first distances and the reciprocal of the mean value of all the first distances is used as the degree of abnormality of the corner point clustering cluster; in this embodiment, the norm normalization function is used to normalize the product of the variance of all the first distances and the reciprocal of the mean value of all the first distances; The greater the known abnormal degree is, the more likely the area corresponding to the corner point clustering cluster is a hyperplastic area. Therefore, in this embodiment, the preset abnormal degree threshold is set to 0.5. The implementer can set the size of the preset abnormal degree threshold according to the actual situation, which is not limited here. When the abnormal degree is greater than the preset abnormal degree threshold, the corresponding corner point clustering cluster is regarded as an abnormal cluster, that is, the corner point clustering cluster corresponding to the hyperplastic area is screened out. Considering that in the process of obtaining the edge line through the Canny edge detection algorithm and during the Gaussian filtering process, it is easy to smooth out some edges of the hyperplastic area. Therefore, in this embodiment, for any abnormal cluster, the complete area formed by connecting the edge corner points in the abnormal cluster through the region growing algorithm is used as an abnormal hyperplastic area in this prostate ultrasound image. Among them, the region growing algorithm is a well-known technology and will not be elaborated here.

[0027] So far, the abnormal hyperplastic areas in each prostate ultrasound image are obtained, that is, the hyperplastic parts in each prostate ultrasound image are determined. It should be noted that in this embodiment, for the prostate ultrasound images without abnormal hyperplastic areas, no subsequent analysis is directly performed.

[0028] Step S3: Match the abnormal hyperplastic areas in the prostate ultrasound images of each examination with those in the adjacent previous examination to obtain the matching areas of each abnormal hyperplastic area; according to the distribution differences of pixel values and edge corner points between each abnormal hyperplastic area and its matching area in each prostate ultrasound image, as well as the distribution differences of blood vessel morphologies, obtain the retention degree of hyperplastic details in each prostate ultrasound image.

[0029] Specifically, the abnormal hyperplastic areas obtained in step S3 are not composed of the real edge lines of the abnormal hyperplastic areas, and there is a deviation between the positioning of the abnormal hyperplastic areas and the real hyperplastic areas, which will affect the subsequent accurate analysis of the changes in the abnormal hyperplastic areas and the accurate evaluation of the efficacy of prostate hyperplasia drugs. Therefore, in this embodiment, it is necessary to analyze the loss of hyperplastic details in each prostate ultrasound image, and then adjust the degree of Gaussian filtering in the Canny edge detection algorithm to further correct the edges of each abnormal hyperplastic area, so that the positioning of the abnormal hyperplastic areas is more consistent with the real hyperplastic areas.

[0030] In actual situations, when the current prostate hyperplasia worsens, the gray value of the hyperplastic area will increase and the edge will become more irregular. Therefore, in this embodiment, by analyzing the gray value change and the change in the number of edge corner points of the abnormal hyperplastic area, the hyperplasia condition of the abnormal hyperplastic area is determined; on the other hand, when the prostate hyperplasia worsens, the local blood flow demand in the hyperplastic area will increase, resulting in blood vessel dilation and increased blood flow, forming a finer and denser blood vessel morphology. Therefore, in this embodiment, the hyperplasia condition of the abnormal hyperplastic area can also be determined by analyzing the change in the blood vessel morphology of the abnormal hyperplastic area. When the hyperplasia condition determined by the gray value change and the change in the number of edge corner points of the abnormal hyperplastic area is more consistent with the hyperplasia condition determined by the blood vessel morphology change, it indicates that the positioning of the abnormal hyperplastic area is more the same as the true hyperplastic area, the degree of edge smoothing is smaller, and the loss of hyperplasia details is less. Therefore, in this embodiment, first, the abnormal hyperplastic areas in the prostate ultrasound images of each examination are matched with those in the adjacent previous examination to obtain the matching areas of each abnormal hyperplastic area; among them, the abnormal hyperplastic area and its matching area are essentially the same hyperplastic area in two adjacent examinations. It should be noted that the hyperplastic area of the prostate will not disappear with drug treatment. Therefore, the same abnormal hyperplastic area in two adjacent examinations will definitely be successfully matched. Then, according to the distribution differences of pixel values and edge corner points, as well as the distribution differences of blood vessel morphology, between each abnormal hyperplastic area and its matching area in each prostate ultrasound image, the degree of retention of hyperplasia details in each prostate ultrasound image is obtained. The greater the degree of retention of hyperplasia details, the less adjustment is required for the Gaussian filtering degree in the Canny edge detection algorithm process corresponding to the prostate ultrasound image.

[0031] Preferably, in a feasible implementation manner of this embodiment, the method for obtaining the matching area is as follows: The prostate ultrasound images obtained in each examination are arranged according to the obtained order to form an image sequence for each examination; among them, the process of obtaining the prostate ultrasound images in each examination is exactly the same. Therefore, the number of prostate ultrasound images corresponding to each examination is the same, and the prostate ultrasound images at the same position in different image sequences correspond to the same prostate area. For any prostate ultrasound image in the image sequence of any examination, the prostate ultrasound image at the same position as this prostate ultrasound image in the image sequence of the adjacent previous examination of this examination is used as the matching image of this prostate ultrasound image; For any abnormal hyperplasia region in the prostate ultrasound image, the DeepSORT object tracking algorithm is used to match the abnormal hyperplasia region with the abnormal hyperplasia region in the matching image, and the matching region of the abnormal hyperplasia region is accurately obtained, where the abnormal hyperplasia region and its matching region correspond to the same hyperplasia site. The DeepSORT object tracking algorithm is a well-known technology and will not be elaborated here. It should be noted that there is no adjacent previous examination for the first examination. Therefore, in this embodiment, the matching regions of each abnormal hyperplasia region in each prostate ultrasound image of the first examination are not obtained.

[0032] Preferably, in a feasible implementation manner of this embodiment, for the method of obtaining the degree of retention of hyperplasia details, please refer to Figure 2 , which shows a flowchart of a method for obtaining the degree of retention of hyperplasia details provided in this embodiment. The method includes the following steps: Step S201: Obtain the hyperplasia quantization factor of each abnormal hyperplasia region according to the distribution differences of pixel values and edge corner points in each abnormal hyperplasia region and its matching region.

[0033] When the gray value of an abnormal hyperplasia region is larger and the number of edge corner points is more compared with its matching region, it indicates that the hyperplasia shown by the abnormal hyperplasia region is more severe. Furthermore, in this embodiment, the hyperplasia quantization factor of each abnormal hyperplasia region is obtained according to the distribution differences of pixel values and edge corner points in each abnormal hyperplasia region and its matching region. The larger the hyperplasia quantization factor, the more severe the hyperplasia shown by the corresponding abnormal hyperplasia region.

[0034] In a feasible implementation manner of this embodiment, the method for obtaining the hyperplasia quantization factor is as follows: For any abnormal hyperplasia region, obtain the average value of the gray values of all pixel points in the abnormal hyperplasia region as the first gray value; obtain the average value of the gray values of all pixel points in the matching region of the abnormal hyperplasia region as the second gray value; take the difference between the first gray value and the second gray value as the first hyperplasia speed analysis value; take the difference between the number of edge corner points in the abnormal hyperplasia region and its matching region as the second hyperplasia speed analysis value; when both the first hyperplasia speed analysis value and the second hyperplasia speed analysis value are larger, it indicates that the abnormal hyperplasia region shows a tendency towards severe hyperplasia; in order to accurately obtain the hyperplasia information represented by the abnormal hyperplasia region, furthermore, take the sum result of the first hyperplasia speed analysis value and the second hyperplasia speed analysis value as the hyperplasia quantization factor of the abnormal hyperplasia region.

[0035] Thus, the hyperplasia quantization factor of each abnormal hyperplasia region is obtained.

[0036] Step S202: Obtain the blood flow dynamic change value of each abnormal hyperplasia region according to the distribution differences of blood vessel morphologies in each abnormal hyperplasia region and its matching region.

[0037] When a certain abnormal hyperplasia region shows more curvature and irregularity in blood vessels compared to its matching region, it indicates that the hyperplasia in this abnormal hyperplasia region is more severe. Furthermore, in this embodiment, according to the distribution difference of blood vessel morphology in each abnormal hyperplasia region and its matching region, the blood flow dynamic change value of each abnormal hyperplasia region is obtained. The larger the blood flow dynamic change value, the more severe the hyperplasia in the corresponding abnormal hyperplasia region.

[0038] In a feasible implementation manner of this embodiment, the method for obtaining the blood flow dynamic change value is as follows: It is known that the change in blood vessel morphology corresponding to the hyperplasia region is small and dense. Therefore, for any abnormal hyperplasia region in this embodiment, the color Doppler ultrasound image of this abnormal hyperplasia region is obtained. Since the blood vessel morphology can be accurately analyzed through the color Doppler ultrasound image, and then the color Doppler ultrasound image of this abnormal hyperplasia region is divided into multiple local regions; in this embodiment, the size of the local region is set to , and the implementer can set the size of the local region according to the actual situation, which is not limited here. It should be noted that when dividing the local regions, if the remaining boundary is less than a complete local region, the local region at the boundary may partially overlap with the adjacent local region. In order to accurately analyze the distribution of blood vessels in this abnormal hyperplasia region, and then analyze each local region separately, for any local region, a straight line is fitted through the least squares method according to the positions of all edge pixel points in this local region as the reference straight line of this local region; when the goodness of fit between a certain edge line and the reference straight line is smaller, it indicates that the bending degree of this edge line is larger. When the bending degrees of all edge lines in this local region are larger, it indicates that the blood flow dynamics characterized by this local region are more obvious. Furthermore, in this embodiment, the mean value of the goodness of fit between each edge line in this local region and the reference straight line is obtained as the first index of this local region; the smaller the first index, the more obvious the blood flow dynamics characterized by this local region. Among them, the methods for fitting a straight line by the least squares method and obtaining the goodness of fit are both well-known technologies and will not be elaborated here; In order to more accurately analyze the blood flow dynamics of this local region, further analyze the distribution density of the edge lines in this local region, and then obtain the ratio of the number of edge pixel points to the number of all pixel points in this local region as the second index of this local region; the larger the second index, it indicates that the blood vessel distribution in this local region is denser, and the blood flow dynamics characterized by this local region are more obvious; In order to accurately represent the blood flow dynamics of this local region, in this embodiment, the product of the result of normalizing and negatively correlating the first index and the second index is used as the blood vessel analysis value of this local region; the larger the blood vessel analysis value, the more obvious the blood flow dynamics of this local region.

[0039] Among them, the calculation formula of the blood vessel analysis value is: ; where is the blood vessel analysis value of the a-th local region; is the first index of the a-th local region; is the number of edge pixel points in the a-th local region; is the number of all pixel points in the a-th local region; is the second index of the a-th local region; norm is a normalization function.

[0040] In order to comprehensively analyze the blood flow dynamics of the abnormal hyperplasia region, the mean value of the blood vessel analysis values of all local regions is taken as the blood flow dynamic characteristic value of the abnormal hyperplasia region; the larger the blood flow dynamic characteristic value, the more obvious the blood flow dynamics of the abnormal hyperplasia region.

[0041] In order to analyze whether the change of the abnormal hyperplasia region tends to be more severe, the difference between the blood flow dynamic characteristic value of the abnormal hyperplasia region and its matching region is taken as the blood flow dynamic change value of the abnormal hyperplasia region. The larger the blood flow dynamic change value, the more information of the abnormal hyperplasia region tends to be more severe.

[0042] Thus, the blood flow dynamic change value of each abnormal hyperplasia region is obtained.

[0043] Step S203: Obtain the degree of preservation of hyperplasia details of each prostate ultrasound image according to the correlation between the hyperplasia quantization factor and the blood flow dynamic change value in each prostate ultrasound image.

[0044] When the hyperplasia quantization factor and the blood flow dynamic change value of each abnormal hyperplasia region in a certain prostate ultrasound image both show the same change trend, it indicates that less details of the hyperplasia region in this prostate ultrasound image are lost. Thus, in this embodiment, the degree of preservation of hyperplasia details of each prostate ultrasound image is obtained according to the correlation between the hyperplasia quantization factor and the blood flow dynamic change value in each prostate ultrasound image.

[0045] In a feasible implementation manner of this embodiment, the method for obtaining the degree of proliferation detail retention is as follows: for any prostate ultrasound image, the proliferation quantization factors and blood flow dynamic change values of each abnormal proliferation region in the prostate ultrasound image are sorted in the same order of abnormal proliferation regions to obtain a proliferation quantization factor sequence and a blood flow dynamic change value sequence respectively; it should be noted that the order of abnormal proliferation regions in this prostate ultrasound image can be randomly set and is not limited here, but it must be ensured that the data at the same position in the proliferation quantization factor sequence and the blood flow dynamic change value sequence correspond to the same abnormal proliferation region. Then, the Pearson correlation coefficient of the proliferation quantization factor sequence and the blood flow dynamic change value sequence is obtained. The larger the Pearson correlation coefficient, the more consistent the change trends of the proliferation quantization factors and blood flow dynamic change values of the abnormal proliferation regions, indirectly indicating that the less proliferation detail is lost in this prostate ultrasound image. Therefore, in this embodiment, the result of normalizing the Pearson correlation coefficient of the proliferation quantization factor sequence and the blood flow dynamic change value sequence is used as the degree of proliferation detail retention of this prostate ultrasound image. In this embodiment, the Pearson correlation coefficient of the proliferation quantization factor sequence and the blood flow dynamic change value sequence is normalized by the norm normalization function. Among them, the method for obtaining the Pearson correlation coefficient is a well-known technology and will not be elaborated here.

[0046] Thus, the degree of proliferation detail retention of each prostate ultrasound image is obtained.

[0047] Step S4: Based on the degree of proliferation detail retention, correct the standard deviation of Gaussian filtering during the Canny edge detection process for each prostate ultrasound image, obtain the corrected standard deviation of Gaussian filtering during the Canny edge detection process for each prostate ultrasound image, and determine the true edge of the abnormal proliferation region in the prostate ultrasound image examined each time.

[0048] It is known that when obtaining the edge line in the prostate ultrasound image through the Canny edge detection algorithm, Gaussian filtering is performed on the prostate ultrasound image. The larger the standard deviation of Gaussian filtering, the greater the filtering degree and the easier it is to lose edge details; the smaller the standard deviation of Gaussian filtering, the smaller the filtering degree and the more edge details are retained. In order to avoid the loss of the edge of the proliferation region, resulting in inaccurate recognition of the proliferation region, in this embodiment, based on the degree of proliferation detail retention, the standard deviation of Gaussian filtering during the Canny edge detection process for each prostate ultrasound image is corrected, and the corrected standard deviation of Gaussian filtering during the Canny edge detection process for each prostate ultrasound image is obtained, so as to accurately determine the true edge of the abnormal proliferation region in the prostate ultrasound image examined each time.

[0049] Preferably, in an implementable manner of this embodiment, the method for obtaining the corrected Gaussian filter standard deviation is as follows: for any prostate ultrasound image, the product of the degree of retention of hyperplastic details of the prostate ultrasound image and the Gaussian filter standard deviation during the Canny edge detection process of the prostate ultrasound image is used as the corrected Gaussian filter standard deviation during the Canny edge detection process of the prostate ultrasound image. It should be noted that the Gaussian filter standard deviation during the Canny edge detection process of the prostate ultrasound image is known.

[0050] Thus, the corrected Gaussian filter standard deviation during the Canny edge detection process of each prostate ultrasound image is obtained, and then the true edges of the abnormal hyperplastic regions in each prostate ultrasound image of each examination are accurately determined, which is beneficial to the subsequent timely and accurate evaluation of the efficacy of prostate hyperplasia drugs.

[0051] Step S5: Evaluate the efficacy of prostate hyperplasia drugs based on the true edges of the abnormal hyperplastic regions.

[0052] Specifically, in this embodiment, the three-dimensional construction model obtains the volume of each abnormal hyperplastic region of each examination according to the true edges of the abnormal hyperplastic regions in each prostate ultrasound image of each examination, and the sum of the volumes of all abnormal hyperplastic regions of each examination is used as the total hyperplastic volume of each examination. It should be noted that the total hyperplastic volume of the first examination is not obtained.

[0053] When the total hyperplastic volume detected for the second time is much larger than the total hyperplastic volume detected for the third time, it indicates that the treatment effect of prostate hyperplasia is better, indirectly reflecting that the efficacy of prostate hyperplasia drugs is better. Furthermore, in this embodiment, the difference between the total hyperplastic volume detected for the second time and the total hyperplastic volume detected for the third time is obtained as the volume change value; then the ratio of the volume change value to the total hyperplastic volume detected for the second time is used as the prostate hyperplasia drug efficacy evaluation value. When the prostate hyperplasia drug efficacy evaluation value is larger, the efficacy of prostate hyperplasia drugs is better. Therefore, in this embodiment, the preset evaluation threshold is set to 0.3, and the implementer can set the size of the preset evaluation threshold according to the actual situation, which is not limited here. When the prostate hyperplasia drug efficacy evaluation value is greater than the preset evaluation threshold, it indicates that the treatment with prostate hyperplasia drugs is effective, and prostate patients can continue to take prostate hyperplasia drugs; when the prostate hyperplasia drug efficacy evaluation value is less than or equal to the preset evaluation threshold, it indicates that the volume change of the hyperplastic region is small, indirectly indicating that the efficacy of prostate hyperplasia drugs is limited. At this time, the doctor needs to adjust the drugs taken by prostate patients to optimize the drug treatment plan for prostate patients and improve the treatment effect of prostate patients.

[0054] In summary, in this embodiment, the prostate ultrasound images of each examination of the prostate patient are obtained; the abnormal hyperplasia region is obtained according to the distribution of the edge corner points in the prostate ultrasound image; the abnormal hyperplasia regions in the prostate ultrasound images of two adjacent examinations are matched to obtain the matching regions of the abnormal hyperplasia regions; according to the distribution differences of the pixel values, edge corner points and blood vessel morphology between the abnormal hyperplasia region and its matching region, the degree of retention of hyperplasia details is obtained to correct the standard deviation of Gaussian filtering in the Canny edge detection of the prostate ultrasound image, and the corrected standard deviation of Gaussian filtering is obtained to determine the true edge of the abnormal hyperplasia region. By correcting the standard deviation of Gaussian filtering in the Canny edge detection, the present invention effectively improves the accuracy of the localization of the hyperplasia region, which is beneficial to accurately evaluating the efficacy of prostate hyperplasia drugs.

[0055] Embodiment 2: The present invention also proposes a system for evaluating the efficacy of prostate hyperplasia drugs based on ultrasound images. Please refer to Figure 3 , which shows the structural diagram of a system for evaluating the efficacy of prostate hyperplasia drugs based on ultrasound images provided by an embodiment of the present invention. The system includes: an image acquisition module 10, an abnormal hyperplasia region acquisition module 20, a hyperplasia detail retention degree acquisition module 30, a true edge acquisition module 40, and an evaluation module 50.

[0056] The image acquisition module 10 is used to acquire each prostate ultrasound image of each examination of the prostate patient.

[0057] The abnormal hyperplasia region acquisition module 20 is used to obtain the abnormal hyperplasia region in each prostate ultrasound image according to the distribution of the edge corner points in each prostate ultrasound image.

[0058] The hyperplasia detail retention degree acquisition module 30 is used to match the abnormal hyperplasia regions in the prostate ultrasound images of each examination with the abnormal hyperplasia regions in the prostate ultrasound image of the previous adjacent examination to obtain the matching regions of each abnormal hyperplasia region; according to the distribution differences of the pixel values and edge corner points between each abnormal hyperplasia region and its matching region in each prostate ultrasound image, as well as the distribution difference of the blood vessel morphology, the hyperplasia detail retention degree of each prostate ultrasound image is obtained.

[0059] The true edge acquisition module 40 is used to correct the standard deviation of Gaussian filtering in the process of Canny edge detection of each prostate ultrasound image based on the hyperplasia detail retention degree, obtain the corrected standard deviation of Gaussian filtering in the process of Canny edge detection of each prostate ultrasound image, and determine the true edge of the abnormal hyperplasia region in the prostate ultrasound image of each examination; The evaluation module 50 evaluates the efficacy of prostate hyperplasia drugs based on the true edge of the abnormal hyperplasia region.

[0060] It should be noted that: For the system provided in the above embodiments, only the division of the above functional modules is used as an example for illustration. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, an embodiment of a system for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images and an embodiment of a method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images provided in the above embodiments belong to the same concept. The specific implementation process can be found in the method embodiments and will not be elaborated here.

[0061] Embodiment 3: The present invention also proposes an apparatus for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images. The apparatus includes a memory and a processor. Among them, executable program code is stored in the memory, and the processor is used to call and execute the executable program code to execute a method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images provided in the embodiments of the present application. The apparatus may specifically be a chip, a component or a module. The chip may include a processor and a memory connected thereto; among them, the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute a method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images provided in the above embodiments.

[0062] In addition, the present invention also protects a computer device. Please refer to Figure 4 , 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. Among them, when the processor 402 executes the computer program 403, the computer device can execute any one of the methods for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images introduced above.

[0063] Embodiment 4: The present invention also provides a computer-readable storage medium. Computer program code is stored in the computer-readable storage medium. When the computer program code runs on a computer, the computer is made to execute the above-related method steps to implement a method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images provided in the above embodiments.

[0064] Embodiment 5: The present invention also provides a computer program product. When the computer program product runs on a computer, the computer is made to execute the above-related steps to implement a method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images provided in the above embodiments.

[0065] Among them, the device, computer-readable storage medium, computer program product or chip provided in this embodiment are all used to execute the corresponding method provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method provided above, and will not be elaborated here.

[0066] It should be noted that: the above sequence of embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0067] Each embodiment in this specification is described in a progressive manner, and the same or similar parts among the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images, characterized in that, The method includes the following steps: Obtain each prostate ultrasound image of each examination of the prostate patient; According to the distribution of edge corner points in each prostate ultrasound image, obtain the abnormal hyperplasia area in each prostate ultrasound image; Match the abnormal hyperplasia areas in the prostate ultrasound images of each examination with those in the adjacent previous examination to obtain the matching areas of each abnormal hyperplasia area; According to the distribution differences of pixel values, edge corner points, and vascular morphology between each abnormal hyperplasia area in each prostate ultrasound image and its matching area, obtain the degree of retention of hyperplasia details in each prostate ultrasound image; Based on the degree of retention of hyperplasia details, correct the standard deviation of Gaussian filtering during the Canny edge detection process for each prostate ultrasound image, obtain the corrected standard deviation of Gaussian filtering during the Canny edge detection process for each prostate ultrasound image, and determine the true edge of the abnormal hyperplasia area in the prostate ultrasound image of each examination; Evaluate the efficacy of prostate hyperplasia drugs based on the true edge of the abnormal hyperplasia area.

2. The method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images according to claim 1, wherein The method for obtaining the abnormal hyperplasia area is as follows: For any prostate ultrasound image and any edge corner point in this prostate ultrasound image, use the preset number of other edge corner points closest to this edge corner point in this prostate ultrasound image as the neighborhood corner points of this edge corner point; Obtain the variance of the Euclidean distances between this edge corner point and each of its neighborhood corner points as the distribution eigenvalue of this edge corner point; For any two edge corner points in this prostate ultrasound image, use the sum of the difference in the distribution eigenvalues of these two edge corner points and the Euclidean distance between these two edge corner points as the distance metric value of these two edge corner points; According to the distance metric values of any two edge corner points in this prostate ultrasound image, cluster the edge corner points in this prostate ultrasound image by the K-means clustering algorithm to obtain the corner point clustering clusters in this prostate ultrasound image; According to the distribution of edge corner points in each corner point clustering cluster, obtain the degree of abnormality of each corner point clustering cluster; When the degree of abnormality is greater than the preset abnormality degree threshold, use the corresponding corner point clustering cluster as an abnormal cluster; For any abnormal cluster, use the complete area formed by connecting the edge corner points in this abnormal cluster by the region growing algorithm as an abnormal hyperplasia area in this prostate ultrasound image.

3. The method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images according to claim 2, wherein The method for obtaining the degree of abnormality is as follows: For any corner point clustering cluster, obtain the Euclidean distance between any two edge corner points in this corner point clustering cluster as the first distance; Use the normalized result of the product of the variance of all first distances and the reciprocal of the mean of all first distances as the degree of abnormality of this corner point clustering cluster.

4. The method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images according to claim 1, wherein, The method for obtaining the matching area is as follows: Arrange the prostate ultrasound images obtained in each examination in the order of acquisition as the image sequence of each examination; Among them, the process of obtaining prostate ultrasound images in each examination is exactly the same; For any prostate ultrasound image in the image sequence of any examination, the prostate ultrasound image at the same position as this prostate ultrasound image in the image sequence of the adjacent previous examination of this examination is used as the matching image of this prostate ultrasound image; For any abnormal hyperplasia region in this prostate ultrasound image, the DeepSORT object tracking algorithm is used to match this abnormal hyperplasia region with the abnormal hyperplasia region in the matching image to obtain the matching region of this abnormal hyperplasia region.

5. The method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images according to claim 1, wherein, The method for obtaining the degree of retention of hyperplasia details is as follows: According to the distribution differences of pixel values and edge corner points in each abnormal hyperplasia region and its matching region, the hyperplasia quantization factor of each abnormal hyperplasia region is obtained; According to the distribution differences of blood vessel morphologies in each abnormal hyperplasia region and its matching region, the blood flow dynamic change value of each abnormal hyperplasia region is obtained; According to the correlation relationship between the hyperplasia quantization factor and the blood flow dynamic change value in each prostate ultrasound image, the degree of retention of hyperplasia details in each prostate ultrasound image is obtained.

6. The method for evaluating the efficacy of drugs for prostate hyperplasia based on ultrasonic images according to claim 5, characterized in that The method for obtaining the hyperplasia quantization factor is as follows: For any abnormal hyperplasia region, the mean value of the gray values of all pixel points in this abnormal hyperplasia region is obtained as the first gray value; The mean value of the gray values of all pixel points in the matching region of this abnormal hyperplasia region is obtained as the second gray value; The difference between the first gray value and the second gray value is used as the first hyperplasia speed analysis value; The difference between the number of edge corner points in this abnormal hyperplasia region and its matching region is used as the second hyperplasia speed analysis value; The sum result of the first hyperplasia speed analysis value and the second hyperplasia speed analysis value is used as the hyperplasia quantization factor of this abnormal hyperplasia region.

7. The method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images according to claim 5, wherein, The method for obtaining the blood flow dynamic change value is as follows: For any abnormal hyperplasia region, the color Doppler ultrasound image of this abnormal hyperplasia region is divided into multiple local regions; For any local region, a straight line is fitted according to the positions of all edge pixel points in this local region by the least squares method as the reference straight line of this local region; The mean value of the goodness of fit of each edge line in this local region to the reference straight line is obtained as the first index of this local region; The ratio of the number of edge pixel points in this local region to the number of all pixel points is obtained as the second index of this local region; The product of the normalized and negatively correlated result of the first index and the second index is used as the blood vessel analysis value of this local region; The mean value of the blood vessel analysis values of all local regions is used as the blood flow dynamic characteristic value of this abnormal hyperplasia region; The difference between the blood flow dynamic characteristic values of this abnormal hyperplasia region and its matching region is used as the blood flow dynamic change value of this abnormal hyperplasia region.

8. The method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images according to claim 5, wherein, The method for obtaining the degree of retention of hyperplasia details is as follows: For any prostate ultrasound image, the hyperplasia quantization factors and blood flow dynamic change values of each abnormal hyperplasia region in this prostate ultrasound image are sorted in the same order of abnormal hyperplasia regions, and the hyperplasia quantization factor sequence and the blood flow dynamic change value sequence are obtained respectively; The result of normalizing the Pearson correlation coefficient between the hyperplasia quantization factor sequence and the blood flow dynamic change value sequence is used as the degree of retention of hyperplasia details in this prostate ultrasound image.

9. The method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images according to claim 1, wherein The method for obtaining the modified Gaussian filter standard deviation is as follows: For any prostate ultrasound image, the product of the degree of retention of hyperplasia details in this prostate ultrasound image and the Gaussian filter standard deviation during the Canny edge detection of this prostate ultrasound image is used as the modified Gaussian filter standard deviation during the Canny edge detection of this prostate ultrasound image.

10. The method for evaluating the efficacy of drugs for benign prostatic hyperplasia based on ultrasonic images according to claim 1, wherein, The method for obtaining the edge corner points is as follows: The edge corner points in each prostate ultrasound image are obtained through the Harris corner detection algorithm.

Images