Search

CN-121982351-A - Scanning image histogram joint analysis method and system

CN121982351ACN 121982351 ACN121982351 ACN 121982351ACN-121982351-A

Abstract

The invention relates to a method and a system for joint analysis of a histogram of a scanned image, wherein the method comprises the steps of collecting the scanned image of a user, determining a target scanned image from the scanned image, expanding the outline of a marked interested object to obtain an expanded object, generating a first gray level histogram of the interested object and a second gray level histogram of the expanded object, acquiring quantitative parameters, comparing the quantitative parameters with quantitative parameter templates corresponding to various classifications to determine classifications, and sending the classifications to a user terminal as joint analysis results. The invention fully utilizes the layering scanning images, fully utilizes the auxiliary information contained in each layer of scanning images in a weight decreasing mode, and greatly improves the joint analysis efficiency.

Inventors

  • JIN JIANGUO

Assignees

  • 杭州市临平区第一人民医院

Dates

Publication Date
20260505
Application Date
20231128

Claims (10)

  1. 1. A method of joint analysis of a histogram of a scanned image, the method comprising: step S1, collecting a plurality of scanning images of a user, wherein the scanning images correspond to a plurality of different scanning levels respectively; Step S2, determining a target scanning image from the scanning images, determining an object of interest from the target scanning image, increasing the determination times c of the target scanning image, and setting a historical similarity Sim c-1 ＝Sim＝Sim c ; the method comprises the steps of determining target scanning images from the scanning images, namely sorting the scanning images according to the size of the object of interest in the scanning images, and selecting the scanning image with the largest size of the object of interest, which is not determined as the target scanning image, in the scanning images as the target scanning image; Step S3, expanding the marked interesting object outline to obtain an expanded object; Step S4, generating a first gray level histogram of the object of interest and a second gray level histogram of the extended object, and acquiring quantitative parameters related to the first gray level histogram and/or the second gray level histogram, wherein the quantitative parameters comprise a first quantitative parameter related to the object of interest, a second quantitative parameter related to the extended object, and a third quantitative parameter related to the object of interest and the extended object at the same time; Step S5, comparing the quantitative parameters with quantitative parameter templates corresponding to all the classifications to determine whether similar classifications exist, if so, sending the classifications to the user terminal as a joint analysis result, otherwise, returning to the step S2; The step S5 specifically includes the following steps: Step S51, pre-storing quantitative parameter templates corresponding to various classifications; Step S52, respectively comparing the quantitative parameters with the first quantitative parameter, the second quantitative parameter and the third quantitative parameter of the quantitative parameter template to obtain the similarity thereof, and adopting the following formula (1) to obtain the current similarity Sim c , wherein Sim1, sim2 and Sim3 are the first similarity, the second similarity and the third similarity corresponding to the first quantitative parameter, the second quantitative parameter and the third quantitative parameter respectively, alpha 1, alpha 2 and alpha 3 are preset weight coefficients, and adopting Euclidean distance to calculate the first similarity, the second similarity and the third similarity; Sim c =α1×Sim1+α2×Sim2+α3×Sim3(1); Step S53, calculating the similarity Sim based on the current similarity and the historical similarity Sim c-1 , specifically, calculating the similarity Sim based on the following formula (5), wherein beta 1 and beta 2 are balance coefficients; Sim=β1×Sim c +β2×Sim c-1 (5); Step S54, if a certain amount of parameter templates exist in quantitative parameter templates corresponding to one classification so that the similarity Sim is smaller than or equal to a similarity threshold, determining that the one classification is similar, sending the similar classification to the user terminal as a joint analysis result, and otherwise, returning to step S2.
  2. 2. The system for joint analysis of histograms of scanned images according to claim 1, characterized in that said increasing target scanned image determines a number c, in particular by setting c=c+1.
  3. 3. The system of claim 2, wherein the dual-phase enhanced scanning is performed using a bolus tracking technique according to a 1ml/kg body mass.
  4. 4. A scanned image histogram joint analysis system as claimed in claim 3, wherein the initial value of c is set to 1.
  5. 5. The system of claim 4, wherein the object of interest is determined from the target scan image by image analysis or manual annotation.
  6. 6. A scanning image histogram joint analysis system is characterized by comprising a user terminal and a joint analysis server, wherein the system is used for realizing the scanning image histogram joint analysis method as set forth in any one of claims 1-5.
  7. 7. A processor for running a program, wherein the program is operative to perform the method of joint analysis of a histogram of a scanned image as claimed in any one of claims 1 to 5.
  8. 8. An execution device comprising a processor coupled to a memory, the memory storing program instructions that when executed by the processor implement the method of joint analysis of a scanned image histogram of any one of claims 1-5.
  9. 9. A computer readable storage medium comprising a program which, when run on a computer, causes the computer to perform the scanned image histogram joint analysis method as claimed in any one of claims 1 to 5.
  10. 10. A cloud server configured to perform the scanned image histogram joint analysis method of any one of claims 1-5.

Description

Scanning image histogram joint analysis method and system [ Field of technology ] The invention belongs to the technical field of intelligent medical treatment, and particularly relates to a method and a system for joint analysis of a scanning image histogram. [ Background Art ] On the one hand, with the continuous development of science and technology, the living standard of the general public is continuously improved, the physical condition and the health level of the general public are more concerned, and under the trend of the calculation development of the Chinese Internet and big data cloud, the country greatly promotes the strategy of Internet plus. The intelligent medical service is a service mode taking the patient as the center of the Internet and medical treatment, and the medical service can be completed by the patient without going out, so that medical resources of hospitals are greatly saved, more energy is applied to the study on the difficult problem in the medical field by comprehensive hospitals, and the development of medical science and technology is promoted. On the other hand, image recognition is one of the most active fields in the field of computer vision, and the main research aim is to utilize the powerful computing capability of a computer to help human beings automatically process massive physical information and to recognize targets in various different modes to replace part of mental labor of the human beings. Image recognition integrates many disciplines including computer science and technology, physics, statistics, neurobiology, etc., and is widely used in many fields such as geological exploration, image remote sensing, robot vision, biomedicine, etc. The medical image is utilized, the slight gray scale difference in the scanning image which can not be distinguished by naked eyes can be distinguished through computer software analysis, and the characteristics of the scanning image are analyzed, so that the scanning image is widely used in clinical assistance and intelligent medical treatment. Medical institutions generate a large number of medical images each day that contain a large amount of potential information, but these potential information are underutilized. In addition, the outline of the interested object is often not clear, an exact labeling result is difficult to form even if manual labeling is carried out, and important information is often omitted only by a traditional scanning image analysis method. Based on the three-dimensional characteristic of the object of interest, the method fully utilizes the layered scanning images, fully utilizes the auxiliary information contained in each layer of scanning images in a weight decreasing mode, and greatly improves the joint analysis efficiency. [ Invention ] In order to solve the above problems in the prior art, the present invention proposes a method for joint analysis of histograms of scanned images and the same, said method comprising: step S1, collecting a plurality of scanning images of a user, wherein the scanning images correspond to a plurality of different scanning levels respectively; Step S2, determining a target scanning image from the scanning images, determining an object of interest from the target scanning image, increasing the determination times c of the target scanning image, and setting a historical similarity Sim c-1＝Sim＝Simc; the method comprises the steps of determining target scanning images from the scanning images, namely sorting the scanning images according to the size of the object of interest in the scanning images, and selecting the scanning image with the largest size of the object of interest, which is not determined as the target scanning image, in the scanning images as the target scanning image; Step S3, expanding the marked interesting object outline to obtain an expanded object; Step S4, generating a first gray level histogram of the object of interest and a second gray level histogram of the extended object, and acquiring quantitative parameters related to the first gray level histogram and/or the second gray level histogram, wherein the quantitative parameters comprise a first quantitative parameter related to the object of interest, a second quantitative parameter related to the extended object, and a third quantitative parameter related to the object of interest and the extended object at the same time; Step S5, comparing the quantitative parameters with quantitative parameter templates corresponding to all the classifications to determine whether similar classifications exist, if so, sending the classifications to the user terminal as a joint analysis result, otherwise, returning to the step S2; The step S5 specifically includes the following steps: Step S51, pre-storing quantitative parameter templates corresponding to various classifications; Step S52, respectively comparing the quantitative parameters with the first quantitative parameter, the second quantitative paramet