Medical image transmission method and system

Abstract

The invention relates to the technical field of image transmission, and particularly discloses a medical image transmission method and system, and the method comprises the steps: obtaining a medical image and a digital medical record, and classifying the medical image and the digital medical record according to the shooting part of the medical image; for all medical images of the same shooting part, inputting the medical images into the DCT transformation model in sequence to obtain a feature matrix; clustering the medical images according to the feature matrix, querying a digital medical record corresponding to each medical image in each type of clustered medical images, and calculating a transmission value of each medical image according to the digital medical records; and when an image downloading request is received, selecting the medical image based on the transmission value, and transmitting the medical image to a request sending end. According to the invention, the transmission value of the medical image after secondary classification is determined according to the difference between the medical image and the medical image of the same kind, and the image with higher learning value is selected according to the transmission value and sent to the internship staff when the acquisition request is oriented.

Description

Technical Field

[0001] The present invention relates to the field of image transmission technology, and in particular to a medical image transmission method and system. Background Art

[0002] Medical imaging is a procedure performed in the radiology or nuclear medicine department and includes X-rays, ultrasound, CT (computed tomography), MRI (magnetic resonance imaging), mammography, fluoroscopy, PET (positron emission tomography) scans and bone scans, and is extremely common in the current technology context.

[0003] Medical imaging is actually a kind of learning material. For interns, they need to read a lot of medical images to improve their personal level. The number of medical images is very large, and it is difficult for individuals to screen out valuable medical images from multiple medical images. Only high-level staff can organize learning regularly and select classic cases for sharing. This is a concentrated learning method. In actual scenarios, many interns have additional learning needs. Providing a platform for additional learning needs can effectively improve the progress of interns. Summary of the invention

[0004] The object of the present invention is to provide a medical image transmission method and system to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A medical image transmission method, the method comprising:

[0007] Obtain medical images and digital medical records, and classify the medical images and digital medical records according to the shooting parts of the medical images;

[0008] For all medical images of the same shooting part, the medical images are sequentially input into the DCT transformation model to obtain a feature matrix;

[0009] Clustering the medical images according to the characteristic matrix, querying the digital medical records corresponding to each medical image in each category of clustered medical images, and calculating the transmission value of each medical image according to the digital medical records; the transmission value is a numerical value used to indicate the transmission priority of the medical image;

[0010] When an image download request is received, the authority of the request sender is verified. When the request sender passes the verification, the medical image is selected based on the transmission value and transmitted to the request sender.

[0011] As a further solution of the present invention: the step of obtaining medical images and digital medical records and classifying the medical images and digital medical records according to the shooting parts of the medical images includes:

[0012] Obtain shooting parameters of medical images;

[0013] For the shooting parameters of any two medical images, the intersection-and-union ratio is calculated, the medical images are classified according to the intersection-and-union ratio, and the shooting parts are determined simultaneously;

[0014] After the classification is completed, the digital medical record corresponding to each medical image is read, the identity information is removed, and the digital medical record and the medical image after the identity information is removed are connected;

[0015] The process of classifying the medical images according to the intersection-combination comparison and simultaneously determining the shooting part is as follows:

[0016] Medical images with an intersection-over-union ratio greater than a preset threshold are classified into one category, and the shooting parts of the same type of medical images are queried in real time, and the mode of the shooting parts is used as the shooting part of the medical image of this category.

[0017] As a further solution of the present invention: the step of sequentially inputting the medical images of the same photographed part into the DCT transformation model to obtain the feature matrix comprises:

[0018] For all medical images of the same photographed part, the medical images are read in sequence;

[0019] Input the read medical image into the DCT transformation model;

[0020] Determine a sampling size according to the photographed part, and cut out a sub-region in the output of the DCT transform model based on the sampling size;

[0021] The data of the intercepted sub-region is calibrated, and each value is converted to the range of zero to one to obtain a feature matrix.

[0022] As a further solution of the present invention, the steps of clustering the medical images according to the feature matrix, querying the digital medical records corresponding to each medical image in each type of clustered medical images, and calculating the transmission value of each medical image according to the digital medical records include:

[0023] Comparing the feature matrices and calculating the matrix distance;

[0024] clustering the medical images according to the matrix distance;

[0025] For the clustered medical images of the same category, read the digital medical records corresponding to each medical image;

[0026] The transmission value of each medical image is calculated according to the digital medical record.

[0027] As a further solution of the present invention: the content of calculating the transmission value of each medical image according to the digital medical record includes:

[0028] Counting all digital medical records to obtain texts with retained affiliation; the affiliation is used to indicate which digital medical record each word belongs to;

[0029] Calculate the word frequency and inverse document frequency of each word in the text, and calculate the importance value of each word based on the word frequency and inverse document frequency;

[0030] Accumulate the importance values ​​of all words belonging to the same digital medical record as the importance value of the digital medical record;

[0031] The transmission value of the corresponding medical image is determined according to the importance value of the digital medical record; the transmission value is proportional to the importance value.

[0032] As a further solution of the present invention: when receiving an image download request, the request sender is authenticated, and when the request sender passes the authentication, the medical image is selected based on the transmission value, and the step of transmitting the medical image to the request sender includes:

[0033] When receiving an image download request, the request sender is authenticated.

[0034] When the request sending end passes the verification, a part upload request is sent to the request sending end, and the photographed part uploaded by the request sending end is received;

[0035] The medical image is located according to the shooting part, the medical image is selected according to the transmission value, and transmitted to the request sender.

[0036] The technical solution of the present invention also provides a medical image transmission system, the system comprising:

[0037] A data classification module is used to obtain medical images and digital medical records, and classify the medical images and digital medical records according to the shooting parts of the medical images;

[0038] The feature extraction module is used to input the medical images of the same shooting part into the DCT transformation model in turn to obtain the feature matrix;

[0039] a transmission value calculation module, used to cluster the medical images according to the characteristic matrix, query the digital medical records corresponding to each medical image in each category of clustered medical images, and calculate the transmission value of each medical image according to the digital medical records; the transmission value is a numerical value used to indicate the transmission priority of the medical image;

[0040] The authority verification module is used to verify the authority of the request sender when receiving an image download request. When the request sender passes the verification, the medical image is selected based on the transmission value and transmitted to the request sender.

[0041] As a further solution of the present invention: the data classification module includes:

[0042] A parameter acquisition unit, used to acquire shooting parameters of medical images;

[0043] An image classification unit, used to calculate the intersection-and-union ratio for the shooting parameters of any two medical images, classify the medical images according to the intersection-and-union ratio, and simultaneously determine the shooting parts;

[0044] A connection unit is used to read the digital medical record corresponding to each medical image after the classification is completed, remove the identity information, and connect the digital medical record and the medical image after the identity information is removed;

[0045] The process of classifying the medical images according to the intersection-combination comparison and simultaneously determining the shooting part is as follows:

[0046] Medical images with an intersection-over-union ratio greater than a preset threshold are classified into one category, and the shooting parts of the same type of medical images are queried in real time, and the mode of the shooting parts is used as the shooting part of the medical image of this category.

[0047] As a further solution of the present invention: the feature extraction module includes:

[0048] An image reading unit, used for sequentially reading all medical images of the same photographed part;

[0049] A DCT transformation unit, used for inputting the read medical image into a DCT transformation model;

[0050] A region interception unit, configured to determine a sampling size according to the photographed part, and intercept a sub-region in the output of the DCT transform model based on the sampling size;

[0051] The data calibration unit is used to perform data calibration on the intercepted sub-region, and each value is converted to a range from zero to one to obtain a feature matrix.

[0052] As a further solution of the present invention: the transmission value calculation module includes:

[0053] A distance calculation unit, used for comparing the feature matrix and calculating the matrix distance;

[0054] A clustering unit, used for clustering medical images according to the matrix distance;

[0055] A medical record reading unit, used for reading digital medical records corresponding to each medical image of the same type after clustering;

[0056] The calculation execution unit is used to calculate the transmission value of each medical image according to the digital medical record.

[0057] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention classifies the medical images once according to the shooting parts, and then classifies the medical images for a second time according to the differences of the medical images themselves. For the medical images after the second classification, the transmission value is determined according to the differences between them and similar medical images. When facing acquisition requests, images with greater learning value are selected according to the transmission value and sent to the interns. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0059] Figure 1 The figure is a flowchart of the medical image transmission method.

[0060] Figure 2 This is a structural block diagram of the medical image transmission system. DETAILED DESCRIPTION

[0061] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0062] Figure 1 : is a flowchart of a medical image transmission method. In an embodiment of the present invention, a medical image transmission method is provided, the method comprising:

[0063] Step S100: obtaining medical images and digital medical records, and classifying the medical images and digital medical records according to the shooting parts of the medical images;

[0064] Medical imaging refers to non-invasive visual examination of the human body through various technical means. In the context of existing technologies, it is extremely common and there are many types of medical imaging. Digital medical records are digitized medical records stored in computer devices, with a very long storage period and easy access. When obtaining medical images, the digital medical records of the person being photographed are queried simultaneously, and then the medical images and digital medical records are classified according to the shooting parts of the medical images. The transmission needs of medical images addressed by the present invention generally only involve a limited number of parts.

[0065] Step S200: for all medical images of the same photographed part, the medical images are sequentially input into a DCT transformation model to obtain a feature matrix;

[0066] Most medical images are grayscale images because they mainly focus on the density or signal intensity of different tissues, organs or structures. The medical images used in this application are grayscale images by default, and non-grayscale images are not processed, which is sufficient to meet most needs; for all medical images of the same shooting part, the medical images are subjected to DCT transformation. The DCT transformation model is particularly suitable for image processing because the energy of the image is mostly concentrated in the low-frequency area, generally the upper left corner. After the DCT transformation, the area is selected from the upper left corner for interception to obtain the feature matrix.

[0067] Step S300: clustering the medical images according to the feature matrix, querying the digital medical records corresponding to each medical image in each category of clustered medical images, and calculating the transmission value of each medical image according to the digital medical records; the transmission value is a numerical value used to indicate the transmission priority of the medical image;

[0068] The feature matrix is ​​used as the feature of the medical image, the feature matrices of different medical images are compared, the matrix distance is calculated, and the medical images are clustered according to the matrix distance. Compared with the initial classification process based on the shooting part, it can be understood as a secondary classification process for all medical images of the same shooting part. The difference is that the classification process is based on the shooting part and is relatively simple, while the clustering process is more complicated and is implemented based on the matrix distance.

[0069] After clustering is completed, for any type of clustered medical images, the digital medical records of each medical image are queried, and the digital medical records are analyzed to determine the differences between them and other digital medical records. The greater the difference, the more special the corresponding medical image is and the higher the transmission value is.

[0070] Step S400: upon receiving an image download request, performing authority verification on the request sender, and when the request sender passes the verification, selecting a medical image based on the transmission value and transmitting it to the request sender;

[0071] When a request for image download is received, the request sender must first be authenticated. When the request sender passes the verification, the medical image is selected based on the transmission value and fed back to the request sender. Regarding the role of this transmission process, it is essentially a data sharing process for internal personnel. In the work process of the staff, they must always keep learning and keep up with the times. The most common way of progress is to consult a large number of medical records. The existing review method is only a review within a small range. When there are many medical records, the staff are required to screen them by themselves. The solution provided in this application helps the staff to screen. Through computer technology, more valuable images are selected from a large number of medical images and then fed back to the request sender.

[0072] Regarding step S100, the step of acquiring medical images and digital medical records and classifying the medical images and digital medical records according to the shooting parts of the medical images includes:

[0073] Obtain shooting parameters of medical images;

[0074] For the shooting parameters of any two medical images, the intersection-and-union ratio is calculated, the medical images are classified according to the intersection-and-union ratio, and the shooting parts are determined simultaneously;

[0075] After the classification is completed, the digital medical record corresponding to each medical image is read, the identity information is removed, and the digital medical record and the medical image after the identity information is removed are connected;

[0076] The process of classifying the medical images according to the intersection-combination comparison and simultaneously determining the shooting part is as follows:

[0077] Medical images with an intersection-over-union ratio greater than a preset threshold are classified into one category, and the shooting parts of the same type of medical images are queried in real time, and the mode of the shooting parts is used as the shooting part of the medical image of this category.

[0078] In an example of the technical solution of the present invention, the classification process is described. When medical images are taken, there will be shooting parameters. The shooting parameters are an array, which depends on the imaging equipment used and the specific imaging purpose. The specific data involved is pre-set by the staff. X-rays include voltage, current, exposure time and focal length, etc., and CT includes tube voltage, tube current, scanning time, layer thickness and pitch, etc.; in addition, the shooting parameters can also introduce position parameters to characterize the shooting position.

[0079] For the shooting parameters of any two medical images, the intersection-and-union ratio is calculated. The larger the intersection-and-union ratio, the more similar the two images are, and they need to be classified into one category. After comparing the shooting parameters of all medical images in pairs, the classification work is completed. Specifically, the process of classifying medical images according to the intersection-and-union ratio and simultaneously determining the shooting parts is as follows:

[0080] Medical images with an intersection-over-union ratio greater than a preset threshold are classified into one category, and the shooting parts of the same type of medical images are queried in real time. The shooting parts of the medical images of this category are taken as the mode of the shooting parts. That is, for the same type of medical images, each medical image corresponds to a shooting part, and most of these shooting parts are the same, and the same majority is called the mode.

[0081] Regarding step S200, the step of sequentially inputting the medical images of the same photographed part into the DCT transformation model to obtain the feature matrix includes:

[0082] For all medical images of the same photographed part, the medical images are read in sequence;

[0083] Input the read medical image into the DCT transformation model;

[0084] Determine a sampling size according to the photographed part, and cut out a sub-region in the output of the DCT transform model based on the sampling size;

[0085] The data of the intercepted sub-region is calibrated, and each value is converted to the range of zero to one to obtain a feature matrix.

[0086] All medical images corresponding to each photographed part are analyzed separately, and the medical images are read in sequence. The read medical images are input into the DCT transformation model, so that the low-frequency information, that is, the useful information, can be transferred to the upper left corner. Then, starting from the upper left corner, a small area within a preset size is intercepted, called a sub-area. Finally, the values ​​in the sub-area are converted to the range of zero to one to complete the data calibration. The values ​​at each row and column position of the sub-area after data calibration are all between zero and one. The row and column positions are used as matrix element positions, and the sub-area after data calibration can be converted into a feature matrix.

[0087] Regarding step S300, the steps of clustering the medical images according to the feature matrix, querying the digital medical records corresponding to each medical image in each type of clustered medical images, and calculating the transmission value of each medical image according to the digital medical records include:

[0088] Comparing the feature matrices and calculating the matrix distance;

[0089] clustering the medical images according to the matrix distance;

[0090] For the clustered medical images of the same category, read the digital medical records corresponding to each medical image;

[0091] The transmission value of each medical image is calculated according to the digital medical record.

[0092] In an example of the technical solution of the present invention, a clustering scheme is provided. For medical images of the same category, the characteristic matrix is ​​used as its eigenvalue. By comparing the characteristic matrices, the matrix distance can be calculated. The medical images are clustered based on the matrix distance. After the clustering is completed, for each category of medical images, the digital medical records corresponding to each medical image are queried. By comparing the digital medical records, the transmission value of each medical image can be calculated.

[0093] Specifically, the content of calculating the transmission value of each medical image according to the digital medical record includes:

[0094] Counting all digital medical records to obtain texts with retained affiliation; the affiliation is used to indicate which digital medical record each word belongs to;

[0095] Calculate the word frequency and inverse document frequency of each word in the text, and calculate the importance value of each word based on the word frequency and inverse document frequency;

[0096] Accumulate the importance values ​​of all words belonging to the same digital medical record as the importance value of the digital medical record;

[0097] The transmission value of the corresponding medical image is determined according to the importance value of the digital medical record; the transmission value is proportional to the importance value.

[0098] In the above content, the calculation process of word frequency is, for a certain word, obtain the total number of times the word appears in the text, divide the total number by the total number of all words in the text to get the word frequency; the role of word frequency is to measure the importance of a word in a document. The more times a word appears, the greater the word frequency; the calculation process of inverse document frequency is to query the total number of digital medical records, use it as the numerator, and then query the number of digital medical records containing a certain word, add one to the number, use it as the denominator, and get a fraction. Take the logarithm of the fraction to get the inverse document frequency; inverse document frequency is used to measure the prevalence of a word in the entire document set. If a word appears in many documents, it means it is a common word, and the IDF value will be very low. The role of IDF is to reduce the impact of common words that appear frequently in all documents on the importance of the text.

[0099] Furthermore, the importance value can be directly taken as the product of term frequency and inverse document frequency.

[0100] Regarding step S400, when receiving the image download request, the request sender is authenticated, and when the request sender passes the authentication, the medical image is selected based on the transmission value, and the step of transmitting the medical image to the request sender includes:

[0101] When receiving an image download request, the request sender is authenticated.

[0102] When the request sending end passes the verification, a part upload request is sent to the request sending end, and the photographed part uploaded by the request sending end is received;

[0103] The medical image is located according to the shooting part, the medical image is selected according to the transmission value, and transmitted to the request sender.

[0104] The request sender is the port that needs to obtain the image. There are permission requirements for the request sender, and the permission requirements are extremely high. When receiving an image download request, the request sender is verified for permission. When the request sender passes the verification of the platform execution platform through this method, a part upload request is sent to the request sender, and the photographed part uploaded by the request sender is received. The medical image is queried by the photographed part (this corresponds to the classification process in the above content). When the query reaches the medical image, the medical image can be selected according to the transmission value. The higher the transmission value, the greater the probability of being selected.

[0105] Figure 2 : is a structural block diagram of a medical image transmission system. In an embodiment of the present invention, a medical image transmission system, the system 10 includes:

[0106] The data classification module 11 is used to obtain medical images and digital medical records, and classify the medical images and digital medical records according to the shooting parts of the medical images;

[0107] The feature extraction module 12 is used to input the medical images of the same photographed part into the DCT transformation model in sequence to obtain a feature matrix;

[0108] The transmission value calculation module 13 is used to cluster the medical images according to the characteristic matrix, query the digital medical records corresponding to each medical image in each type of clustered medical images, and calculate the transmission value of each medical image according to the digital medical records; the transmission value is a numerical value used to indicate the transmission priority of the medical image;

[0109] The authority verification module 14 is used to perform authority verification on the request sender when receiving an image download request. When the request sender passes the verification, the medical image is selected based on the transmission value and transmitted to the request sender.

[0110] Furthermore, the data classification module 11 includes:

[0111] A parameter acquisition unit, used to acquire shooting parameters of medical images;

[0112] An image classification unit, used to calculate the intersection-and-union ratio for the shooting parameters of any two medical images, classify the medical images according to the intersection-and-union ratio, and simultaneously determine the shooting parts;

[0113] A connection unit is used to read the digital medical record corresponding to each medical image after the classification is completed, remove the identity information, and connect the digital medical record and the medical image after the identity information is removed;

[0114] The process of classifying the medical images according to the intersection-combination comparison and simultaneously determining the shooting part is as follows:

[0115] Medical images with an intersection-over-union ratio greater than a preset threshold are classified into one category, and the shooting parts of the same type of medical images are queried in real time, and the mode of the shooting parts is used as the shooting part of the medical image of this category.

[0116] Specifically, the feature extraction module 12 includes:

[0117] An image reading unit, used for sequentially reading all medical images of the same photographed part;

[0118] A DCT transformation unit, used for inputting the read medical image into a DCT transformation model;

[0119] A region interception unit, configured to determine a sampling size according to the photographed part, and intercept a sub-region in the output of the DCT transform model based on the sampling size;

[0120] The data calibration unit is used to perform data calibration on the intercepted sub-region, and each value is converted to a range from zero to one to obtain a feature matrix.

[0121] In addition, the transmission value calculation module 13 includes:

[0122] A distance calculation unit, used for comparing the feature matrix and calculating the matrix distance;

[0123] A clustering unit, used for clustering medical images according to the matrix distance;

[0124] A medical record reading unit, used for reading digital medical records corresponding to each medical image of the same type after clustering;

[0125] The calculation execution unit is used to calculate the transmission value of each medical image according to the digital medical record.

[0126] The system provided by the technical solution of the present invention can classify medical images once according to the shooting part, and then classify the medical images for a second time according to the differences in the medical images themselves. For the medical images after the second classification, the transmission value is determined according to the differences between them and similar medical images. When facing acquisition requests, images with greater learning value are selected according to the transmission value and sent to the intern staff.

[0127] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware. Based on such an understanding, the above-mentioned technical solution can essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0128] Finally, it should be noted that the medical image transmission method and system disclosed in the embodiments of the present invention are only preferred embodiments of the present invention, and are only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A medical image transmission method, characterized in that: The method comprises: Obtain medical images and digital medical records, and classify the medical images and digital medical records according to the shooting parts of the medical images; For all medical images of the same shooting part, the medical images are sequentially input into the DCT transformation model to obtain a feature matrix; Clustering the medical images according to the characteristic matrix, querying the digital medical records corresponding to each medical image in each category of clustered medical images, and calculating the transmission value of each medical image according to the digital medical records; the transmission value is a numerical value used to indicate the transmission priority of the medical image; When an image download request is received, the authority of the request sender is verified. When the request sender passes the verification, the medical image is selected based on the transmission value and transmitted to the request sender.

2. The medical image transmission method according to claim 1, characterized in that: The steps of obtaining medical images and digital medical records and classifying the medical images and digital medical records according to the shooting parts of the medical images include: Obtain shooting parameters of medical images; For the shooting parameters of any two medical images, the intersection-and-union ratio is calculated, the medical images are classified according to the intersection-and-union ratio, and the shooting parts are determined simultaneously; After the classification is completed, the digital medical record corresponding to each medical image is read, the identity information is removed, and the digital medical record and the medical image after the identity information is removed are connected; The process of classifying the medical images according to the intersection-combination comparison and simultaneously determining the shooting part is as follows: Medical images with an intersection-over-union ratio greater than a preset threshold are classified into one category, and the shooting parts of the same type of medical images are queried in real time, and the mode of the shooting parts is used as the shooting part of the medical images of this category.

3. The medical image transmission method according to claim 2, characterized in that: The step of sequentially inputting the medical images of the same photographed part into the DCT transformation model to obtain the feature matrix comprises: For all medical images of the same photographed part, the medical images are read in sequence; Input the read medical image into the DCT transformation model; Determine a sampling size according to the photographed part, and extract a sub-region in the output of the DCT transform model based on the sampling size; The data of the intercepted sub-region is calibrated, and each value is converted to the range of zero to one to obtain a feature matrix.

4. The medical image transmission method according to claim 1, characterized in that: The steps of clustering the medical images according to the feature matrix, querying the digital medical records corresponding to each medical image in each type of clustered medical images, and calculating the transmission value of each medical image according to the digital medical records include: Comparing the feature matrices and calculating the matrix distance; clustering the medical images according to the matrix distance; For the clustered medical images of the same category, read the digital medical records corresponding to each medical image; The transmission value of each medical image is calculated based on the digital medical record.

5. The medical image transmission method according to claim 4, characterized in that: The content of calculating the transmission value of each medical image according to the digital medical record includes: Counting all digital medical records to obtain texts with retained affiliation; the affiliation is used to indicate which digital medical record each word belongs to; Calculate the word frequency and inverse document frequency of each word in the text, and calculate the importance value of each word based on the word frequency and inverse document frequency; The importance values ​​of all words belonging to the same digital medical record are accumulated as the importance value of the digital medical record; The transmission value of the corresponding medical image is determined according to the importance value of the digital medical record; the transmission value is proportional to the importance value.

6. The medical image transmission method according to claim 5, characterized in that: When receiving the image download request, the step of performing authority verification on the request sender, and when the request sender passes the verification, selecting the medical image based on the transmission value and transmitting it to the request sender includes: When receiving an image download request, the request sender is authenticated. When the request sending end passes the verification, a part upload request is sent to the request sending end, and the photographed part uploaded by the request sending end is received; The medical image is located according to the shooting part, the medical image is selected according to the transmission value, and transmitted to the request sender.

7. A medical image transmission system, characterized in that: The system comprises: A data classification module is used to obtain medical images and digital medical records, and classify the medical images and digital medical records according to the shooting parts of the medical images; The feature extraction module is used to input the medical images of the same shooting part into the DCT transformation model in turn to obtain the feature matrix; a transmission value calculation module, used to cluster the medical images according to the characteristic matrix, query the digital medical records corresponding to each medical image in each category of clustered medical images, and calculate the transmission value of each medical image according to the digital medical records; the transmission value is a numerical value used to indicate the transmission priority of the medical image; The authority verification module is used to verify the authority of the request sender when receiving an image download request. When the request sender passes the verification, the medical image is selected based on the transmission value and transmitted to the request sender.

8. The medical image transmission system according to claim 7, characterized in that: The data classification module comprises: A parameter acquisition unit, used to acquire shooting parameters of medical images; An image classification unit, used to calculate the intersection-and-union ratio for the shooting parameters of any two medical images, classify the medical images according to the intersection-and-union ratio, and simultaneously determine the shooting parts; A connection unit is used to read the digital medical record corresponding to each medical image after the classification is completed, remove the identity information, and connect the digital medical record and the medical image after the identity information is removed; The process of classifying the medical images according to the intersection-combination comparison and simultaneously determining the shooting part is as follows: Medical images with an intersection-over-union ratio greater than a preset threshold are classified into one category, and the shooting parts of the same type of medical images are queried in real time, and the mode of the shooting parts is used as the shooting part of the medical images of this category.

9. The medical image transmission system according to claim 8, characterized in that: The feature extraction module comprises: An image reading unit, used for sequentially reading all medical images of the same photographed part; A DCT transformation unit, used for inputting the read medical image into a DCT transformation model; A region interception unit, configured to determine a sampling size according to the photographed part, and intercept a sub-region in the output of the DCT transform model based on the sampling size; The data calibration unit is used to perform data calibration on the intercepted sub-region, and each value is converted to a range from zero to one to obtain a feature matrix.

10. The medical image transmission system according to claim 7, characterized in that: The transmission value calculation module includes: A distance calculation unit, used for comparing the feature matrix and calculating the matrix distance; A clustering unit, used for clustering medical images according to the matrix distance; A medical record reading unit, used for reading digital medical records corresponding to each medical image of the same type after clustering; The calculation execution unit is used to calculate the transmission value of each medical image according to the digital medical record.

Images