An intelligent analysis method and system for the composition and structure of sacral nerves

Abstract

The present invention relates to the field of sacral nerve testing technology, and is a method and system for intelligent analysis of the composition structure of the sacral nerve, including: screening a subject information set to obtain an available information set, extracting available information from the available information set in sequence, obtaining multiple sacral nerve image sets corresponding to the available information, and performing the following operations on the sacral nerve image sets in the multiple sacral nerve image sets: calculating the walking data set in the sacral nerve image set based on the sacral nerve image set, using the walking data set to construct the sacral nerve identification data corresponding to the sacral nerve image set, summarizing the sacral nerve identification data to obtain the sacral nerve identification data set, parsing the sacral nerve identification data set, obtaining a sacral nerve report corresponding to the available information, importing the sacral nerve report into the available information, and completing the intelligent analysis of the composition structure of the sacral nerve. The present invention can optimize the process of identifying the composition structure of the sacral nerve, accurately extract the composition structure data of the sacral nerve, and avoid wasting medical resources.

Description

Technical Field

[0001] The present invention relates to the technical field of sacral nerve testing, and in particular to an intelligent analysis method and system for the composition structure of sacral nerves. Background Art

[0002] The sacral nerve is a nerve in the human pelvis. If the sacral nerve becomes abnormal, it will cause imbalance in the regulation of various organs in the human spine and pelvis, causing disease. Therefore, during health checks, the composition and structure of the sacral nerve should be screened to ensure the health of the body.

[0003] In the prior art, when a large number of subjects wish to undergo pelvic nerve screening, the screening information is mostly manually screened by nurses, and an MRI or CT scan is performed on the screened subjects. Generally, several physicians with many years of imaging experience observe the scan results and manually measure the anterior or lateral angle of the sacral nerve from the image to determine whether the sacral nerve structure is abnormal.

[0004] Although the above method can measure the anterior and lateral angles of the sacral nerve, when measuring on the image, the measurement results may be inaccurate due to errors in the measurement points selected by the physician or improper locations of the selected measurement points. At the same time, manual information screening is time-consuming and laborious and needs to be optimized. Summary of the invention

[0005] The present invention provides an intelligent analysis method for the composition structure of sacral nerves, a computer can optimize the process of identifying the composition structure of sacral nerves, accurately extract the composition structure data of sacral nerves, and avoid wasting medical resources.

[0006] To achieve the above objectives, the present invention provides an intelligent analysis method for the composition structure of the sacral nerve, comprising:

[0007] Obtain a subject information set, filter the subject information set, obtain an available information set, extract available information from the available information set in sequence, and perform the following operations on the extracted available information:

[0008] A plurality of sacral nerve image sets corresponding to the available information are obtained, wherein the plurality of sacral nerve image sets are: an S1 nerve image set, an S2 nerve image set, an S3 nerve image set, and an S4 nerve image set, and the following operations are performed on the sacral nerve image sets in the plurality of sacral nerve image sets:

[0009] Calculating a running data set in the sacral nerve image set based on the sacral nerve image set, and constructing sacral nerve identification data corresponding to the sacral nerve image set using the running data set, wherein the running data set includes an outer running angle or a front running angle;

[0010] Summarizing the sacral nerve identification data to obtain a sacral nerve identification data set, parsing the sacral nerve identification data set to obtain a sacral nerve report corresponding to the available information, and importing the sacral nerve report into the available information to obtain updated information;

[0011] The updated information is summarized to obtain an updated information set corresponding to the available information set, and intelligent analysis of the composition structure of the sacral nerve is completed based on the updated information set.

[0012] Optionally, the obtaining of the subject information set, screening the subject information set, and obtaining the available information set includes:

[0013] Acquire a medical entity set using a pre-constructed corpus, wherein the medical entity set includes a plurality of medical entities, and the medical entities are included in a pre-constructed spine entity set or a pre-constructed pelvic entity set;

[0014] Obtain a subject information set, extract subject information from the subject information set in sequence, and perform the following operations on the extracted subject information:

[0015] Obtaining an electronic medical record corresponding to the subject information, and obtaining a set of sentences to be identified based on the electronic medical record, wherein the set of sentences to be identified includes zero, one or more sentences to be identified;

[0016] If the sentence set to be identified is an empty set, the subject information is confirmed as suboptimal information;

[0017] If the set of sentences to be identified is not an empty set, extracting the sentences to be identified from the set of sentences to be identified in sequence, calculating the first similarity based on the extracted sentences to be identified and the medical entity set, summarizing the first similarities, and obtaining a first similarity set corresponding to the set of sentences to be identified;

[0018] Extracting a high similarity set from the first similarity set, wherein the high similarity set includes zero, one or more first similarities greater than a preset similarity threshold;

[0019] If the high similarity set is an empty set, the sentence set to be recognized and the medical entity set are introduced into the pre-built second language recognition model to obtain a second similarity set.

[0020] Extracting a second high similarity set from the second similarity set, if the second high similarity set is an empty set, confirming the extracted subject information as suboptimal information, otherwise, skipping the extracted subject information;

[0021] If the high similarity set includes one or more first similarities greater than the similarity threshold, skipping the extracted subject information;

[0022] Suboptimal information is aggregated to obtain a suboptimal information set, and a usable information set is obtained based on the suboptimal information set.

[0023] Optionally, the calculation formula of the first similarity is as follows:

[0024] ;

[0025] in, represents the first similarity, represents a medical entity set, and , Represents the first Medical entities, Represents the first Medical entities, Represents the first Medical entities, Represents the first Medical entities, represents the sentence to be recognized, Indicates The serial number of the medical entity, Represents the total number of medical entities in the medical entity set, Indicates The weight of the medical entity in the medical entity set, Indicates the first The frequency of occurrence of medical entities, Indicates Semantic discrimination of characteristic words of medical entities, represents the first hyperparameter, represents the second hyperparameter, Indicates the length of the sentence to be recognized. Indicates The length of the medical entity in characters.

[0026] Optionally, acquiring the available information set based on the suboptimal information set includes:

[0027] The following operations are performed on the suboptimal information in the suboptimal information set:

[0028] acquiring sacral nerve scanning data based on the suboptimal information, constructing a sacral nerve model based on the sacral nerve scanning data, and performing an image reconstruction operation on the sacral nerve model to obtain a plurality of sacral nerve image sets;

[0029] The following operations are performed on the sacral nerve image sets in the multiple sacral nerve image sets:

[0030] Sacral nerve images are extracted from the sacral nerve image set in sequence, and the following operations are performed on the extracted sacral nerve images:

[0031] Obtaining a clarity value calculation formula, calculating the clarity value of the extracted sacral nerve image based on the clarity value calculation formula, comparing the clarity value with a preset clarity threshold, and if the clarity value is less than the clarity threshold, confirming the extracted sacral nerve image as an unclear image, otherwise, skipping the extracted sacral nerve image;

[0032] Summarizing the unclear images to obtain unclear image sets of multiple sacral nerve image sets, and if the unclear image sets are empty sets, confirming suboptimal information corresponding to the multiple sacral nerve image sets as available information, otherwise, skipping the suboptimal information corresponding to the multiple sacral nerve image sets;

[0033] Summarize the available information to obtain the available information set.

[0034] Optionally, the obtaining of a clarity value calculation formula, and calculating the clarity value of the extracted sacral nerve image based on the clarity value calculation formula, includes:

[0035] Constructing a horizontal convolution template and a vertical convolution template, and using a pre-constructed image coordinate system to obtain the horizontal convolution direction and the vertical convolution direction of the extracted sacral nerve image;

[0036] Using the horizontal convolution template to perform convolution calculation in the horizontal convolution direction to obtain multiple horizontal gradient values, and using the vertical convolution template to perform convolution calculation in the vertical convolution direction to obtain multiple vertical gradient values, wherein the horizontal gradient values ​​and the vertical gradient values ​​correspond to each other one by one;

[0037] The clarity value of the extracted sacral nerve image is calculated based on multiple horizontal gradient values, multiple vertical gradient values ​​and a clarity value calculation formula, wherein the clarity value calculation formula is as follows:

[0038] ;

[0039] in, represents the clarity value of the extracted sacral nerve image, Represents the horizontal coordinate of the pixel point in the sacral nerve image in the image coordinate system, Represents the ordinate of the pixel point in the sacral nerve image in the image coordinate system, Represents the available gradient value of a pixel in the sacral nerve image.

[0040] Optionally, the calculation formula of the available gradient value is as follows:

[0041] ;

[0042] in, represents the gradient value of the pixel in the sacral nerve image, Indicates the available threshold value of the gradient value;

[0043] The calculation formula of the gradient value is as follows:

[0044] ;

[0045] ;

[0046] ;

[0047] in, Represents the gray value of the pixel in the sacral nerve image, represents the convolution operator, represents the horizontal convolution template, represents the vertical convolution template, Represents the horizontal gradient value of the pixel in the sacral nerve image, Represents the vertical gradient value of the pixel in the sacral nerve image.

[0048] Optionally, the calculating of the running data set in the sacral nerve image set based on the sacral nerve image set includes:

[0049] The identification name is confirmed based on the sacral nerve image set, sacral nerve images are extracted from the sacral nerve image set in sequence, and the following operations are performed on the extracted sacral nerve images:

[0050] Confirming the running direction based on the extracted sacral nerve image, extracting a running detection area image from the extracted sacral nerve image, performing a binarization operation on the running detection area image to obtain a binary image, acquiring a detection image coordinate system based on the binary image, and extracting all neural pixel points from the binary image to obtain a neural pixel point set, wherein the neural pixel point is a pixel point with a gray value of 255 in the binary image;

[0051] Using the detection image coordinate system and the pre-constructed straight line fitting method, the neural pixel point set is fitted into a straight line in the detection image coordinate system to obtain the sacral nerve straight line, and the sacral nerve straight line equation is obtained according to the sacral nerve straight line. The straight line slope is extracted from the sacral nerve straight line equation, and the absolute value of the extracted straight line slope is taken to obtain the running angle, and the running data is generated according to the identification name, running angle and running direction;

[0052] The running data are summarized to obtain the running data set corresponding to the sacral nerve image set.

[0053] Optionally, parsing the sacral nerve identification dataset to obtain a sacral nerve report corresponding to available information includes:

[0054] Extract the left S1 nerve anterior running angle, the left S2 nerve anterior running angle, and the S3 nerve anterior running angle from the sacral nerve identification data set, and construct a left anterior running angle sequence based on the left S1 nerve anterior running angle, the left S2 nerve anterior running angle, and the S3 nerve anterior running angle sequence, and obtain a left anterior running angle report based on the left anterior running angle sequence;

[0055] Obtain the right anterior course angle report based on the sacral nerve marker dataset;

[0056] Extract the external running angle of the left S1 nerve, the external running angle of the left S2 nerve, the external running angle of the left S3 nerve, and the external running angle of the outer S4 nerve from the sacral nerve identification data set, and construct a left external running angle sequence based on the external running angle of the left S1 nerve, the external running angle of the left S2 nerve, the external running angle of the left S3 nerve, and the external running angle of the outer S4 nerve, and obtain a left external running angle report based on the left external running angle sequence;

[0057] Get the right external course angle report based on the sacral nerve identification dataset;

[0058] The left anterior running angle report, the right anterior running angle report, the left outer running angle report and the right outer running angle report are summarized to obtain the sacral nerve report corresponding to the available information.

[0059] Optionally, obtaining a left front running angle report based on a left front running angle sequence includes:

[0060] Extract the target angles from the left front running angle sequence in sequence, and perform the following operations on the extracted target angles:

[0061] Acquire an adjacent angle based on the target angle, wherein the adjacent angle is the next target angle in the left front running angle sequence adjacent to the target angle;

[0062] Compare the target angle and the adjacent angles. If the target angle is smaller than the adjacent angle, a normal report is generated based on the target angle and the adjacent angles. Otherwise, an abnormal report is generated.

[0063] Normal reports are summarized to obtain a normal report set, abnormal reports are summarized to obtain an abnormal report set, and a left front running angle report is obtained based on the normal report set and the abnormal report set.

[0064] To achieve the above objectives, the present invention also provides an intelligent analysis system for the composition and structure of the sacral nerve, comprising:

[0065] An information screening module, used to obtain a subject information set, screen the subject information set, and obtain a usable information set;

[0066] The sacral nerve identification data acquisition module is used to extract available information from the available information set in sequence, and perform the following operations on the extracted available information: obtain multiple sacral nerve image sets corresponding to the available information, wherein the multiple sacral nerve image sets are respectively: S1 nerve image set, S2 nerve image set, S3 nerve image set and S4 nerve image set, and perform the following operations on the sacral nerve image sets in the multiple sacral nerve image sets: calculate the running data set in the sacral nerve image set based on the sacral nerve image set, and construct the sacral nerve identification data corresponding to the sacral nerve image set using the running data set, wherein the running data set includes an external running angle or a front running angle;

[0067] A parsing module, for summarizing sacral nerve identification data to obtain a sacral nerve identification data set, parsing the sacral nerve identification data set to obtain a sacral nerve report corresponding to available information, and importing the sacral nerve report into available information to obtain updated information;

[0068] The summarizing module is used to summarize the updated information, obtain the updated information set corresponding to the available information set, and complete the intelligent analysis of the composition structure of the sacral nerve based on the updated information set.

[0069] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0070] A memory storing at least one instruction;

[0071] The processor executes the instructions stored in the memory to implement the above-mentioned intelligent analysis method for the composition structure of the sacral nerve.

[0072] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned intelligent analysis method for the composition structure of the sacral nerve.

[0073] The present invention is to solve the problem described in the background technology. The present invention obtains a subject information set, screens the subject information set, obtains an available information set, and extracts available information from the available information set in sequence. The present invention uses a language model to screen the subject's electronic medical record, removes the subject information with a history of spinal lesions and a history of pelvic lesions, and also removes the information of the sacral nerve image with poor quality in the sacral nerve image set, aiming to extract the subject information without a history of spinal lesions and a history of pelvic lesions, and the sacral nerve images in multiple sacral nerve image sets are all clear images, so as to screen out the information that can meet the health detection needs of the subject by only performing feature detection of the sacral nerve composition structure. The present invention performs the following operations on the extracted available information: obtain multiple sacral nerve image sets corresponding to the available information, and perform the following operations on the sacral nerve image sets in the multiple sacral nerve image sets: calculate the running data set in the sacral nerve image set based on the sacral nerve image set, and use the running data set to construct the sacral nerve identification data corresponding to the sacral nerve image set. The sacral nerve identification data constructed by the image calculation method can describe the external running angle and the front running angle of the sacral nerve, thereby realizing the extraction of the composition structure data of the sacral nerve. Sacral nerve identification data are summarized to obtain a sacral nerve identification data set, and the sacral nerve identification data set is analyzed to obtain a sacral nerve report corresponding to the available information. The present invention uses the statistical significance of the anterior and lateral running angles between the existing different sacral nerves to perform intelligent analysis of the sacral nerve composition structure of the subjects with available information corresponding to the sacral nerve identification data set to determine whether the sacral nerve of the subject is abnormal. Therefore, the present invention can optimize the process of identifying the composition structure of the sacral nerve, accurately extract the composition structure data of the sacral nerve, and avoid wasting medical resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 A schematic diagram of a flow chart of an intelligent analysis method for the composition structure of sacral nerves provided by an embodiment of the present invention;

[0075] Figure 2 A functional module diagram of an intelligent analysis system for the composition structure of sacral nerves provided by an embodiment of the present invention;

[0076] Figure 3 A schematic diagram of the structure of an electronic device for implementing the intelligent analysis method for the composition structure of the sacral nerves provided by an embodiment of the present invention.

[0077] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0078] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0079] The embodiment of the present application provides a method for intelligent analysis of the composition structure of the sacral nerve. The execution subject of the method for intelligent analysis of the composition structure of the sacral nerve includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for intelligent analysis of the composition structure of the sacral nerve can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0080] Reference Figure 1 FIG. 1 is a flow chart of a method for intelligently analyzing the structure of sacral nerves according to an embodiment of the present invention. In this embodiment, the method for intelligently analyzing the structure of sacral nerves includes:

[0081] S1. Obtain a subject information set, filter the subject information set, and obtain an available information set.

[0082] It is understandable that the subject information set is a collection of multiple subject information, and the subject information set includes multiple subject information, and the subject information is the information of the person who participates in the analysis of the sacral nerve composition structure, including but not limited to: name, ID number, age, electronic medical record, etc. The embodiment of the present invention aims to realize intelligent recognition of the composition structure of the sacral nerve, thereby realizing the detection of sacral nerve characteristics of subjects who have no history of spinal lesions and pelvic lesions, and optimizing the steps of using imaging to detect the sacral nerve structure.

[0083] For example, there is a group of subjects, some of whom are known to have a history of spinal and pelvic lesions. If a person has a history of spinal and pelvic lesions, the sacral nerve will be affected by the lesions and a more comprehensive examination is required. Therefore, only the characteristic detection of the sacral nerve structure cannot meet the health detection needs of the subjects. Therefore, the testing department removes the information of these patients and only retains the subjects who have no history of spinal and pelvic lesions.

[0084] Furthermore, the obtaining of the subject information set, screening the subject information set, and obtaining the available information set includes:

[0085] Acquire a medical entity set using a pre-constructed corpus, wherein the medical entity set includes a plurality of medical entities, and the medical entities are included in a pre-constructed spine entity set or a pre-constructed pelvic entity set;

[0086] Obtain a subject information set, extract subject information from the subject information set in sequence, and perform the following operations on the extracted subject information:

[0087] Obtaining an electronic medical record corresponding to the subject information, and obtaining a set of sentences to be identified based on the electronic medical record, wherein the set of sentences to be identified includes zero, one or more sentences to be identified;

[0088] If the sentence set to be identified is an empty set, the subject information is confirmed as suboptimal information;

[0089] If the set of sentences to be identified is not an empty set, extracting the sentences to be identified from the set of sentences to be identified in sequence, calculating the first similarity based on the extracted sentences to be identified and the medical entity set, summarizing the first similarities, and obtaining a first similarity set corresponding to the set of sentences to be identified;

[0090] Extracting a high similarity set from the first similarity set, wherein the high similarity set includes zero, one or more first similarities greater than a preset similarity threshold;

[0091] If the high similarity set is an empty set, the sentence set to be recognized and the medical entity set are introduced into the pre-built second language recognition model to obtain a second similarity set.

[0092] Extracting a second high similarity set from the second similarity set, if the second high similarity set is an empty set, confirming the extracted subject information as suboptimal information, otherwise, skipping the extracted subject information;

[0093] If the high similarity set includes one or more first similarities greater than the similarity threshold, skipping the extracted subject information;

[0094] Suboptimal information is aggregated to obtain a suboptimal information set, and a usable information set is obtained based on the suboptimal information set.

[0095] It is understandable that the corpus is a database including many medical entities, for example, the CCKS 2017 training corpus database, the CCKS 2019 training corpus database, etc., which are all prior art and will not be elaborated here. Medical entities are entities extracted from medical texts, such as diseases, body parts, drug information, etc. The spinal entity set is a set of medical entities related to spinal diseases, such as scoliosis, herniated disc, laminectomy, etc. The pelvic entity set is a set of medical entities related to pelvic diseases. Multiple medical entities in the embodiments of the present invention are included in the spinal entity set or the pelvic entity set.

[0096] Furthermore, the set of sentences to be recognized is a set of multiple sentences obtained by dividing the medical text in the electronic medical record. Optionally, the electronic medical record is divided using a syntactic analyzer. Since the electronic medical record has certain specifications when it is generated, the division of the medical text in the electronic medical record in the embodiment of the present invention is as follows: extracting the content between two punctuation marks to obtain sentences to be recognized. For example, if the text in the electronic medical record is: "Swelling of the left cheek, a small amount of fluid in the right pleural cavity", the text in the electronic medical record is divided into two sentences to be recognized: "Swelling of the left cheek" and "A small amount of fluid in the right pleural cavity", and the set of sentences to be recognized is a set consisting of the two sentences to be recognized.

[0097] It is understandable that the suboptimal information is the information for identifying subjects who have no history of spinal lesions and pelvic lesions in the electronic medical record. When the set of sentences to be identified is an empty set, it means that the subject has not undergone treatment in the hospital, and therefore it is considered that the subject has no history of spinal lesions and pelvic lesions. The first similarity is the degree of similarity between the extracted sentence to be identified and the medical entity set. Due to the characteristics of medical texts, the sentence to be identified is a short text. If the first similarity is high, it means that the subject may have a history of spinal lesions and pelvic lesions. Therefore, the present invention sets a similarity threshold and extracts a high similarity set from the first similarity set. The similarity threshold is the minimum value for considering that the sentence to be identified is similar to the medical entity set. When the first similarity or the second similarity is higher than the similarity threshold, it means that the sentence to be identified is similar to the medical entity set, thereby indicating that the electronic medical record includes the medical entity set, so the subject is a subject with a history of spinal lesions and pelvic lesions.

[0098] It should be noted that the first similarity set is a set consisting of multiple first similarities, and the high similarity set is a set consisting of first similarities greater than a similarity threshold.

[0099] Furthermore, if the high similarity set is an empty set, it means that none of the first similarities in the first similarity set is greater than the similarity threshold, and the subject may not have a history of spinal lesions and pelvic lesions. However, recognizing electronic medical records based on only one text recognition model may result in recognition errors, and therefore the sentence set to be recognized needs to be retested. The second language recognition model is a model that can recognize medical entities in medical texts. The second language recognition model integrates semantic features. Therefore, if the subject is still considered to have no history of spinal lesions and pelvic lesions after recognition by the second language recognition model, the subject information corresponding to the subject is confirmed as suboptimal information. Among them, the meaning of having no history of spinal lesions and pelvic lesions is that there is neither a history of spinal lesions nor a history of pelvic lesions.

[0100] It is understandable that the second similarity set is a set composed of multiple second similarities, the second similarity is the degree of similarity between the extracted sentence to be recognized and the medical entity set calculated using the second language recognition model, and the language model used by the second similarity is different from the language model used by the first similarity. The second high similarity set is a set constructed by the second similarity greater than the similarity threshold.

[0101] It should be noted that if the high similarity set includes one or more first similarities greater than the similarity threshold, it means that one or more first similarities in the first similarity set are greater than the similarity threshold. At this time, the embodiment of the present invention believes that the subject has a history of spinal diseases and pelvic diseases, and therefore needs to be skipped.

[0102] Furthermore, the calculation formula of the first similarity is as follows:

[0103] ;

[0104] in, represents the first similarity, represents a medical entity set, and , Represents the first Medical entities, Represents the first Medical entities, Represents the first Medical entities, Represents the first Medical entities, represents the sentence to be recognized, Indicates The serial number of the medical entity, Represents the total number of medical entities in the medical entity set, Indicates The weight of the medical entity in the medical entity set, Indicates the first The frequency of occurrence of medical entities, Indicates Semantic discrimination of characteristic words of medical entities, represents the first hyperparameter, represents the second hyperparameter, Indicates the length of the sentence to be recognized. Indicates The length of the medical entity in characters.

[0105] It should be noted that the medical entities in the medical entity set are not ranked. Medical entities, Medical entities, Medical Entity and The serial number in the medical entity is only used to distinguish multiple medical entities in the medical entity. In the embodiment of the present invention, each of the multiple medical entities has an equal weight in the medical entity set. The parameters of the semantic features of medical entities are integrated with the semantic features during the calculation of the first similarity. Therefore, when performing the first similarity calculation, it is not only judged according to whether the characters contained in the sentence to be identified are the same as the characters of the medical entity. For example, the sentence to be identified is: "Normal spine morphology", and "scoliosis" exists in the medical entity set. After integrating the semantic features, the first similarity between "normal spine morphology" and the medical entity set is less than the similarity threshold, and it is considered that "normal spine morphology" is not similar to the medical entity set. The semantic discrimination of the feature words used to describe the semantic features is a prior art and will not be repeated here.

[0106] It should be noted that the first hyperparameter and the second hyperparameter are both hyperparameters and will not be described in detail here.

[0107] Further, the obtaining of the available information set based on the suboptimal information set includes:

[0108] The following operations are performed on the suboptimal information in the suboptimal information set:

[0109] acquiring sacral nerve scanning data based on the suboptimal information, constructing a sacral nerve model based on the sacral nerve scanning data, and performing an image reconstruction operation on the sacral nerve model to obtain a plurality of sacral nerve image sets;

[0110] The following operations are performed on the sacral nerve image sets in the multiple sacral nerve image sets:

[0111] Sacral nerve images are extracted from the sacral nerve image set in sequence, and the following operations are performed on the extracted sacral nerve images:

[0112] Obtaining a clarity value calculation formula, calculating the clarity value of the extracted sacral nerve image based on the clarity value calculation formula, comparing the clarity value with a preset clarity threshold, and if the clarity value is less than the clarity threshold, confirming the extracted sacral nerve image as an unclear image, otherwise, skipping the extracted sacral nerve image;

[0113] Summarizing the unclear images to obtain unclear image sets of multiple sacral nerve image sets, and if the unclear image sets are empty sets, confirming suboptimal information corresponding to the multiple sacral nerve image sets as available information, otherwise, skipping the suboptimal information corresponding to the multiple sacral nerve image sets;

[0114] Summarize the available information to obtain the available information set.

[0115] It should be noted that the suboptimal information set is a collection of multiple suboptimal information. After the embodiment of the present invention screens out patients without a history of spinal lesions and pelvic lesions based on the electronic medical records, magnetic resonance imaging scans are performed on all subjects corresponding to the suboptimal information set to obtain sacral nerve scanning data. According to the principle of magnetic resonance imaging, the sacral nerve scanning data is a series of 2D images. Through the sacral nerve scanning data, a 3D sacral nerve model can be constructed. This technology is a prior art and will not be repeated here. In nuclear magnetic resonance imaging scanning technology, the signals between nerves and fat are different, and the signals corresponding to nerves are stronger than fat signals. Therefore, according to the sacral nerve scanning data, using maximum signal intensity projection and multi-plane reconstruction technology, sacral nerve images of different perspectives can be extracted from the sacral nerve model, thereby displaying the structure and shape of the sacral nerve image. Therefore, the technology used in the image reconstruction operation is maximum signal intensity projection and multi-plane reconstruction technology. Both maximum signal intensity projection and multi-plane reconstruction technology are prior art and will not be repeated here.

[0116] Furthermore, the clarity value is a numerical value used to describe whether the sacral nerve image is clear. In the embodiment of the present invention, the composition and structural characteristics of the patient's sacral nerve are measured by image intelligence. When the sacral nerve image is not clear, it is difficult to accurately determine the composition and structural characteristics of the sacral nerve through the image. Therefore, the suboptimal information of the unclear sacral nerve image needs to be eliminated.

[0117] Specifically, the clarity threshold is a preset minimum clarity value for considering the sacral nerve image to be clear. Optionally, before calculating the clarity value of the extracted sacral nerve image, multiple people select a clear sacral nerve image as an example image, calculate the clarity value of the example image using a clarity value calculation formula, and use the clarity value of the example image as the clarity threshold. Therefore, when the clarity value of the extracted sacral nerve image is less than the clarity threshold, it means that the extracted sacral nerve image is not clear. An unclear image is a sacral nerve image whose clarity value is lower than the clarity threshold. An unclear image set is a set of unclear images.

[0118] It can be understood that the available information is information about a subject who has no history of spinal lesions and pelvic lesions and whose sacral nerve images in multiple sacral nerve image sets are all clear images. The available information set is a collection of multiple available information.

[0119] Furthermore, the obtaining of the clarity value calculation formula, and calculating the clarity value of the extracted sacral nerve image based on the clarity value calculation formula, includes:

[0120] Constructing a horizontal convolution template and a vertical convolution template, and using a pre-constructed image coordinate system to obtain the horizontal convolution direction and the vertical convolution direction of the extracted sacral nerve image;

[0121] Using the horizontal convolution template to perform convolution calculation in the horizontal convolution direction to obtain multiple horizontal gradient values, and using the vertical convolution template to perform convolution calculation in the vertical convolution direction to obtain multiple vertical gradient values, wherein the horizontal gradient values ​​and the vertical gradient values ​​correspond to each other one by one;

[0122] The clarity value of the extracted sacral nerve image is calculated based on multiple horizontal gradient values, multiple vertical gradient values ​​and a clarity value calculation formula, wherein the clarity value calculation formula is as follows:

[0123] ;

[0124] in, represents the clarity value of the extracted sacral nerve image, Represents the horizontal coordinate of the pixel point in the sacral nerve image in the image coordinate system, Represents the ordinate of the pixel point in the sacral nerve image in the image coordinate system, Represents the available gradient value of a pixel in the sacral nerve image.

[0125] Furthermore, the calculation formula of the available gradient value is as follows:

[0126] ;

[0127] in, represents the gradient value of the pixel in the sacral nerve image, Indicates the available threshold value of the gradient value;

[0128] The calculation formula of the gradient value is as follows:

[0129] ;

[0130] ;

[0131] ;

[0132] in, Represents the gray value of the pixel in the sacral nerve image, represents the convolution operator, represents the horizontal convolution template, represents the vertical convolution template, Represents the horizontal gradient value of the pixel in the sacral nerve image, Represents the vertical gradient value of the pixel in the sacral nerve image.

[0133] It can be understood that the horizontal convolution template and the vertical convolution template are respectively a template for performing convolution calculation in the horizontal direction in the sacral nerve image and a template for performing convolution calculation in the vertical direction in the sacral nerve image.

[0134] Exemplarily, the horizontal convolution template and the vertical convolution template use the horizontal convolution template and the vertical convolution template corresponding to the Sobel operator clarity function, and the horizontal convolution template and the vertical convolution template are as follows:

[0135] ;

[0136] ;

[0137] in, represents the horizontal convolution template, Represents a vertical convolution template.

[0138] Furthermore, the image coordinate system is the image coordinate system of the extracted sacral nerve image, and the construction of the image coordinate system is a prior art and will not be described in detail here. The horizontal convolution direction is the direction in which the horizontal coordinate increases in the image coordinate system, and the vertical convolution direction is the direction in which the vertical coordinate increases in the image coordinate system. The horizontal gradient value is the gradient value obtained by convolving the pixels in the sacral nerve image using the horizontal convolution template, and the vertical gradient value is the gradient value obtained by convolving the pixels in the sacral nerve image using the vertical convolution template.

[0139] It should be noted that, when observing two images of the same content through human vision, one of which is blurred and the other is not, the clearer image has more edges and details in the image and a larger gradient value than the blurrier image. In other words, in two pictures of the same content, the gradient value in the blurrier image is smaller than the gradient value in the clearer image. The embodiment of the present invention sets an available threshold, and it is considered that only the gradient value exceeding the available threshold has an impact on the clarity value of the sacral nerve image. The available threshold is the minimum gradient value included in the clarity value calculation of the sacral nerve image. The embodiment of the present invention uses the available threshold to screen the gradient values ​​of the pixel points in the sacral nerve image, and retains the gradient values ​​greater than the available threshold.

[0140] S2. Extract available information from the available information set in sequence, and perform the following operations on the extracted available information: obtain multiple sacral nerve image sets corresponding to the available information, wherein the multiple sacral nerve image sets are: S1 nerve image set, S2 nerve image set, S3 nerve image set and S4 nerve image set.

[0141] It should be noted that the S1 neural image set is a set of images corresponding to multiple S1 nerves, the S2 neural image set is a set of images corresponding to multiple S2 nerves, the S3 neural image set is a set of images corresponding to multiple S3 nerves, and the S4 neural image set is a set of images corresponding to multiple S4 nerves.

[0142] Furthermore, the S1 nerve, S2 nerve, S3 nerve and S4 nerve are the first sacral nerve, the second sacral nerve, the third sacral nerve and the fourth sacral nerve, respectively, which are professional terms in medicine and will not be described herein.

[0143] It is understandable that when the S1 nerve image set is acquired, the S1 nerve image in the S1 nerve image set can display the external course angle and the front course angle of the S1 nerve. For example, the S1 nerve image set includes but is not limited to the coronal image of the S1 nerve of the subject, the sagittal image of the S1 nerve of the subject, etc. The S2 nerve image set, the S3 nerve image set, and the S4 nerve image set can all display the external course angle and the front course angle of the corresponding nerve like the S1 nerve image set, and will not be repeated here.

[0144] S3. Perform the following operations on all sacral nerve image sets in the multiple sacral nerve image sets: calculate a running data set in the sacral nerve image set based on the sacral nerve image set, and construct sacral nerve identification data corresponding to the sacral nerve image set using the running data set, wherein the running data set includes an outer running angle or a front running angle.

[0145] It should be noted that the lateral course angle is the angle between the sacral nerve behind the sacral ganglion and the long axis of the human body measured on the coronal sacral nerve image, and the anterior course angle is the angle between the sacral nerve behind the sacral ganglion and the long axis of the human body measured on the sagittal sacral nerve image.

[0146] Furthermore, the step of calculating the running data set in the sacral nerve image set based on the sacral nerve image set includes:

[0147] The identification name is confirmed based on the sacral nerve image set, sacral nerve images are extracted from the sacral nerve image set in sequence, and the following operations are performed on the extracted sacral nerve images:

[0148] Confirming the running direction based on the extracted sacral nerve image, extracting a running detection area image from the extracted sacral nerve image, performing a binarization operation on the running detection area image to obtain a binary image, acquiring a detection image coordinate system based on the binary image, and extracting all neural pixel points from the binary image to obtain a neural pixel point set, wherein the neural pixel point is a pixel point with a gray value of 255 in the binary image;

[0149] Using the detection image coordinate system and the pre-constructed straight line fitting method, the neural pixel point set is fitted into a straight line in the detection image coordinate system to obtain the sacral nerve straight line, and the sacral nerve straight line equation is obtained according to the sacral nerve straight line. The straight line slope is extracted from the sacral nerve straight line equation, and the absolute value of the extracted straight line slope is taken to obtain the running angle, and the running data is generated according to the identification name, running angle and running direction;

[0150] The running data are summarized to obtain the running data set corresponding to the sacral nerve image set.

[0151] It should be noted that the identification name is the name of the nerve corresponding to the sacral nerve image set, which is used to distinguish different sacral nerve image sets. For example, the sacral nerve image set is: If the sacral nerve image set is the S1 nerve image set, the corresponding identification name is: S1 nerve. If the sacral nerve image set is the S2 nerve image set, the corresponding identification name is: S2 nerve. And so on, it will not be repeated here. The running direction is the direction of the sacral nerve in the sacral nerve image. For example, the S1 nerve on the left runs forward or the S1 nerve on the left runs outward. When the extracted sacral nerve image is a sacral nerve image in the coronal position, the extracted sacral nerve image measures the external running angle, and thus the running direction confirmed from the extracted sacral nerve image is: left-outer running.

[0152] Specifically, the walking detection area image is an image extracted from the sacral nerve image and only includes the sacral nerve. In the extracted sacral nerve image, there are many other human structures, such as human organs, human bones, etc. The human organs are not obvious in the sacral nerve image. The nerve signal is stronger than the human organ signal, and the human bone signal is also stronger than the fat signal. Therefore, in the sacral nerve image, in addition to the sacral nerve, human bones can be clearly observed. The walking detection area image is an image directly captured from the sacral nerve area. The walking detection area image reduces the influence of the bones in the sacral nerve image on the measurement of the sacral nerve structure.

[0153] It is understood that the binarization operation is an operation of converting the grayscale values ​​corresponding to the pixels in the image into 0 or 255. This technology is prior art and will not be described in detail here. The binary image is an image obtained by binarizing the walking detection area image. In the embodiment of the present invention, the grayscale values ​​of the pixels corresponding to the sacral nerves in the walking detection area image are all converted to 255, and the grayscale values ​​of all pixels other than the pixels corresponding to the sacral nerves in the walking detection area image are all converted to 0.

[0154] Furthermore, the detection image coordinate system is an image coordinate system in a binary image. A neural pixel is a pixel with a grayscale value of 255 in a binary image, and represents a pixel of a sacral nerve in a binary image. A neural pixel set is a set of neural pixel points. A straight line fitting method is a method of fitting the coordinates corresponding to a plurality of discrete neural pixel points into a straight line. This method is a prior art, such as the least squares method, and will not be described in detail here. The sacral nerve straight line is a straight line fitted by the neural pixel set in the detection image coordinate system using a straight line fitting method, and the sacral nerve straight line equation is the straight line equation of the sacral nerve straight line in the detection image coordinate system. The running angle is the angle between the sacral nerve after the sacral ganglion measured on the sacral nerve image and the long axis of the human body. The running data is used to describe the data measured in the extracted sacral nerve image. For example, the running direction is: left side - outside running, the running angle is: 25.9°, and the identification name is: S1 nerve, then the running data is: S1-left-outside-25.9°. At this time, the running data includes the outside running angle. Since the S1 nerves are two symmetrical nerves on the left and right, there should be two outside running angles corresponding to S1. The other running data is: S1-right-outside-25.2°. At this time, the shape data includes the outside running angle.

[0155] It should be noted that the above-mentioned definitions of left and right are the medical definitions of left and right, for example, the left and right in the left chest cavity and the right chest cavity, which will not be repeated here.

[0156] Furthermore, the sacral nerve identification data is data used to identify the external course angle and the front course angle of the sacral nerve corresponding to the sacral nerve image set. For example, in the sacral nerve image set corresponding to the S1 nerve image set, through measurement, the two course data corresponding to the S1 nerve image set for indicating the external course angle are: S1-left-outer-25.9° and S1-right-outer-25.2°, and the S1 nerve image set also has two course data for indicating the front course angle, which are assumed to be: S1-left-front-6.9° and S1-right-front-6.6°, then the above four course data are merged to obtain the sacral nerve identification data: [S1-left-outer-25.9°, S1-right-outer-25.2°, S1-left-front-6.9°, S1-right-front-6.6°].

[0157] S4. Summarize the sacral nerve identification data to obtain a sacral nerve identification data set, parse the sacral nerve identification data set to obtain a sacral nerve report corresponding to the available information, import the sacral nerve report into the available information, and obtain updated information.

[0158] It is understandable that the sacral nerve identification data set is a collection of multiple sacral nerve identification data. It should be noted that one available information corresponds to one sacral nerve identification data set, and the sacral nerve identification data set includes the sacral nerve identification data corresponding to the S1 nerve of the subject of the available information, the sacral nerve identification data corresponding to the S2 nerve, the sacral nerve identification data corresponding to the S3 nerve, and the sacral nerve identification data corresponding to the S4 nerve. The updated information is the available information including the sacral nerve report.

[0159] It is understandable that the parsing of the sacral nerve identification data set to obtain a sacral nerve report corresponding to the available information includes:

[0160] Extract the left S1 nerve anterior running angle, the left S2 nerve anterior running angle, and the S3 nerve anterior running angle from the sacral nerve identification data set, and construct a left anterior running angle sequence based on the left S1 nerve anterior running angle, the left S2 nerve anterior running angle, and the S3 nerve anterior running angle sequence, and obtain a left anterior running angle report based on the left anterior running angle sequence;

[0161] Obtain the right anterior course angle report based on the sacral nerve marker dataset;

[0162] Extract the external running angle of the left S1 nerve, the external running angle of the left S2 nerve, the external running angle of the left S3 nerve, and the external running angle of the outer S4 nerve from the sacral nerve identification data set, and construct a left external running angle sequence based on the external running angle of the left S1 nerve, the external running angle of the left S2 nerve, the external running angle of the left S3 nerve, and the external running angle of the outer S4 nerve, and obtain a left external running angle report based on the left external running angle sequence;

[0163] Get the right external course angle report based on the sacral nerve identification dataset;

[0164] The left anterior running angle report, the right anterior running angle report, the left outer running angle report and the right outer running angle report are summarized to obtain the sacral nerve report corresponding to the available information.

[0165] Further, obtaining a left front running angle report based on the left front running angle sequence includes:

[0166] Extract the target angles from the left front running angle sequence in sequence, and perform the following operations on the extracted target angles:

[0167] Acquire an adjacent angle based on the target angle, wherein the adjacent angle is the next target angle in the left front running angle sequence adjacent to the target angle;

[0168] Compare the target angle and the adjacent angles. If the target angle is smaller than the adjacent angle, a normal report is generated based on the target angle and the adjacent angles. Otherwise, an abnormal report is generated.

[0169] Normal reports are summarized to obtain a normal report set, abnormal reports are summarized to obtain an abnormal report set, and a left front running angle report is obtained based on the normal report set and the abnormal report set.

[0170] It can be understood that the anterior running angle of the left S1 nerve, the anterior running angle of the left S2 nerve and the anterior running angle of the S3 nerve are the anterior running angles corresponding to the left S1 nerve, the left S2 nerve and the left S3 nerve respectively, and the left anterior running angle sequence is a sequence constructed by the anterior running angle of the left S1 nerve, the anterior running angle of the left S2 nerve and the anterior running angle of the S3 nerve, and the sequence order is: the anterior running angle of the left S1 nerve, the anterior running angle of the left S2 nerve and the anterior running angle of the S3 nerve. For example, the anterior running angle of the left S1 nerve is 6.9°, the anterior running angle of the left S2 nerve is 10.4°, and the anterior running angle of the S3 nerve is 15.1°, then the left anterior running angle sequence is: [6.9, 10.4, 15.1].

[0171] It should be noted that the target angle is the angle extracted from the left anterior running angle sequence, and when extracting the target angle from the left anterior running angle sequence, it is extracted in the order of the left S1 nerve anterior running angle, the left S2 nerve anterior running angle, and the S3 nerve anterior running angle. In the anterior running angle, if the subject is healthy, the S1 nerve anterior running angle, the left S2 nerve anterior running angle, and the S3 nerve anterior running angle increase in sequence. The left anterior sacral nerve abnormality report indicates that the extracted target angle is smaller than the adjacent angle, indicating that the subject may have sacral nerve abnormalities, such as: sacral nerve compression

[0172] For example, the left front running angle sequence is: [11.4, 10.4, 15.1], 11.4 is extracted as the target angle, and the adjacent angle is 10.4, then an abnormal report is generated, and the abnormal report content is: the left front running angle of the S1 nerve and the S2 nerve is abnormal, 10.4 is extracted as the target angle, 15.1 is the adjacent angle, then a normal report is generated, and the normal report content is: the left front running angle of the S2 nerve and the S3 nerve is normal. At this time, the left front running angle report includes an abnormal report and an abnormal report.

[0173] Furthermore, when there is no normal report in the left front running angle sequence, the normal report set is an empty set, and the abnormal report set is the same, which will not be repeated here.

[0174] It should be noted that the right front running angle report is a report used to describe whether the front running angles of the right S1 nerve, the right S2 nerve and the right S3 nerve are abnormal. The implementation method of the right front running angle report is the same as that of the left front running angle report, and can achieve the same effect, which will not be repeated here.

[0175] Further, the S1 nerve external running angle, the left S2 nerve external running angle, the left S3 nerve external running angle and the outer S4 nerve external running angle are the external running angles corresponding to the left S1 nerve, the left S2 nerve, the left S3 nerve and the left S4 nerve respectively. The left external running angle sequence is a sequence constructed by the S1 nerve external running angle, the left S2 nerve external running angle, the left S3 nerve external running angle and the outer S4 nerve external running angle.

[0176] It should be noted that in the left front running angle report and the right front running angle report, when the front running angle of the nerve increases in the order of S1 nerve, S2 nerve and S3 nerve, it is considered normal. In the left outer running angle report and the right outer running angle report, when the outer running angle of the nerve decreases in the order of S1 nerve, S2 nerve, S3 nerve and S4 nerve, it is considered normal. The left outer running angle report is a report used to describe whether the outer running angles of the left S1 nerve, the left S2 nerve, the left S3 nerve and the left S4 nerve are abnormal. The right outer running angle report is a report used to describe whether the outer running angles of the right S1 nerve, the right S2 nerve, the right S3 nerve and the right S4 nerve are abnormal. The implementation method of the left outer running angle report and the right outer running angle report is the same as that of the left front running angle report, and can achieve the same effect, which will not be repeated here.

[0177] In particular, the sacral nerve report is a report used to describe whether the sacral nerve of the subject corresponding to the available information is abnormal.

[0178] S5. Summarize the updated information to obtain an updated information set corresponding to the available information set, and complete intelligent analysis of the composition structure of the sacral nerve based on the updated information set.

[0179] It should be noted that the update information set is a collection of update information.

[0180] The present invention is to solve the problem described in the background technology. The present invention obtains a subject information set, screens the subject information set, obtains an available information set, and extracts available information from the available information set in sequence. The present invention uses a language model to screen the subject's electronic medical record, removes the subject information with a history of spinal lesions and a history of pelvic lesions, and also removes the information of the sacral nerve image with poor quality in the sacral nerve image set, aiming to extract the subject information without a history of spinal lesions and a history of pelvic lesions, and the sacral nerve images in multiple sacral nerve image sets are all clear images, so as to screen out the information that can meet the health detection needs of the subject by only performing feature detection of the sacral nerve composition structure. The present invention performs the following operations on the extracted available information: obtain multiple sacral nerve image sets corresponding to the available information, and perform the following operations on the sacral nerve image sets in the multiple sacral nerve image sets: calculate the running data set in the sacral nerve image set based on the sacral nerve image set, and use the running data set to construct the sacral nerve identification data corresponding to the sacral nerve image set. The sacral nerve identification data constructed by the image calculation method can describe the external running angle and the front running angle of the sacral nerve, thereby realizing the extraction of the composition structure data of the sacral nerve. Sacral nerve identification data are summarized to obtain a sacral nerve identification data set, and the sacral nerve identification data set is analyzed to obtain a sacral nerve report corresponding to the available information. The present invention uses the statistical significance of the anterior and lateral running angles between the existing different sacral nerves to perform intelligent analysis of the sacral nerve composition structure of the subjects with available information corresponding to the sacral nerve identification data set to determine whether the sacral nerve of the subject is abnormal. Therefore, the present invention can optimize the process of identifying the composition structure of the sacral nerve, accurately extract the composition structure data of the sacral nerve, and avoid wasting medical resources.

[0181] like Figure 2 1 is a functional module diagram of an intelligent analysis system for the composition structure of the sacral nerve provided by an embodiment of the present invention.

[0182] The intelligent analysis system 100 for the composition structure of the sacral nerves of the present invention can be installed in an electronic device. According to the functions to be implemented, the intelligent analysis system 100 for the composition structure of the sacral nerves can include an information screening module 101, a sacral nerve identification data acquisition module 102, an analysis module 103 and a summary module 104. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0183] The information screening module 101 is used to obtain a subject information set, screen the subject information set, and obtain a usable information set;

[0184] The sacral nerve identification data acquisition module 102 is used to extract available information from the available information set in sequence, and perform the following operations on the extracted available information: obtain multiple sacral nerve image sets corresponding to the available information, wherein the multiple sacral nerve image sets are respectively: S1 nerve image set, S2 nerve image set, S3 nerve image set and S4 nerve image set, and perform the following operations on the sacral nerve image sets in the multiple sacral nerve image sets: calculate the running data set in the sacral nerve image set based on the sacral nerve image set, and construct the sacral nerve identification data corresponding to the sacral nerve image set using the running data set, wherein the running data set includes an external running angle or a front running angle;

[0185] The parsing module 103 is used to summarize the sacral nerve identification data to obtain a sacral nerve identification data set, parse the sacral nerve identification data set to obtain a sacral nerve report corresponding to the available information, and import the sacral nerve report into the available information to obtain updated information;

[0186] The summarizing module 104 is used to summarize the updated information, obtain the updated information set corresponding to the available information set, and complete the intelligent analysis of the composition structure of the sacral nerve based on the updated information set.

[0187] In detail, the modules in the intelligent analysis system 100 for the composition structure of the sacral nerve in the embodiment of the present invention are used in the same manner as described above. Figure 1 The same technical means as the intelligent analysis method for the composition structure of the sacral nerve described in the text can produce the same technical effects, so I will not go into details here.

[0188] like Figure 3 FIG. 1 is a schematic diagram of the structure of an electronic device for implementing an intelligent analysis method for the composition structure of sacral nerves provided by an embodiment of the present invention.

[0189] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a program for an intelligent analysis method for the composition structure of the sacral nerve.

[0190] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 11 also includes an internal storage unit of the electronic device 1 and an external storage device. The memory 11 can not only be used to store application software and various types of data installed in the electronic device 1, such as the code of the program of the intelligent analysis method for the composition structure of the sacral nerve, but also can be used to temporarily store data that has been output or is to be output.

[0191] The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules stored in the memory 11 (for example, a program for intelligent analysis of the structure of the sacral nerve, etc.), and calls data stored in the memory 11, so as to execute various functions of the electronic device 1 and process data.

[0192] The bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0193] Figure 3 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 3The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0194] For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 10 through a power management system, so that the power management system can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0195] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0196] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0197] The program of the intelligent analysis method for the composition structure of the sacral nerve stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve:

[0198] Obtain a subject information set, filter the subject information set, obtain an available information set, extract available information from the available information set in sequence, and perform the following operations on the extracted available information:

[0199] A plurality of sacral nerve image sets corresponding to the available information are obtained, wherein the plurality of sacral nerve image sets are: an S1 nerve image set, an S2 nerve image set, an S3 nerve image set, and an S4 nerve image set, and the following operations are performed on the sacral nerve image sets in the plurality of sacral nerve image sets:

[0200] Calculating a running data set in the sacral nerve image set based on the sacral nerve image set, and constructing sacral nerve identification data corresponding to the sacral nerve image set using the running data set, wherein the running data set includes an outer running angle or a front running angle;

[0201] Summarizing the sacral nerve identification data to obtain a sacral nerve identification data set, parsing the sacral nerve identification data set to obtain a sacral nerve report corresponding to the available information, and importing the sacral nerve report into the available information to obtain updated information;

[0202] The updated information is summarized to obtain an updated information set corresponding to the available information set, and intelligent analysis of the composition structure of the sacral nerve is completed based on the updated information set.

[0203] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figures 1 to 3 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0204] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or system that can carry the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0205] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement:

[0206] Obtain a subject information set, filter the subject information set, obtain an available information set, extract available information from the available information set in sequence, and perform the following operations on the extracted available information:

[0207] A plurality of sacral nerve image sets corresponding to the available information are obtained, wherein the plurality of sacral nerve image sets are: an S1 nerve image set, an S2 nerve image set, an S3 nerve image set, and an S4 nerve image set, and the following operations are performed on the sacral nerve image sets in the plurality of sacral nerve image sets:

[0208] Calculating a running data set in the sacral nerve image set based on the sacral nerve image set, and constructing sacral nerve identification data corresponding to the sacral nerve image set using the running data set, wherein the running data set includes an outer running angle or a front running angle;

[0209] Summarizing the sacral nerve identification data to obtain a sacral nerve identification data set, parsing the sacral nerve identification data set to obtain a sacral nerve report corresponding to the available information, and importing the sacral nerve report into the available information to obtain updated information;

[0210] The updated information is summarized to obtain an updated information set corresponding to the available information set, and intelligent analysis of the composition structure of the sacral nerve is completed based on the updated information set.

[0211] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, and actual implementation may have other division methods.

[0212] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0213] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0214] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0215] Finally, it should be noted that the above embodiments 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 preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. An intelligent analysis method for the composition structure of sacral nerves, characterized in that: The method comprises: Obtaining a subject information set, screening the subject information set, and obtaining an available information set, wherein the subjects corresponding to the subject information in the available information set only retain subjects who have no history of spinal lesions and pelvic lesions; Extract available information from the available information set in sequence, and perform the following operations on the extracted available information: A plurality of sacral nerve image sets corresponding to the available information are obtained, wherein the plurality of sacral nerve image sets are respectively: an S1 nerve image set, an S2 nerve image set, an S3 nerve image set, and an S4 nerve image set, and the following operations are performed on the sacral nerve image sets in the plurality of sacral nerve image sets: Calculating a running data set in the sacral nerve image set based on the sacral nerve image set, and constructing sacral nerve identification data corresponding to the sacral nerve image set using the running data set, wherein the running data set includes an outer running angle or a front running angle; Summarizing the sacral nerve identification data to obtain a sacral nerve identification data set, parsing the sacral nerve identification data set to obtain a sacral nerve report corresponding to the available information, and importing the sacral nerve report into the available information to obtain updated information; Summarize the updated information to obtain the updated information set corresponding to the available information set, and complete the intelligent analysis of the composition structure of the sacral nerve based on the updated information set; Wherein, the obtaining of the subject information set, screening the subject information set, and obtaining the available information set includes: Acquire a medical entity set using a pre-constructed corpus, wherein the medical entity set includes a plurality of medical entities, and the medical entities are included in a pre-constructed spine entity set or a pre-constructed pelvic entity set; Obtain a subject information set, extract subject information from the subject information set in sequence, and perform the following operations on the extracted subject information: Obtaining an electronic medical record corresponding to the subject information, and obtaining a set of sentences to be identified based on the electronic medical record, wherein the set of sentences to be identified includes zero, one or more sentences to be identified; If the sentence set to be identified is an empty set, the subject information is confirmed as suboptimal information; If the set of sentences to be identified is not an empty set, extracting the sentences to be identified from the set of sentences to be identified in sequence, calculating the first similarity based on the extracted sentences to be identified and the medical entity set, summarizing the first similarities, and obtaining a first similarity set corresponding to the set of sentences to be identified; Extracting a high similarity set from the first similarity set, wherein the high similarity set includes zero, one or more first similarities greater than a preset similarity threshold; If the high similarity set is an empty set, the sentence set to be recognized and the medical entity set are introduced into the pre-built second language recognition model to obtain a second similarity set. Extracting a second high similarity set from the second similarity set, if the second high similarity set is an empty set, confirming the extracted subject information as suboptimal information, otherwise, skipping the extracted subject information; If the high similarity set includes one or more first similarities greater than the similarity threshold, skipping the extracted subject information; Summarize suboptimal information to obtain a suboptimal information set, and obtain a usable information set based on the suboptimal information set; The step of parsing the sacral nerve identification data set to obtain a sacral nerve report corresponding to the available information includes: Extract the left S1 nerve anterior running angle, the left S2 nerve anterior running angle, and the S3 nerve anterior running angle from the sacral nerve identification data set, and construct a left anterior running angle sequence based on the left S1 nerve anterior running angle, the left S2 nerve anterior running angle, and the S3 nerve anterior running angle sequence, and obtain a left anterior running angle report based on the left anterior running angle sequence; Obtain the right anterior course angle report based on the sacral nerve marker dataset; Extract the external running angle of the left S1 nerve, the external running angle of the left S2 nerve, the external running angle of the left S3 nerve, and the external running angle of the outer S4 nerve from the sacral nerve identification data set, and construct a left external running angle sequence based on the external running angle of the left S1 nerve, the external running angle of the left S2 nerve, the external running angle of the left S3 nerve, and the external running angle of the outer S4 nerve, and obtain a left external running angle report based on the left external running angle sequence; Get the right external course angle report based on the sacral nerve identification dataset; Summarize the left front running angle report, the right front running angle report, the left outer running angle report and the right outer running angle report to obtain the sacral nerve report corresponding to the available information; Among them, the left anterior running angle sequence is a sequence constructed by the left S1 nerve anterior running angle, the left S2 nerve anterior running angle and the S3 nerve anterior running angle, and the sequence order is: the left S1 nerve anterior running angle, the left S2 nerve anterior running angle and the S3 nerve anterior running angle; The step of obtaining a left front running angle report based on a left front running angle sequence includes: Extract the target angles from the left front running angle sequence in sequence, and perform the following operations on the extracted target angles: Acquire an adjacent angle based on the target angle, wherein the adjacent angle is the next target angle in the left front running angle sequence adjacent to the target angle; Compare the target angle and the adjacent angles. If the target angle is smaller than the adjacent angle, a normal report is generated based on the target angle and the adjacent angles. Otherwise, an abnormal report is generated. Normal reports are summarized to obtain a normal report set, abnormal reports are summarized to obtain an abnormal report set, and a left front running angle report is obtained based on the normal report set and the abnormal report set.

2. The intelligent analysis method for the composition structure of the sacral nerve according to claim 1, characterized in that: The calculation formula of the first similarity is as follows: ； in, represents the first similarity, represents a medical entity set, and , Represents the first Medical entities, Represents the first Medical entities, Represents the first Medical entities, Represents the first Medical entities, represents the sentence to be recognized, Indicates The serial number of the medical entity, Represents the total number of medical entities in the medical entity set, Indicates The weight of the medical entity in the medical entity set, Indicates the first The frequency of occurrence of medical entities, Indicates Semantic discrimination of characteristic words of medical entities, represents the first hyperparameter, represents the second hyperparameter, Indicates the length of the sentence to be recognized. Indicates The length of the medical entity in characters.

3. The intelligent analysis method for the composition structure of the sacral nerve according to claim 2, characterized in that: The obtaining of the available information set based on the suboptimal information set includes: The following operations are performed on the suboptimal information in the suboptimal information set: acquiring sacral nerve scanning data based on the suboptimal information, constructing a sacral nerve model based on the sacral nerve scanning data, and performing an image reconstruction operation on the sacral nerve model to obtain a plurality of sacral nerve image sets; The following operations are performed on the sacral nerve image sets in the multiple sacral nerve image sets: Sacral nerve images are extracted from the sacral nerve image set in sequence, and the following operations are performed on the extracted sacral nerve images: Obtaining a clarity value calculation formula, calculating the clarity value of the extracted sacral nerve image based on the clarity value calculation formula, comparing the clarity value with a preset clarity threshold, and if the clarity value is less than the clarity threshold, confirming the extracted sacral nerve image as an unclear image, otherwise, skipping the extracted sacral nerve image; Summarizing the unclear images to obtain unclear image sets of multiple sacral nerve image sets, and if the unclear image sets are empty sets, confirming suboptimal information corresponding to the multiple sacral nerve image sets as available information, otherwise, skipping the suboptimal information corresponding to the multiple sacral nerve image sets; Summarize the available information to obtain the available information set.

4. The intelligent analysis method for the composition structure of the sacral nerve according to claim 3, characterized in that: The obtaining of the clarity value calculation formula, and calculating the clarity value of the extracted sacral nerve image based on the clarity value calculation formula, includes: Constructing a horizontal convolution template and a vertical convolution template, and using a pre-constructed image coordinate system to obtain the horizontal convolution direction and the vertical convolution direction of the extracted sacral nerve image; Using the horizontal convolution template to perform convolution calculation in the horizontal convolution direction to obtain multiple horizontal gradient values, and using the vertical convolution template to perform convolution calculation in the vertical convolution direction to obtain multiple vertical gradient values, wherein the horizontal gradient values ​​and the vertical gradient values ​​correspond to each other one by one; The clarity value of the extracted sacral nerve image is calculated based on multiple horizontal gradient values, multiple vertical gradient values ​​and a clarity value calculation formula, wherein the clarity value calculation formula is as follows: ； in, represents the clarity value of the extracted sacral nerve image, Represents the horizontal coordinate of the pixel point in the sacral nerve image in the image coordinate system, Represents the ordinate of the pixel point in the sacral nerve image in the image coordinate system, Represents the available gradient value of a pixel in the sacral nerve image.

5. The intelligent analysis method for the composition structure of the sacral nerve according to claim 4, characterized in that: The calculation formula of the available gradient value is as follows: ； in, represents the gradient value of the pixel in the sacral nerve image, Indicates the available threshold value of the gradient value; The calculation formula of the gradient value is as follows: ； ； ； in, Represents the grayscale value of the pixel in the sacral nerve image, represents the convolution operator, represents the horizontal convolution template, represents the vertical convolution template, Represents the horizontal gradient value of the pixel in the sacral nerve image, Represents the vertical gradient value of the pixel in the sacral nerve image.

6. The intelligent analysis method for the composition structure of the sacral nerve according to claim 5, characterized in that: The step of calculating the running data set in the sacral nerve image set based on the sacral nerve image set includes: The identification name is confirmed based on the sacral nerve image set, sacral nerve images are extracted from the sacral nerve image set in sequence, and the following operations are performed on the extracted sacral nerve images: Confirming the running direction based on the extracted sacral nerve image, extracting a running detection area image from the extracted sacral nerve image, performing a binarization operation on the running detection area image to obtain a binary image, acquiring a detection image coordinate system based on the binary image, and extracting all neural pixel points from the binary image to obtain a neural pixel point set, wherein the neural pixel point is a pixel point with a gray value of 255 in the binary image; Using the detection image coordinate system and the pre-constructed straight line fitting method, the neural pixel point set is fitted into a straight line in the detection image coordinate system to obtain the sacral nerve straight line, and the sacral nerve straight line equation is obtained according to the sacral nerve straight line. The straight line slope is extracted from the sacral nerve straight line equation, and the absolute value of the extracted straight line slope is taken to obtain the running angle, and the running data is generated according to the identification name, running angle and running direction; The running data are summarized to obtain the running data set corresponding to the sacral nerve image set.

7. An intelligent analysis system for the composition structure of sacral nerves, applied to an intelligent analysis method for the composition structure of sacral nerves as claimed in any one of claims 1 to 6, characterized in that: The system comprises: An information screening module, used to obtain a subject information set, screen the subject information set, and obtain a usable information set; The sacral nerve identification data acquisition module is used to extract available information from the available information set in sequence, and perform the following operations on the extracted available information: obtain multiple sacral nerve image sets corresponding to the available information, wherein the multiple sacral nerve image sets are respectively: S1 nerve image set, S2 nerve image set, S3 nerve image set and S4 nerve image set, and perform the following operations on the sacral nerve image sets in the multiple sacral nerve image sets: calculate the running data set in the sacral nerve image set based on the sacral nerve image set, and construct the sacral nerve identification data corresponding to the sacral nerve image set using the running data set, wherein the running data set includes an external running angle or a front running angle; A parsing module, configured to aggregate sacral nerve identification data to obtain a sacral nerve identification data set, parse the sacral nerve identification data set to obtain a sacral nerve report corresponding to available information, and import the sacral nerve report into the available information to obtain updated information; The summarizing module is used to summarize the updated information, obtain the updated information set corresponding to the available information set, and complete the intelligent analysis of the composition structure of the sacral nerve based on the updated information set.

Images