Ultrasonic anesthesia positioning method and system based on AI preoperative evaluation and storage medium

Abstract

The invention discloses an ultrasonic anesthesia positioning method and system based on AI preoperative evaluation and a storage medium, and relates to the technical field of medical assistance. The method comprises the following steps: comprehensively collecting body health data of a patient, including a physical examination result, a laboratory examination result and a medical history, reasoning the body health data of the patient based on an operation risk knowledge graph model, and determining an operation risk level; if the operation risk level is higher than a preset risk level value, the operation risk is evaluated again in combination with past operation case information, and if the operation condition is met, the crowd is judged to be the high-risk crowd of the operation; an ultrasonic image of an in-vivo structure of an operation high-risk crowd is collected by utilizing an ultrasonic technology, the ultrasonic image is recognized by utilizing an image recognition AI algorithm, the relationship between an anatomical position and an adjacent part is determined, and then an anesthesia position is determined. The medical quality and the service level can be improved, and safer and more efficient surgical treatment experience can be brought to patients.

Description

Technical Field

[0001] The present invention relates to the field of medical assistance technology, and more specifically to an ultrasonic anesthesia positioning method, system and storage medium based on AI preoperative evaluation. Background Art

[0002] Brain science and artificial intelligence (AI) are the commanding heights of the current human science and technology development strategy. The two are intertwined and integrated. In recent years, the deep neural network (DNN) and machine learning (ML) algorithms derived from neurobiology have risen rapidly, greatly promoting the development of AI algorithms; accordingly, AI helps to make deep breakthroughs in brain science research. Through biological principles, mathematical modeling of structures such as neurons and synapses can be achieved. In the future, intelligent algorithm systems will be able to present the intelligent structure of the human brain and realize the simulation of the biological brain. AI belongs to computer science and is defined as a general term for any computer program with human intelligence. It gives computers the ability to perform text recognition, state reasoning and decision-making, and follows specific rules and algorithms to perform human intelligence tasks. ML is a technical discipline that uses computers to predict and analyze data by building probabilistic statistical models to achieve AI. At present, AI and ML have penetrated into many fields of medical health, and their application in the field of anesthesiology is also booming, including but not limited to airway management, ultrasound-assisted diagnosis, intelligent drug infusion system, intraoperative precise monitoring and early warning, perioperative complications and mortality prediction, intensive care treatment, etc., which can change clinical anesthesia practice, optimize treatment processes and improve patient prognosis. Mastering AI technology is essential to provide safe, efficient and more cost-effective clinical anesthesia. However, the current research on AI is still in its early stages, and how to achieve preoperative anesthesia evaluation still requires further research by technical personnel in this field. Summary of the invention

[0003] In view of this, the present invention provides an ultrasonic anesthesia positioning method, system and storage medium based on AI preoperative evaluation, which helps to improve the quality of medical care and service level, and can also bring patients a safer and more efficient surgical treatment experience.

[0004] In order to achieve the above object, the present invention adopts the following technical solution:

[0005] An ultrasonic anesthesia positioning method based on AI preoperative evaluation includes the following steps:

[0006] Comprehensively collect the patient's physical health data, including physical examination results, laboratory test results, and medical history, and infer the patient's physical health data based on the surgical risk knowledge graph model to determine the surgical risk level;

[0007] If the surgical risk level is higher than the preset risk level, the surgical risk will be re-evaluated in combination with past surgical case information. If the surgical conditions are met, the patient will be identified as a high-risk surgical population;

[0008] Ultrasound technology is used to collect ultrasound images of the internal structures of people at high risk of surgery, and image recognition AI algorithms are used to identify the ultrasound images, determine the relationship between the anatomical position and adjacent parts, and then determine the anesthesia position.

[0009] Optionally, the surgical risk knowledge graph model works as follows:

[0010] Construct an initial knowledge graph based on the patient's health data and all the information about previous successful surgeries, convert the knowledge graph into a graph structure, and predict the associations between entities in the graph structure;

[0011] Based on the geometric Brownian motion model, the data transfer relationship between entities in the graph structure is calculated, and the data transfer relationship is defined as a logical relationship;

[0012] The K-means clustering algorithm is used in combination with the semantic similarity between entities to cluster the logical relationships between entities to obtain clusters. Based on the clusters and the patient's health data in the knowledge graph, the missing data in the knowledge graph is inferred. The missing data is the surgical risk level.

[0013] Optionally, the K-means clustering algorithm is used in combination with the semantic similarity between entities to cluster the logical relationships between entities to obtain cluster clusters, specifically: determine any node in the logical relationship as the core point, use the K-means algorithm to cluster the nodes in the logical relationship to expand the core point, extract all points within the core point domain radius, and summarize all points within the domain radius that are densely connected to the core point into the first cluster cluster; traverse all logical relationship nodes until all clustering is completed to obtain the final cluster cluster.

[0014] Optionally, it also includes optimization of logical relationships, as follows: obtaining a set of data transfer relationships, the data transfer relationship includes multiple rule statements, and the rule statements are matched one by one with preset matching rules, and the rule statements are used to describe the corresponding matching rules; for every two rule statements, determining whether there is a logical inclusion relationship between the two matching rules corresponding to the two rule statements; if it is an inclusion relationship, then removing the rule statements corresponding to the included matching rules from the rule set to obtain an updated logical relationship.

[0015] Optionally, an image recognition AI algorithm is used to identify the image and determine the relationship between the anatomical position and adjacent parts. The specific steps are as follows:

[0016] The ultrasound image is input into a convolutional neural network for feature extraction to obtain a feature map, and three-dimensional reconstruction is performed based on the feature map to obtain a three-dimensional reconstruction model of the anatomical position and adjacent parts;

[0017] The vector relationship between the anatomical position and the adjacent parts in the three-dimensional reconstructed model is calculated to determine the relationship between the anatomical position and the adjacent parts.

[0018] Optionally, the anesthesia position may be changed according to the ultrasound image display in different parts to ensure the optimal anesthesia plan.

[0019] An ultrasonic anesthesia positioning system based on AI preoperative evaluation, comprising:

[0020] Data collection and risk level determination module: used to comprehensively collect patients' health data, including physical examination results, laboratory test results and medical history, and to infer patients' health data based on the surgical risk knowledge graph model to determine the surgical risk level;

[0021] Surgical risk reassessment module: If the surgical risk level is higher than the preset risk level value, it is used to reassess the surgical risk in combination with past surgical case information. If the surgical conditions are met, the patient is determined to be a high-risk surgical population;

[0022] Anesthesia position determination module: It is used to use ultrasound technology to collect images of the internal structures of people at high risk of surgery, and use image recognition AI algorithms to recognize the images, determine the relationship between the anatomical position and adjacent parts, and then determine the anesthesia position.

[0023] A computer storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of any one of the ultrasonic anesthesia positioning methods based on AI preoperative evaluation are implemented.

[0024] It can be seen from the above technical solutions that, compared with the prior art, the present invention provides an ultrasonic anesthesia positioning method, system and storage medium based on AI preoperative evaluation, which have the following beneficial effects:

[0025] 1. By comprehensively collecting the patient's physical health data and using the surgical risk knowledge graph model for reasoning, the surgical risk level can be assessed more accurately. This invention is more systematic and comprehensive than traditional methods, and helps to detect potential risks in advance so that preventive measures can be taken. After determining the high-risk population for surgery, the anesthesia location can be accurately determined by combining ultrasound technology and image recognition AI algorithms. It fully considers the individual differences and surgical needs of patients, helps to develop a more personalized anesthesia plan, and reduces anesthesia-related complications.

[0026] 2. Using ultrasound technology to collect ultrasound images of the internal structures of people at high risk of surgery allows doctors to intuitively see the anatomical structure and pathological conditions of the patient. This visualization technology greatly improves the accuracy and reliability of anesthesia positioning. Automatic analysis and recognition of ultrasound images through image recognition AI algorithms can quickly determine the relationship between anatomical positions and adjacent parts, avoiding errors and subjectivity in human judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0028] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0029] Figure 2 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] The embodiment of the present invention discloses an ultrasonic anesthesia positioning method based on AI preoperative evaluation, such as Figure 1 As shown, the following steps are included:

[0032] Step 1: Comprehensively collect the patient's health data, including physical examination results, laboratory test results, and medical history, and infer the patient's health data based on the surgical risk knowledge graph model to determine the surgical risk level;

[0033] Among them, the patient's systemic data includes but is not limited to medical history (previous diseases, surgical history, drug allergy history, etc.), physical examination results (weight, height, BMI, cardiopulmonary function assessment, etc.) and laboratory test results (blood tests, urine analysis, imaging reports, etc.).

[0034] Step 2: If the surgical risk level is higher than the preset risk level, the surgical risk is re-evaluated in combination with past surgical case information. If the surgical conditions are met, the patient is identified as a high-risk surgical population;

[0035] Step 3: Use ultrasound technology to collect ultrasound images of the internal structures of people at high risk of surgery, use image recognition AI algorithms to identify the ultrasound images, determine the relationship between the anatomical position and adjacent parts, and then determine the anesthesia position.

[0036] Furthermore, in step 1, the working principle of the surgical risk knowledge graph model is as follows:

[0037] S11: Build an initial knowledge graph based on the patient's health data and all the information of previous successful surgical cases, convert the knowledge graph into a graph structure, and predict the association between entities in the graph structure;

[0038] S12: Based on the geometric Brownian motion model, the data transfer relationship between entities in the graph structure is calculated, and the data transfer relationship is defined as a logical relationship;

[0039] S13: K-means clustering algorithm is used in combination with semantic similarity between entities to cluster the logical relationships between entities to obtain clusters. Based on the clusters and the patient health data in the knowledge graph, the missing data in the knowledge graph is inferred. The missing data is the surgical risk level.

[0040] Furthermore, in S13, the K-means clustering algorithm is used in combination with the semantic similarity between entities to cluster the logical relationships between entities to obtain cluster clusters, specifically: any node in the logical relationship is determined as the core point, the nodes in the logical relationship are clustered using the K-means algorithm to expand the core point, all points within the core point domain radius are extracted, and all points within the domain radius that are densely connected to the core point are summarized as the first cluster cluster; all logical relationship nodes are traversed until all clustering is completed to obtain the final cluster cluster.

[0041] This embodiment also includes optimization of logical relationships, as follows: obtaining a data transfer relationship set, the data transfer relationship includes multiple rule statements, and the rule statements are matched one by one with preset matching rules, and the rule statements are used to describe the corresponding matching rules; for every two rule statements, determining whether there is a logical inclusion relationship between the two matching rules corresponding to the two rule statements; if it is an inclusion relationship, then removing the rule statements corresponding to the included matching rules from the rule set to obtain an updated logical relationship.

[0042] Furthermore, in step 2, the doctor needs to review the data of similar surgical cases, especially those with high surgical risks but ultimately completed successfully. These cases provide doctors with valuable experience and reference. Based on collecting patient information and referring to past cases, the doctor needs to comprehensively assess the patient's surgical risk. The assessment includes but is not limited to the patient's physical condition, surgical difficulty, possible complications and preventive measures. If after a comprehensive assessment, the patient is deemed to meet the surgical conditions but has a high surgical risk, he or she will be identified as a high-risk surgical population. This determination result will provide an important reference for subsequent surgical preparation and anesthesia plan formulation.

[0043] For people at high risk of surgery, doctors need to develop personalized surgical plans, including choosing the appropriate surgical method, anesthesia method, timing of surgery, etc. At the same time, a detailed surgical emergency plan needs to be developed to deal with various situations that may arise. For people at high risk of surgery, preoperative preparation is particularly important. Doctors need to communicate fully with patients, inform them of the risks of surgery and possible complications, and obtain the understanding and consent of patients. At the same time, it is necessary to strengthen preoperative examinations and monitoring of patients to ensure that patients are in good physical condition and can tolerate surgery. During the operation, doctors need to strengthen monitoring and management of patients to ensure that the operation goes smoothly. Close follow-up and observation are also required after surgery to promptly detect and deal with possible complications and adverse reactions.

[0044] Furthermore, in step three, an image recognition AI algorithm is used to recognize the image and determine the relationship between the anatomical position and the adjacent parts. The specific steps are as follows:

[0045] S31: inputting the ultrasound image into a convolutional neural network to extract features, obtain a feature map, perform three-dimensional reconstruction based on the feature map, and obtain a three-dimensional reconstruction model of the anatomical position and adjacent parts;

[0046] S32: Calculate the vector relationship between the anatomical position and the adjacent parts in the three-dimensional reconstructed model to determine the relationship between the anatomical position and the adjacent parts.

[0047] Finally, this embodiment also discloses changing the anesthesia position according to the imaging conditions of the ultrasound image at different parts to ensure the optimal anesthesia plan, as follows:

[0048] Nerve block anesthesia: For example, for brachial plexus block anesthesia, ultrasound images can show the direction and distribution of the brachial plexus nerves. The anesthesiologist can adjust the position and angle of the anesthetic needle based on the image to ensure that the anesthetic drug accurately acts on the brachial plexus nerves. For lower limb nerve block anesthesia such as sciatic nerve block anesthesia, ultrasound images can also be used to locate the nerves and adjust the anesthesia position.

[0049] Spinal anesthesia: Ultrasound images can be used to display the anatomical structure of the spine and the location of the epidural space. The anesthesiologist can select the appropriate puncture point and needle insertion depth based on the images to ensure that the anesthetic is accurately injected into the epidural space.

[0050] General anesthesia: Although general anesthesia does not directly rely on ultrasound images to locate the anesthesia position, ultrasound images can be used to assess the patient's respiratory and circulatory function, providing an important reference for the safe implementation of general anesthesia.

[0051] and Figure 1 Corresponding to the method shown, the present invention also discloses an ultrasonic anesthesia positioning system based on AI preoperative evaluation for Figure 1 The implementation of the method, the specific structure is as follows Figure 2 As shown, including:

[0052] Data collection and risk level determination module: used to comprehensively collect patients' health data, including physical examination results, laboratory test results and medical history, and to infer patients' health data based on the surgical risk knowledge graph model to determine the surgical risk level;

[0053] Surgical risk reassessment module: If the surgical risk level is higher than the preset risk level value, it is used to reassess the surgical risk in combination with past surgical case information. If the surgical conditions are met, the patient is determined to be a high-risk surgical population;

[0054] Anesthesia position determination module: It is used to use ultrasound technology to collect images of the internal structures of people at high risk of surgery, and use image recognition AI algorithms to recognize the images, determine the relationship between the anatomical position and adjacent parts, and then determine the anesthesia position.

[0055] Finally, this embodiment discloses a computer storage medium, on which a computer program is stored. When the computer program is executed by a processor, any one of the steps of the ultrasonic anesthesia positioning method based on AI preoperative evaluation is implemented.

[0056] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0057] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An ultrasonic anesthesia positioning method based on AI preoperative evaluation, characterized in that: The following steps are involved: Comprehensively collect the patient's physical health data, including physical examination results, laboratory test results, and medical history, and infer the patient's physical health data based on the surgical risk knowledge graph model to determine the surgical risk level; If the surgical risk level is higher than the preset risk level, the surgical risk will be re-evaluated in combination with past surgical case information. If the surgical conditions are met, the patient will be identified as a high-risk group for surgery. Ultrasound technology is used to collect ultrasound images of the internal structures of people at high risk of surgery, and image recognition AI algorithms are used to identify the ultrasound images, determine the relationship between the anatomical position and adjacent parts, and then determine the anesthesia position.

2. The ultrasonic anesthesia positioning method based on AI preoperative evaluation according to claim 1, characterized in that: The working principle of the surgical risk knowledge graph model is as follows: Construct an initial knowledge graph based on the patient's health data and all the information about previous successful surgeries, convert the knowledge graph into a graph structure, and predict the associations between entities in the graph structure; Based on the geometric Brownian motion model, the data transfer relationship between entities in the graph structure is calculated, and the data transfer relationship is defined as a logical relationship; The K-means clustering algorithm is used in combination with the semantic similarity between entities to cluster the logical relationships between entities to obtain clusters. Based on the clusters and the patient's health data in the knowledge graph, the missing data in the knowledge graph is inferred. The missing data is the surgical risk level.

3. The ultrasonic anesthesia positioning method based on AI preoperative evaluation according to claim 2 is characterized in that: The K-means clustering algorithm is used in combination with the semantic similarity between entities to cluster the logical relationships between entities to obtain cluster clusters. Specifically, any node in the logical relationship is determined as the core point, and the nodes in the logical relationship are clustered using the K-means algorithm to expand the core point, extract all points within the core point domain radius, and summarize all points within the domain radius that are densely connected to the core point into the first cluster cluster; traverse all logical relationship nodes until all clustering is completed to obtain the final cluster cluster.

4. The ultrasonic anesthesia positioning method based on AI preoperative evaluation according to claim 2 is characterized in that: It also includes the optimization of logical relationships, as follows: obtaining a data transfer relationship set, the data transfer relationship includes multiple rule statements, and the rule statements are matched one by one with preset matching rules, and the rule statements are used to describe the corresponding matching rules; for every two rule statements, determining whether there is a logical inclusion relationship between the two matching rules corresponding to the two rule statements; if it is an inclusion relationship, then removing the rule statements corresponding to the included matching rules from the rule set to obtain an updated logical relationship.

5. The ultrasonic anesthesia positioning method based on AI preoperative evaluation according to claim 1, characterized in that: Image recognition AI algorithm is used to identify the image and determine the relationship between the anatomical position and adjacent parts. The specific steps are as follows: The ultrasound image is input into a convolutional neural network for feature extraction to obtain a feature map, and three-dimensional reconstruction is performed based on the feature map to obtain a three-dimensional reconstruction model of the anatomical position and adjacent parts; The vector relationship between the anatomical position and the adjacent parts in the three-dimensional reconstructed model is calculated to determine the relationship between the anatomical position and the adjacent parts.

6. The ultrasonic anesthesia positioning method based on AI preoperative evaluation according to claim 1, characterized in that: It also includes changing the anesthesia position according to the ultrasound image in different parts of the body to ensure the optimal anesthesia plan.

7. An ultrasonic anesthesia positioning system based on AI preoperative evaluation, characterized in that: include: Data collection and risk level determination module: used to comprehensively collect patients' health data, including physical examination results, laboratory test results and medical history, and to infer patients' health data based on the surgical risk knowledge graph model to determine the surgical risk level; Surgical risk reassessment module: If the surgical risk level is higher than the preset risk level value, it is used to reassess the surgical risk in combination with past surgical case information. If the surgical conditions are met, the patient is determined to be a high-risk surgical population; Anesthesia position determination module: It is used to use ultrasound technology to collect images of the internal structures of people at high risk of surgery, and use image recognition AI algorithms to recognize the images, determine the relationship between the anatomical position and adjacent parts, and then determine the anesthesia position.

8. A computer storage medium, characterized in that The computer storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the ultrasonic anesthesia positioning method based on AI preoperative evaluation as described in any one of claims 1 to 6 are implemented.

Images