Medical scheme generation method and device, electronic equipment and storage medium

Abstract

The invention discloses a medical scheme generation method and device, electronic equipment and a storage medium. Determining a first statement, and performing statement analysis on the first statement to obtain first information; generating first diagnosis information according to the first information and the first image; and obtaining a second statement, and correcting the first diagnosis information according to the second statement to obtain second diagnosis information. The method comprises the following steps: analyzing a first statement to obtain first information; the first diagnosis information is generated according to the first information and the first image, the obtained first diagnosis information is corrected through the second statement, the second diagnosis information is obtained, the diagnosis information can be corrected in combination with multi-source data information, omission of key information can be avoided through the bidirectional feedback determination process, and meanwhile the diagnosis efficiency is improved. And the diagnosis information can be subjected to bidirectional correction, so that the accuracy of the second diagnosis information is improved.

Description

Technical Field

[0001] This invention relates to the field of medical technology, and in particular to a method, apparatus, electronic device, and storage medium for generating medical solutions. Background Technology

[0002] In the current field of medical diagnosis, AI technology has demonstrated certain advantages in image analysis and data processing, but problems such as poor collaboration with doctors and fragmented decision-making still exist, specifically: AI and doctor decision-making are disconnected: Existing AI systems mostly output analysis results independently (such as image lesion identification), which doctors only use as a reference. The lack of dynamic interaction and decision-making integration between the two can easily lead to misinterpretation or neglect of AI conclusions. Lack of intent understanding and feedback: AI cannot accurately understand the doctor's clinical thinking and implicit needs, and doctors also find it difficult to naturally correct the AI's analysis direction, forming a one-way process of "AI output—doctor's judgment." Ambiguous role division: The core responsibilities of AI and doctors in diagnosis are not clearly defined (e.g., AI focuses on quantitative image analysis, while doctors are responsible for integrating clinical background), leading to repetitive work or the omission of key information. Insufficient multimodal information fusion: Multi-source information such as image data, clinical history, and laboratory tests are scattered throughout the decision-making process of AI and doctors, making it difficult to achieve deep integration to support comprehensive treatment decisions. All of these problems can result in medical plans obtained by AI not meeting the patient's actual symptoms. Summary of the Invention

[0003] This invention provides a method, apparatus, electronic device, and storage medium for generating medical solutions, in order to address the problem that AI-generated medical solutions cannot solve patients' symptoms.

[0004] According to one aspect of the present invention, a method for generating a medical treatment plan is provided, comprising:

[0005] The first statement is determined, and the first statement is parsed to obtain the first information; the first statement is natural language input from the interactive interface, used to characterize the feature information of the object; the first information is used to describe the probability of the disease occurring and the cause; the object is an object suffering from a disease that needs to be diagnosed.

[0006] First diagnostic information is generated based on the first information and the first image; the first image is a medical imaging image of the object; the first diagnostic information is used to characterize the disease type of the object.

[0007] Obtain the second statement, and correct the first diagnostic information according to the second statement to obtain the second diagnostic information; the second statement is information input from the interactive interface to supplement the object features and / or correction information to correct the first diagnostic information; the second diagnostic information is the object's state change information.

[0008] According to another aspect of the present invention, a medical treatment plan generation apparatus is provided, comprising:

[0009] The first information determination module is used to determine a first statement, parse the first statement to obtain first information; the first statement is natural language input from the interactive interface, used to characterize the feature information of the object; the first information is used to describe the probability of the disease occurring and the cause; the object is an object suffering from a disease that needs to be diagnosed.

[0010] The first diagnostic information determination module is used to generate first diagnostic information based on first information and a first image; the first image is a medical contrast image of the object; the first diagnostic information is used to characterize the disease type of the object.

[0011] The second diagnostic information determination module is used to acquire a second statement, and correct the first diagnostic information according to the second statement to obtain second diagnostic information; the second statement is information input from the interactive interface to supplement the object features and / or correction information to correct the first diagnostic information; the second diagnostic information is the object's state change information.

[0012] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0013] At least one processor; and

[0014] A memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the medical plan generation method according to any embodiment of the present invention.

[0016] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the medical plan generation method according to any embodiment of the present invention.

[0017] The technical solution of this invention involves determining a first statement, parsing the first statement to obtain first information, and obtaining the first information to accurately characterize the state changes of an object, providing a basis for determining diagnostic information. First diagnostic information is generated based on the first information and a first image, describing the cause and type of disease caused by the object, providing a basis for disease diagnosis. A second statement is then obtained, and the first diagnostic information is corrected based on the second statement to obtain second diagnostic information, making the obtained second diagnostic information more closely match the individual characteristics of the object and improving the accuracy of disease diagnosis. This method, by parsing the first statement to obtain first information, generating first diagnostic information based on the first information and a first image, and correcting the obtained first diagnostic information using the second statement to obtain second diagnostic information, can combine multi-source data to correct diagnostic information. The bidirectional feedback determination process avoids the omission of key information and enables bidirectional correction of diagnostic information, thereby improving the accuracy of the second diagnostic information.

[0018] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating a medical plan generation method provided in an embodiment of the present invention;

[0021] Figure 2 This is a schematic diagram of the structure of a diagnostic information generation system provided in an embodiment of the present invention;

[0022] Figure 3 This is a schematic diagram of a medical solution generation device provided in an embodiment of the present invention;

[0023] Figure 4 A schematic diagram of the structure of an electronic device for implementing the medical solution generation method of this invention. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0026] Figure 1 This is a flowchart illustrating a method for generating a treatment plan based on a patient's condition, provided by an embodiment of the present invention. This embodiment is applicable to situations where a treatment plan is generated based on a patient's condition. The method can be executed by a device for generating a treatment plan based on a patient's condition. This device can be implemented in hardware and / or software, and can be configured in any electronic device with network communication capabilities. Figure 1 As shown, the method includes:

[0027] S110. Determine the first statement, parse the first statement to obtain the first information; the first statement is natural language input from the interactive interface, used to characterize the object's feature information; the first information is used to describe the probability of the disease occurring and the cause; the object is an object suffering from a disease that needs to be diagnosed.

[0028] The first statement can describe the state changes caused by the disease affecting the object and the object's own characteristics.

[0029] Furthermore, the first statement can be entered through the input module of the interactive interface of the interactive terminal.

[0030] The state change information describes the manifest and latent states of the disease on the object. The manifest state refers to the state change of the object that can be visually detected; the latent state refers to the state change of the object that is detected by the detection device.

[0031] Furthermore, the use of the status change information and the characteristic information of the object itself in this application has been authorized.

[0032] Furthermore, the first statement can be either audio or text information.

[0033] Specifically, the user interface's input module takes into account information describing the object's disease-related state changes and the object's own characteristics, resulting in a first statement. A large-scale model then segments this first statement according to medical terminology rules, yielding at least one first segmented phrase. These at least one first segmented phrase are then marked with their positions, resulting in at least one first phrase. Based on the relationship between medical terms and concepts, and the structural logic of the statement, these at least one first phrase are interpreted, yielding a first definition. Logical reasoning is then applied to the first definition using a pre-defined knowledge graph to determine the first disease type and its probability of occurrence that caused the object's state change. The obtained first disease type and its probability of occurrence are then organized to obtain first information.

[0034] Furthermore, before parsing the first statement, if the first statement is voice input information, it needs to be converted into text.

[0035] Among them, the large-scale model can adopt Med-PaLM (a large-scale language model specifically for the medical field). Med-PaLM is a large-scale language model for the medical field. It is based on the Transformer architecture and is trained on a large amount of data such as medical literature and electronic object feature information. It has the ability to parse medical-related sentences, that is, to parse the first sentence.

[0036] Furthermore, the generation of initial information can be achieved through the intent understanding and feedback module. This module contains a large model.

[0037] Furthermore, such as Figure 2 As shown, the diagnostic information generation system also includes: an intelligent core module and a multimodal data fusion module.

[0038] The intelligent core module is used to generate first and second diagnostic information. It is also used to determine whether there is any missing data in the first information.

[0039] The multimodal data fusion module integrates multi-source information, including image data, laboratory test results, and electronic medical records, into the diagnostic information generation system. The intent understanding and feedback module can access the full dataset; the display terminal can then display the data. The construction of the multimodal data fusion module avoids decision-making biases caused by information silos.

[0040] S120. Generate first diagnostic information based on the first information and the first image; the first image is a medical imaging image of the object; the first diagnostic information is used to characterize the disease type of the object.

[0041] The first image is an image obtained after detection by medical testing instruments, which can characterize the changes in the object's invisible state.

[0042] Specifically, the multimodal data fusion module acquires a first image from the detection device and sends it to the intelligent core module. Upon receiving the first image, the intelligent construct module convolves the first image using a preset convolution kernel to extract texture information of objects within it. Anomaly analysis is performed on the acquired texture information, and lesion regions are generated based on the anomaly analysis results. The first image is then segmented based on the acquired lesion regions, and second information is determined based on the segmented image. The second information is used to match the corresponding disease type from a preset knowledge graph to obtain a second disease type. The acquired second disease type is then used to filter the disease types in the first information to obtain a third disease type. After filtering, since at least one third disease type is obtained, the disease induction probability is generated based on the third disease type and the first interpretation. The obtained induction probability is then mapped to the corresponding third disease type to obtain a first quantitative feature. Third diagnostic information is acquired from the multimodal data fusion module, and the first quantitative feature is correlated and filtered based on the third diagnostic information. A fourth disease type is determined based on the filtering results. Based on the fourth disease type and the first interpretation, a corresponding treatment plan is matched from the preset knowledge graph. The acquired treatment plan is then organized to generate the first diagnostic information.

[0043] The second information includes: the size information, density information, and liquid flow rate information of the objects within the cut image.

[0044] S130. Obtain the second statement, and correct the first diagnostic information according to the second statement to obtain the second diagnostic information; the second statement is information input from the interactive interface to supplement the object features and / or correction information to correct the first diagnostic information; the second diagnostic information is the object's state change information.

[0045] The process involves inputting an evaluation result of the first diagnostic information from the input module of the interactive interface, resulting in a second statement. The second statement is then segmented into phrases using a large model, yielding at least one second segmented phrase. The position of each second segment is marked, resulting in at least one second phrase. The first statement is then interpreted to obtain modification information. Based on the modification information and the first diagnostic information, the disease type causing the object's disease and resulting in state change information is re-identified from a pre-defined knowledge graph. The cause of the disease is then matched based on the acquired disease type. The first diagnostic information is then corrected based on the acquired disease type and cause, resulting in second diagnostic information. If the second diagnostic information does not meet the requirements, correction continues until the requirements are met.

[0046] Optionally, the first statement can be parsed to obtain the first information, including steps A1-A3:

[0047] Step A1: Split the first sentence into phrases to obtain at least one first phrase.

[0048] Specifically, the first sentence is segmented into phrases according to the word segmentation rules for medical terminology, resulting in at least one first segmented phrase. The position of each of the at least one first segmented phrases is then marked, resulting in at least one first phrase.

[0049] Furthermore, the position marking of at least one first split phrase obtained is as follows: the sentence structure of the first sentence is analyzed, phrases containing punctuation or located at sentence boundaries are obtained, and position markings are added to the phrases.

[0050] Furthermore, if punctuation marks are detected in the first statement, punctuation marks are selected according to their type, and punctuation marks with special meanings are retained and marked in position.

[0051] Punctuation marks with special meanings are those that can indicate the state of the first statement. For example, ? represents a question; ! represents emphasis.

[0052] Step A2: Semantically associate at least one first phrase to obtain the first definition.

[0053] Specifically, the first phrase is conceptually mapped based on the association between medical terms and medical concepts to obtain a conceptual mapping relationship. Based on the structural logic of the statement, the logical relationship between at least one first phrase is determined. Based on the obtained logical relationship and conceptual mapping relationship, at least one first phrase is interpreted to obtain a first definition.

[0054] The logical structure of the statements is as follows: the grammatical role and logical relationship of each first phrase in the first statement. The grammatical role is that of a first phrase, either as a subject, predicate, or attributive. The logical relationship is the causal or progressive relationship between different first phrases.

[0055] Furthermore, the process of determining the concept mapping relationship is as follows: based on the medical terminology's analytical concept in medicine, the medical terminology is understood, and the concept mapping relationship is generated based on the understanding result.

[0056] For example, assuming the first word is "treatment", the concept mapping result is: interventions taken to address the causes and symptoms of a disease, including medication, surgery, and lifestyle modifications.

[0057] Step A3: Perform logical reasoning on the first interpretation to obtain the first information.

[0058] Specifically, logical reasoning is performed on the first interpretation based on the preset knowledge graph to determine the first disease type, the probability of occurrence of the first disease type, and the cause of the cause that triggers the change in the state of the object. The obtained first disease type, the probability of occurrence of the first disease type, and the cause of the cause are then organized to obtain the first information.

[0059] Furthermore, the specific steps of logical reasoning are as follows: Based on the state change information in the first interpretation, determine the disease type that causes the object's disease and thus generates state change information from the preset knowledge graph; match the cause of the disease based on the obtained disease type; and calculate the probability that the disease corresponding to the disease type will occur on the object based on the obtained disease type, the cause of the disease, and the object's characteristic information.

[0060] The preset knowledge graph is a knowledge graph constructed based on medical diagnostic results and medical literature, which can represent different diseases, information on changes in state caused by diseases, characteristic information of different objects, and treatment plans.

[0061] Optionally, first diagnostic information is generated based on the first information and the first image, including steps B1-B3:

[0062] Step B1: Cut out the lesion area and extract features from the first image to obtain the second information; the second information is used to describe the lesion changes in the lesion area.

[0063] Specifically, the multimodal data fusion module acquires a first image from the detection device and sends it to the intelligent core module. Upon receiving the first image, the intelligent construct module convolves it with a preset kernel to extract texture information of objects within the image. Anomaly analysis is performed on the acquired texture information, and lesion regions are generated based on the analysis results. The first image is then segmented based on the acquired lesion regions to obtain a second feature image. Size information and object density are extracted from the objects in the second feature image. The flow rate of liquid within the pipes inside the objects in the second feature image is also determined. Finally, the extracted size information, object density, and flow rate of liquid within the pipes are combined to obtain second information.

[0064] Step B2: Generate a first quantitative feature based on the first information, the second information, and the preset knowledge graph; the first quantitative feature is the probability of the object being induced by at least one disease.

[0065] Specifically, based on the second information, the corresponding disease type is matched from a preset knowledge graph to obtain a second disease type. The obtained second disease type is then used to filter the disease types in the first information. If overlapping disease types exist, they are retained to obtain a third disease type. If no overlapping disease types exist, the first and second information are associated, and disease types are re-matched from the preset knowledge graph based on the association result, becoming the third disease type. After filtering, since at least one third disease type is obtained, the disease's induction probability is generated based on the third disease type and the first definition. The obtained induction probability is then mapped to the corresponding third disease type to obtain a first quantitative feature.

[0066] Step B3: Associate the first quantitative feature with the third diagnostic information to obtain the first diagnostic information; the third diagnostic information is the subject's historical disease outcome information.

[0067] Specifically, third diagnostic information is obtained from the multimodal data fusion module. Based on this third diagnostic information, the first quantitative features are correlated and filtered. If the first quantitative features contain the third diagnostic information, overlapping disease types are retained as the fourth disease type. If the first quantitative features do not contain the third diagnostic information, a correlation between disease types is constructed based on the third diagnostic information and the disease types in the first quantitative features, and the fourth disease type is generated based on this correlation. According to the fourth disease type and its first definition, corresponding treatment plans are matched from a preset knowledge graph. The obtained treatment plans are then organized to generate the first diagnostic information.

[0068] Optionally, the lesion region of the first image is segmented and its features are extracted to obtain the second information, including steps C1-C3:

[0069] Step C1: Extract texture features from the first image using a preset convolution kernel to obtain the first feature image.

[0070] Specifically, the multimodal data fusion module acquires the first image from the detection device and sends the first image to the intelligent core module. After receiving the first image, the intelligent construct module performs convolution on the first image according to the preset convolution kernel, extracts the texture information of the objects in the first image, and obtains the first feature image.

[0071] Step C2: Determine the lesion region based on the first feature image, and cut the first image according to the lesion region to obtain the second feature image.

[0072] Specifically, anomaly analysis is performed on the acquired texture information, and lesion regions are generated based on the anomaly analysis results. The first image is then segmented based on the acquired lesion regions to obtain a second feature image.

[0073] Furthermore, the anomaly analysis involves comparing the state of the objects contained in the acquired first feature image with a preset state. If the states are the same, the area corresponding to the state is a normal area; if the states are different, the area corresponding to the state is a lesion area.

[0074] The preset states are the color and size states of a normal object as defined in advance.

[0075] Step C3: Extract features from the second feature image to obtain the second information.

[0076] Specifically, the size information and density of the object in the second feature image are extracted. The flow rate of the liquid inside the pipe within the object in the second feature image is then determined. The extracted size information, object density, and flow rate of the liquid inside the pipe are combined to obtain the second information.

[0077] Optionally, feature extraction is performed based on the second feature image to obtain second information, including steps D1-D4:

[0078] Step D1: Measure the size of the second feature image to obtain the first feature data.

[0079] Specifically, the size of the object in the second feature image is measured using a measuring tool, and the obtained size information is used as the first feature data.

[0080] The dimensional information includes at least the object's three-dimensional volume, longest diameter, and surface area.

[0081] Step D2: Determine the second feature data based on the signal distribution intensity of the second feature image; the second feature data is used to describe the density of the lesion area at different locations.

[0082] Specifically, the average density and density standard deviation of the object are determined based on the signal distribution intensity at different locations of the object in the second feature image. The obtained average density and density standard deviation are used as the second feature data.

[0083] Furthermore, the specific process for determining the second feature data is as follows: determine the signal value of each pixel in the second feature image, average the acquired signal values ​​to obtain the average density, and determine the density standard deviation based on the average density.

[0084] Step D3: Evaluate the liquid flow rate of the second feature image to obtain the third feature data.

[0085] Specifically, signal enhancement is performed on the second feature image. The enhanced signal values ​​and original information values ​​of each pixel of the enhanced object are obtained. Normalization is then performed based on the obtained enhanced signal values ​​and original information values. The flow rate of the liquid is determined based on the normalized enhanced signal values ​​and original information values ​​of the current second feature image, the normalized enhanced signal values ​​and original information values ​​of the previous second feature image, and the scan time of the second feature image, thus obtaining the third feature data.

[0086] Step D4: Generate second information based on the first feature data, the second feature data, and the third feature data.

[0087] Specifically, the acquired first feature data, second feature data, and third feature data are correlated and organized to obtain the second information.

[0088] Optionally, before generating the first quantified feature based on the first information, the second information, and the preset knowledge graph, steps E1-E3 are included:

[0089] Step E1: Determine the missing information for the first definition to obtain the first result; the missing information is the feature description information of the object.

[0090] Specifically, the core module of the intelligent agent determines whether there is any missing information based on the completeness of the first information and the generation specifications of the first information.

[0091] Furthermore, the missing information determination also includes: matching the information contained in the first definition with the information recorded in the configuration information table stored in the database; if there is a missing information, the first result is that the information is missing; if there is no missing information, the first result is that there is no missing information.

[0092] The generation criteria for the first piece of information are: whether it clearly describes the onset time, development process, and accompanying symptoms of the disease; or whether there is a reasonable logical connection between the diagnostic conclusion and the symptoms and examination results, i.e., whether there are any semantic incompleteness or logical inconsistencies.

[0093] Step E2: Generate a data missing instruction based on the first result.

[0094] Specifically, the first result will generate a data missing instruction and be displayed on the interactive interface to provide a reminder for external methods of completing the missing information.

[0095] Step E3: Complete the first information according to the data missing instruction.

[0096] Specifically, the external interface parses the missing data command to obtain the missing information. Based on the missing information, it matches the corresponding supplementary information from the state change information database and sends the supplementary information to the core module of the intelligent agent through the interactive interface.

[0097] Optionally, the first diagnostic information is modified according to the second statement to obtain the second diagnostic information, including steps F1-F2:

[0098] Step F1: Analyze the second statement to determine the modification information.

[0099] Specifically, the evaluation result of the first diagnostic information is input from the input module of the interactive interface, resulting in a second statement. The second statement is then segmented into phrases using a large model, yielding at least one second segmented phrase. The position of each second segment is marked, resulting in at least one second phrase. The first statement is then interpreted to obtain modification information.

[0100] Step F2: Correct the first diagnostic information based on the modified information to obtain the second diagnostic information.

[0101] Specifically, based on the modified information and the first diagnostic information, the disease type that caused the object's disease and thus generated the state change information is re-matched from the preset knowledge graph. The cause of the disease is then matched based on the obtained disease type. The first diagnostic information is then corrected based on the obtained disease type and the cause of the disease to obtain the second diagnostic information.

[0102] The technical solution of this embodiment involves determining a first statement, parsing the first statement to obtain first information, and obtaining the first information to accurately characterize the state changes of the object, providing a basis for determining diagnostic information. First diagnostic information is generated based on the first information and a first image, which describes the cause and type of disease caused by the object, providing a basis for disease diagnosis. A second statement is then obtained, and the first diagnostic information is corrected based on the second statement to obtain second diagnostic information, making the obtained second diagnostic information more closely match the individual characteristics of the object and improving the accuracy of disease diagnosis. This method, by parsing the first statement to obtain first information, generating first diagnostic information based on the first information and a first image, and correcting the obtained first diagnostic information using the second statement to obtain second diagnostic information, can combine multi-source data to correct diagnostic information. The bidirectional feedback determination process avoids the omission of key information and also enables bidirectional correction of diagnostic information, thereby improving the accuracy of the second diagnostic information.

[0103] Figure 3 This is a schematic diagram of a device for generating treatment plans based on a patient's condition, provided in an embodiment of the present invention. This embodiment is applicable to situations requiring the generation of treatment plans based on a patient's condition. The device can be implemented in hardware and / or software, and can be configured in any electronic device with network communication capabilities. Figure 3 As shown, the device includes: a first information determination module 210, a first diagnostic information determination module 220, and a second diagnostic information determination module 230, wherein,

[0104] First Information Determination Module 210: Used to determine the first statement, parse the first statement to obtain the first information; the first statement is natural language input from the interactive interface, used to characterize the feature information of the object; the first information is used to describe the probability of the disease occurring and the cause; the object is an object suffering from a disease that needs to be diagnosed.

[0105] First diagnostic information determination module 220: used to generate first diagnostic information based on first information and first image; the first image is a medical contrast image of the object; the first diagnostic information is used to characterize the disease type of the object;

[0106] The second diagnostic information determination module 230 is used to obtain a second statement, correct the first diagnostic information based on the second statement, and obtain the second diagnostic information; the second statement is information input from the interactive interface to supplement the object features and / or correction information to correct the first diagnostic information; the second diagnostic information is the object's state change information.

[0107] Optionally, the first information determination module 210 includes:

[0108] First phrase determination unit: used to split the first statement into phrases to obtain at least one first phrase;

[0109] First semantic relation determination unit: used to semantically associate at least one first phrase to obtain a first definition;

[0110] First information determination unit: used to perform logical reasoning on the first interpretation to obtain the first information.

[0111] Optionally, the first diagnostic information determination module 220 includes:

[0112] The second information determination unit is used to cut out the lesion region and extract features from the first image to obtain second information; the second information is used to describe the lesion changes in the lesion region.

[0113] First quantitative feature determination unit: used to generate a first quantitative feature based on first information, second information and a preset knowledge graph; the first quantitative feature is the probability of the object being induced by at least one disease;

[0114] First diagnostic information determination unit: used to associate the first quantitative feature with the third diagnostic information to obtain the first diagnostic information; the third diagnostic information is the object's historical disease outcome information.

[0115] Optionally, the second information determining unit includes:

[0116] First feature image determination subunit: used to extract texture features from the first image using a preset convolution kernel to obtain the first feature image;

[0117] Second feature image determination subunit: used to determine the lesion region based on the first feature image, and to cut the first image according to the lesion region to obtain the second feature image;

[0118] The second information determination subunit is used to extract features based on the second feature image to obtain the second information.

[0119] Optionally, the second information determines the sub-unit, specifically used for:

[0120] The size of the second feature image is measured to obtain the first feature data;

[0121] The second feature data is determined based on the signal distribution intensity of the second feature image; the second feature data is used to describe the density of the lesion area at different locations.

[0122] Liquid flow rate is evaluated using the second feature image to obtain the third feature data;

[0123] The second information is generated based on the first feature data, the second feature data, and the third feature data.

[0124] Optionally, the first diagnostic information determination module 220 includes:

[0125] First Result Determination Unit: Used to determine the missing information in the first interpretation and obtain the first result; the missing information is the feature description information of the object.

[0126] Data Missing Instruction Determination Unit: Used to generate a data missing instruction based on the first result;

[0127] Data completion unit: used to complete the first information according to the data missing instruction.

[0128] Optionally, the second diagnostic information determination module 230 includes:

[0129] Modification information determination unit: used to parse the second statement and determine the modification information;

[0130] Second diagnostic information determination unit: used to correct the first diagnostic information based on the modified information to obtain the second diagnostic information.

[0131] The device for generating treatment plans based on the condition provided in the embodiments of the present invention can execute the method for generating treatment plans based on the condition provided in any of the embodiments of the present invention, and has the corresponding functions and beneficial effects of executing the method for generating treatment plans based on the condition. For details, please refer to the relevant operations of the method for generating treatment plans based on the condition in the foregoing embodiments.

[0132] Figure 4 This is a schematic diagram of an electronic device for implementing the method of generating treatment plans based on medical conditions according to embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0133] like Figure 4As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0134] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0135] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as methods for generating treatment plans based on a patient's condition.

[0136] In some embodiments, the method for generating a treatment plan based on a patient's condition may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or mounted on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for generating a treatment plan based on a patient's condition described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the method for generating a treatment plan based on a patient's condition by any other suitable means (e.g., by means of firmware).

[0137] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0138] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0139] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0140] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0141] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0142] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0143] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0144] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for generating a medical plan, characterized in that, include: Determine the first statement, parse the first statement, and obtain the first information; The first statement is natural language input through the interactive interface, used to represent the characteristic information of the object; The first information is used to describe the probability of the disease occurring and the causes; the object is an individual suffering from a disease who needs to be diagnosed. First diagnostic information is generated based on the first information and the first image; the first image is a medical imaging image of the object; the first diagnostic information is used to characterize the disease type of the object. Obtain the second statement, and correct the first diagnostic information according to the second statement to obtain the second diagnostic information; The second statement is information input through the interactive interface to supplement the object's characteristics and / or to correct the first diagnostic information; the second diagnostic information is the object's state change information.

2. The method according to claim 1, characterized in that, The step of parsing the first statement to obtain the first information includes: The first sentence is split into phrases to obtain at least one first phrase; Semantically associate the at least one first phrase to obtain a first definition; Logical reasoning is performed on the first interpretation to obtain the first piece of information.

3. The method according to claim 1, characterized in that, The step of generating first diagnostic information based on first information and first image includes: The first image is segmented into lesion regions and its features are extracted to obtain second information; the second information is used to describe the lesion changes in the lesion regions. A first quantitative feature is generated based on the first information, the second information, and a preset knowledge graph; the first quantitative feature is the probability of the object being induced by at least one disease; The first quantitative feature is correlated with the third diagnostic information to obtain the first diagnostic information; the third diagnostic information is the object's historical disease outcome information.

4. The method according to claim 3, characterized in that, The step of segmenting the lesion region and extracting features from the first image to obtain the second information includes: Texture features are extracted from the first image using a preset convolution kernel to obtain the first feature image; The lesion region is determined based on the first feature image, and the first image is segmented based on the lesion region to obtain the second feature image; Feature extraction is performed on the second feature image to obtain the second information.

5. The method according to claim 4, characterized in that, The step of extracting features from the second feature image to obtain the second information includes: The size of the second feature image is measured to obtain the first feature data; The second feature data is determined based on the signal distribution intensity of the second feature image; the second feature data is used to describe the density of the lesion area at different locations. Liquid flow rate is evaluated on the second feature image to obtain the third feature data; Second information is generated based on the first feature data, the second feature data, and the third feature data.

6. The method according to claim 3, characterized in that, Before generating the first quantified feature based on the first information, the second information, and the preset knowledge graph, the process includes: The first interpretation is evaluated for missing information to obtain a first result; the missing information is the feature description information of the object. Generate a data missing instruction based on the first result; The first information is completed according to the data missing instruction.

7. The method according to claim 1, characterized in that, The step of correcting the first diagnostic information according to the second statement to obtain the second diagnostic information includes: The second statement is parsed to determine the modification information; The first diagnostic information is corrected based on the modified information to obtain the second diagnostic information.

8. A medical treatment plan generation device, characterized in that, include: The first information determination module is used to determine the first statement, parse the first statement, and obtain the first information. The first statement is natural language input through the interactive interface, used to represent the characteristic information of the object; The first information is used to describe the probability of the disease occurring and the causes; the object is an individual suffering from a disease who needs to be diagnosed. The first diagnostic information determination module is used to generate first diagnostic information based on first information and a first image; the first image is a medical contrast image of the object; the first diagnostic information is used to characterize the disease type of the object. The second diagnostic information determination module is used to obtain a second statement, and to correct the first diagnostic information according to the second statement to obtain the second diagnostic information. The second statement is information input through the interactive interface to supplement the object's characteristics and / or to correct the first diagnostic information; the second diagnostic information is the object's state change information.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the medical protocol generation method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the medical protocol generation method according to any one of claims 1-7.

Images