Method, device and equipment for realizing rapid migration of multi-school-of-thought syndrome diagnosis of traditional Chinese medicine

Abstract

The present application relates to the field of model hosting and the field of digital medicine, and discloses a method, device and equipment for realizing rapid migration of traditional Chinese medicine syndrome differentiation and diagnosis of multiple schools, which comprises the following steps: querying candidate syndrome type labels of case samples, representing syndrome differentiation node states of the candidate syndrome type labels, and constructing a ring network of the candidate syndrome type labels based on the syndrome differentiation node states; performing vector coding on diagnosis text to obtain diagnosis vectors, splicing the diagnosis vectors and school vectors to obtain spliced vectors, reconstructing the school vectors by using the spliced vectors based on the ring network, and obtaining reconstructed school vectors; constructing a diagnosis data batch of the diagnosis text, selecting a target school vector from the reconstructed school vectors by using the diagnosis data batch, performing comparative learning training on the target school vector based on the diagnosis data batch, and obtaining a trained school vector; and performing syndrome type school migration by using the trained school vector and the diagnosis text to obtain a school-migrated syndrome type. The present application can flexibly switch between multiple syndrome differentiation systems of traditional Chinese medicine.

Description

Technical Field

[0001] This invention relates to the fields of model hosting and digital healthcare, and in particular to a method, apparatus, and equipment for the rapid transfer of diagnostic methods from multiple schools of thought in Traditional Chinese Medicine. Background Technology

[0002] Differentiation of syndromes is a unique feature of disease diagnosis in Traditional Chinese Medicine (TCM). Throughout its historical development, various schools of thought on differentiation of syndromes have emerged, each with its own emphasis. Modern TCM, on the other hand, has integrated the strengths of these schools to form a comprehensive method of differentiation of syndromes, such as differentiation of syndromes based on the Eight Principles, differentiation of syndromes based on the Zang-Fu organs, differentiation of syndromes based on Qi, Blood and Body Fluids, and differentiation of syndromes based on the Six Channels.

[0003] With the rise of machine learning technology, AI-powered medical consultations can support functions such as disease-assisted diagnosis, health management, and remote consultation. However, the same disease can have different syndrome names under different diagnostic systems, which poses a challenge to the standardization and intelligentization of TCM diagnosis and treatment. For example, "Syndrome A" and "Syndrome B" are different. Existing intelligent TCM diagnostic technologies build models based on data from the same diagnostic system. The problem is that in real-world TCM diagnosis and treatment data, syndrome diagnosis is based on the physician's own habits regarding the methods of their inherited system, which has a great deal of freedom. As a result, there are more than a thousand types of TCM syndromes, leading to an extremely uneven distribution of case data. In addition, there is the problem of semantic overlap. Syndrome is a description of the patient's state. The same symptom can be identified from the perspective of the internal organs or from the perspective of Wei Qi Ying Xue. Therefore, from a modeling perspective, it is difficult to identify the same symptom from both perspectives at the same time. Given the excessive number and uneven distribution of syndrome types mentioned above, it is impractical to build a model for every syndrome under every diagnostic system. Overall, there is currently a lack of technology or products that can flexibly switch between the various diagnostic systems in Traditional Chinese Medicine. Summary of the Invention

[0004] This invention provides a method, apparatus, and equipment for rapid migration of syndrome differentiation and diagnosis among multiple schools of thought in Traditional Chinese Medicine (TCM). Its main purpose is to enable flexible switching between various syndrome differentiation systems in TCM.

[0005] To achieve the above objectives, this invention provides a method for rapid transfer of diagnostic methods from multiple schools of thought in Traditional Chinese Medicine, comprising: Obtain case samples, query candidate syndrome type labels of the case samples, characterize the dialectical node state of the candidate syndrome type labels, and construct a ring network of the candidate syndrome type labels based on the dialectical node state. Obtain the school of thought vector in the dialectical node state and the diagnostic text of the case sample, perform vector encoding on the diagnostic text to obtain the diagnostic vector, concatenate the diagnostic vector and the school of thought vector to obtain the concatenated vector, and reconstruct the school of thought vector based on the ring network to obtain the reconstructed school of thought vector. Construct a diagnostic data batch of the diagnostic text, select a target school vector from the reconstructed school vector using the diagnostic data batch, and perform comparative learning training on the target school vector based on the diagnostic data batch to obtain a training school vector; The training school vectors and the diagnostic text are used to perform school transfer of syndrome types to obtain school transfer syndrome types.

[0006] Optionally, querying the candidate syndrome labels of the case sample includes: Query the initial syndrome label of the case sample; The initial certificate type label is decomposed into certificate element labels; Obtain the syndrome type samples from multiple dialectical systems corresponding to the case samples; Identify the character edit distance between the evidence element label and the evidence type sample; When the character editing distance is less than the preset editing distance, the certificate type sample and the initial certificate type label are used as the candidate certificate type label.

[0007] Optionally, constructing the ring network of candidate evidence type labels based on the dialectical node states includes: Obtain the network node corresponding to the dialectical node state, and determine the node path of the network node using the following format: in, This represents the node path of the network node. Nodes represent the syndrome type labels under different diagnostic systems within a specific case data set. This indicates switching from a certificate type under one school of thought to a certificate type under another school of thought; The network containing the dialectical node states and the node paths is defined as the ring network.

[0008] Optionally, concatenating the diagnostic vector and the school vector to obtain the concatenated vector includes: The school of thought vector is randomly initialized to obtain an initialization vector; The diagnostic vector and the school of thought vector are concatenated using the following format to obtain the concatenated vector: in, This represents the concatenated vector. This represents the initialization vector. Let i represent the diagnostic vector, where i represents the vector index.

[0009] Optionally, the step of reconstructing the school vector using the spliced ​​vector based on the ring network to obtain the reconstructed school vector includes: The concatenated vector is input into a preset certificate detection model; The certificate type detection model is used to output the certificate type label sequence of the concatenated vector; Query whether the switching relationship between the school vector in the spliced ​​vector and the school vector corresponding to the certificate type label sequence conforms to the switching relationship of the school vector in the ring network; When the switching relationship between the school vector in the spliced ​​vector and the school vector corresponding to the certificate type label sequence does not conform to the switching relationship of the school vector in the ring network, the target vector of the school vector in the spliced ​​vector is queried in the certificate type label sequence, and the target vector is replaced with the school vector to obtain the reconstructed school vector.

[0010] Optionally, the step of performing comparative learning training on the target school vector based on the diagnostic data batch to obtain a training school vector includes: The diagnostic data is batch-input into a preset contrastive learning model; Obtain data batch samples and their parent syndrome type samples from the diagnostic data batch; based on the parent syndrome type samples and the target school vector, query the diagnostic dissimilar samples and school identical samples in the data batch samples. The semantic distance between the diagnostic dissimilar samples is expanded using the contrastive learning model, resulting in an expanded semantic distance. The semantic distance of samples with the same genre is reduced by using the contrastive learning model, resulting in reduced semantic distance. The target genre vector, which includes the expanded semantic distance and the reduced semantic distance, is used as the training genre vector.

[0011] To address the aforementioned problems, the present invention also provides a device for rapid transfer of diagnostic methods across multiple schools of thought in Traditional Chinese Medicine, the device comprising: The network construction module is used to acquire case samples, query candidate syndrome type labels of the case samples, characterize the dialectical node state of the candidate syndrome type labels, and construct a ring network of the candidate syndrome type labels based on the dialectical node state. The vector reconstruction module is used to obtain the school vector in the dialectical node state and the diagnostic text of the case sample, perform vector encoding on the diagnostic text to obtain the diagnostic vector, concatenate the diagnostic vector and the school vector to obtain the concatenated vector, and reconstruct the school vector based on the ring network to obtain the reconstructed school vector. The vector training module is used to construct a diagnostic data batch of the diagnostic text, select a target school vector from the reconstructed school vector using the diagnostic data batch, and perform comparative learning training on the target school vector based on the diagnostic data batch to obtain a training school vector. The school of thought transfer module is used to perform school of thought transfer using the trained school of thought vectors and the diagnostic text to obtain the school of thought transfer certificate.

[0012] To address the above problems, the present invention also provides an electronic device, the electronic device comprising: 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, and the computer program is executed by the at least one processor to realize the above-described method for rapid transfer of diagnosis and treatment based on multiple schools of thought in traditional Chinese medicine.

[0013] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one computer program, which is executed by a processor in an electronic device to implement the aforementioned method for rapid migration of diagnosis based on multiple schools of thought in traditional Chinese medicine.

[0014] As can be seen, this embodiment of the invention queries the candidate syndrome type labels of the case samples to find multiple candidate syndrome type labels that may exist for the same case across different syndrome differentiation systems. Furthermore, this embodiment constructs a circular network of the candidate syndrome type labels based on the state of the syndrome nodes to determine the switching relationship between syndrome types of the same case in different syndrome differentiation systems. The advantage of this is that it is equivalent to changing the position of only one node in the circular network each time, while the remaining sequence order remains unchanged. This is a sampling based on a Markov process, sampling the transformation relationship between syndrome differentiation and syndrome differentiation styles, making it easier to implicitly mine from the data which syndrome differentiation styles are more likely to transition. Furthermore, this embodiment reconstructs the school of thought vector based on the circular network using the spliced ​​vector. This is used to minimize the loss caused by the conversion between these syndrome type labels when different physicians diagnose the same case and arrive at syndrome type conclusions belonging to different syndrome differentiation systems, thereby obtaining the representation results of each syndrome differentiation system / school of thought. Therefore, the method, apparatus, and equipment for rapid migration of TCM multi-school syndrome differentiation and diagnosis proposed in this embodiment of the invention can flexibly switch between various TCM syndrome differentiation systems. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating a method for rapid transfer of diagnosis from multiple schools of thought in traditional Chinese medicine, provided in an embodiment of the present invention. Figure 2 A schematic diagram of a module for a rapid transfer device for diagnosis of multiple schools of thought in traditional Chinese medicine provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the internal structure of an electronic device for implementing a rapid transfer method for syndrome differentiation and diagnosis across multiple schools of traditional Chinese medicine, as provided in an embodiment of the present invention. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0016] This invention provides a method for rapid migration of TCM multi-school syndrome differentiation diagnosis. The execution subject of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this invention: a server, a terminal, etc. In other words, the method for rapid migration of TCM multi-school syndrome differentiation diagnosis 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. The server can be an independent server or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0017] Reference Figure 1 The diagram shown is a flowchart illustrating a method for rapid transfer of TCM multi-school syndrome differentiation diagnosis according to an embodiment of the present invention. In this embodiment, the method includes the following steps S1-S4: S1. Obtain case samples, query candidate syndrome type labels of the case samples, characterize the dialectical node state of the candidate syndrome type labels, and construct a ring network of the candidate syndrome type labels based on the dialectical node state.

[0018] In this embodiment of the invention, the case samples include multiple different case samples, and the content of the case samples varies in different medical scenarios. For example, in the scenario of diagnosing headaches, the case samples include "wind-cold headache: manifested as aversion to wind and cold, pain extending to the neck and back, aggravated by wind and cold, white tongue coating, and floating and tight pulse; treatment is to dispel wind and cold, commonly used drugs include Chuanxiong Tea Powder, Duliang Pill, etc." and "wind-heat headache: manifested as headache with distension, headache like splitting, fever or aversion to wind, flushed face and red eyes, thirst, constipation, yellow urine, red tongue, yellow coating, and floating and rapid pulse." The treatment principle is to dispel wind and clear heat. Commonly used drugs include Chuanxiong Zhishi Gao Tang, Huanglian Shangqing Wan, Sangju Yin, Qingxuan Wan, etc. "Rheumatic headache: manifested as headache as if wrapped in a cloth, heaviness in the limbs, poor appetite, chest tightness, difficulty in urination, loose stools, white and greasy tongue coating, and weak pulse. Commonly used drugs include Qianghuo Shenshi Tang with modifications, Huanglian Xiangru Yin with modifications, etc." In the context of stomach pain diagnosis, the case samples mentioned include "Stomach pain in traditional Chinese medicine is also called epigastric pain, which is mostly caused by external pathogens invading the stomach, food damaging the stomach, emotional distress, and inherent spleen deficiency, etc., and is a disease with pain in the upper abdomen near the heart as the main symptom."

[0019] Furthermore, in this embodiment of the invention, candidate syndrome type labels of the case sample are queried to find multiple candidate syndrome type labels that may exist for the same case across different syndrome differentiation systems.

[0020] In one embodiment of the present invention, querying the candidate syndrome type labels of the case sample includes: querying the initial syndrome type label of the case sample; decomposing the initial syndrome type label into syndrome element labels; obtaining syndrome type samples within multiple dialectical systems corresponding to the case sample; identifying the character edit distance between the syndrome element labels and the syndrome type samples; and when the character edit distance is less than a preset edit distance, using the syndrome type samples and the initial syndrome type label as the candidate syndrome type labels.

[0021] For example, since the name of a syndrome describes the mechanism of a disease, it can be further broken down into syndrome elements. Therefore, even syndromes across different systems have commonalities in name and syndrome elements. For example, the syndrome elements of "liver fire disturbing the heart" are "liver, heart, yin deficiency, fire"; the syndrome elements of "yin deficiency and fire excess" are "yin deficiency, fire". First, a heuristic search is performed using the character edit distance of the syndrome name and syndrome elements to find multiple candidate syndrome labels that may exist in the same case across different syndrome differentiation systems. Then, experts conduct a second confirmation.

[0022] In this embodiment of the invention, the dialectical node state refers to the state of the syndrome type under different dialectical systems, the school vector of the dialectical system to which the syndrome type belongs, and the case text vector corresponding to the syndrome type.

[0023] In one embodiment of the present invention, the dialectical node state of the candidate evidence type label is characterized using the following format: in, This indicates the state of the dialectical node. Represents a genre vector. This represents the encoded vector of the case text. and Let R represent the dimensions of the vector. A set of.

[0024] Furthermore, in this embodiment of the invention, a circular network of candidate syndrome type labels is constructed based on the state of the dialectical nodes. This circular network is used to determine the switching relationship between syndrome types of the same case in different dialectical systems. The advantage of doing so is that it is equivalent to changing the position of only one node in the circular network each time, while the order of the remaining sequence remains unchanged. This is a sampling based on a Markov process, which samples the transformation relationship between dialectics and dialectical styles, making it easier to implicitly mine from the data which dialectical styles are more likely to have a transformation relationship.

[0025] In one embodiment of the present invention, constructing a ring network of candidate dialectical labels based on the dialectical node state includes: obtaining the network node corresponding to the dialectical node state, and determining the node path of the network node using the following format: in, This represents the node path of the network node. Nodes represent the syndrome type labels under different diagnostic systems within a specific case data set. This indicates switching from a certificate type under one school of thought to a certificate type under another school of thought; The network containing the dialectical node states and the node paths is defined as the ring network.

[0026] S2. Obtain the school vector in the dialectical node state and the diagnostic text of the case sample, perform vector encoding on the diagnostic text to obtain a diagnostic vector, concatenate the diagnostic vector and the school vector to obtain a concatenated vector, and reconstruct the school vector based on the ring network using the concatenated vector to obtain a reconstructed school vector.

[0027] In this embodiment of the invention, the diagnostic text refers to the text used to diagnose the case sample.

[0028] In one embodiment of the present invention, the step of vector encoding the diagnostic text to obtain a diagnostic vector includes: performing random addition and deletion processing on the diagnostic text to obtain randomly added and deleted text; performing text sequence modeling on the randomly added and deleted text to obtain a text sequence model; extracting the hidden layer vector of the text sequence model, and using the hidden layer vector as the diagnostic vector.

[0029] For example, enter the diagnosis text of a case. Perform random character deletion and random Chinese character insertion operations on it, and repeat. Next, a text sequence with noise is generated: ,right The text sequence is modeled using Bi-LSTM, with the last hidden layer of the LSTM used as a vector representation of the entire diagnostic text.

[0030] In one embodiment of the present invention, concatenating the diagnostic vector and the school vector to obtain a concatenated vector includes: randomly initializing the school vector to obtain an initialization vector; and concatenating the diagnostic vector and the school vector using the following format to obtain the concatenated vector: in, This represents the concatenated vector. This represents the initialization vector. Let i represent the diagnostic vector, where i represents the vector index.

[0031] Furthermore, in this embodiment of the invention, the school vector is reconstructed using the spliced ​​vector based on the ring network. This is used to minimize the loss caused by the conversion between these syndrome labels when different physicians diagnose the same case and arrive at syndrome conclusions belonging to different syndrome systems. This results in the representation of each syndrome system / school.

[0032] In one embodiment of the present invention, the step of reconstructing the school vector using the spliced ​​vector based on the ring network to obtain the reconstructed school vector includes: inputting the spliced ​​vector into a preset certificate type detection model; outputting a certificate type label sequence of the spliced ​​vector using the certificate type detection model; querying whether the switching relationship between the school vector in the spliced ​​vector and the school vector corresponding to the certificate type label sequence conforms to the switching relationship of school vectors in the ring network; when the switching relationship between the school vector in the spliced ​​vector and the school vector corresponding to the certificate type label sequence does not conform to the switching relationship of school vectors in the ring network, querying the target vector of the school vector in the spliced ​​vector in the certificate type label sequence, replacing the target vector with the school vector, and obtaining the reconstructed school vector.

[0033] For example, using the BERT+CRF+cross-entropy loss model, the input... Output the label sequence of the certificate type. Training results It is obtained based on the modeling of the transformation relationship of multiple syndrome types of the same disease; wherein, the switching relationship between the school vector in the splicing vector and the school vector corresponding to the syndrome label sequence refers to the relationship between network nodes switching through node paths.

[0034] S3. Construct a diagnostic data batch of the diagnostic text, select a target school vector from the reconstructed school vector using the diagnostic data batch, and perform comparative learning training on the target school vector based on the diagnostic data batch to obtain a training school vector.

[0035] In one embodiment of the present invention, selecting a target school vector from the reconstructed school vector using the diagnostic data batch includes: querying the school system category corresponding to each diagnostic data batch in the diagnostic data batch; querying the reconstructed school vector corresponding to the school system category in the reconstructed school vector to obtain the target school vector.

[0036] In one embodiment of the present invention, the step of performing comparative learning training on the target school vector based on the diagnostic data batch to obtain a training school vector includes: inputting the diagnostic data batch into a preset comparative learning model; obtaining data batch samples and their parent class symptom samples in the diagnostic data batch; querying the diagnostic dissimilar samples and school-same samples in the data batch samples according to the parent class symptom samples and the target school vector; expanding the semantic distance of the diagnostic dissimilar samples using the comparative learning model to obtain expanded semantic distance; reducing the semantic distance of the school-same samples using the comparative learning model to obtain reduced semantic distance; and using the target school vector containing the expanded semantic distance and the reduced semantic distance as the training school vector.

[0037] For example, a contrastive learning algorithm is used; let each Samples in Compared with other locations Samples within The semantic distance should be as large as possible from itself, and from other samples of the same school of thought that are located in the parent class of the same type of evidence. The semantic distance is smaller.

[0038] S4. Use the trained school vectors and the diagnostic text to perform school transfer to obtain the school transfer certificate.

[0039] In one embodiment of the present invention, the step of performing syndrome transfer using the training school vector and the diagnostic text to obtain a school transfer syndrome includes: concatenating the training school vector and the diagnostic text to obtain a school-diagnosis vector; and classifying the school-diagnosis vector to obtain the school transfer syndrome.

[0040] As can be seen, this embodiment of the invention queries the candidate syndrome type labels of the case samples to find multiple candidate syndrome type labels that may exist for the same case across different syndrome differentiation systems. Furthermore, this embodiment constructs a circular network of the candidate syndrome type labels based on the state of the syndrome nodes to determine the switching relationship between syndrome types of the same case in different syndrome differentiation systems. The advantage of this is that it is equivalent to changing the position of only one node in the circular network each time, while the order of the remaining sequences remains unchanged. This is a sampling method based on a Markov process, sampling the transformation relationship between syndrome differentiation and syndrome differentiation styles, making it easier to implicitly mine from the data which syndrome differentiation styles are more likely to transition. Further, this embodiment reconstructs the school of thought vector based on the circular network using the spliced ​​vector. This is used to minimize the loss caused by the conversion between these syndrome type labels when different physicians diagnose the same case and arrive at syndrome type conclusions belonging to different syndrome differentiation systems, thereby obtaining the representation results of each syndrome differentiation system / school of thought. Therefore, the fast transfer method for multi-school syndrome differentiation diagnosis in Traditional Chinese Medicine proposed in this embodiment can flexibly switch between multiple syndrome differentiation systems in Traditional Chinese Medicine.

[0041] like Figure 2 The diagram shown is a functional module diagram of the rapid transfer device for TCM multi-school syndrome differentiation and diagnosis according to the present invention.

[0042] The rapid migration device 100 for implementing TCM multi-school syndrome differentiation and diagnosis according to the present invention can be installed in an electronic device. Depending on the functions implemented, the rapid migration device may include a network construction module 101, a vector reconstruction module 102, a vector training module 103, and a school migration module 104. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, and is stored in the memory of the electronic device.

[0043] In this embodiment, the functions of each module / unit are as follows: The network construction module 101 is used to acquire case samples, query candidate syndrome type labels of the case samples, characterize the dialectical node state of the candidate syndrome type labels, and construct a ring network of the candidate syndrome type labels based on the dialectical node state. The vector reconstruction module 102 is used to obtain the school vector in the dialectical node state and the diagnostic text of the case sample, perform vector encoding on the diagnostic text to obtain a diagnostic vector, concatenate the diagnostic vector and the school vector to obtain a concatenated vector, and reconstruct the school vector using the concatenated vector based on the ring network to obtain a reconstructed school vector. The vector training module 103 is used to construct a diagnostic data batch of the diagnostic text, select a target school vector from the reconstructed school vector using the diagnostic data batch, and perform comparative learning training on the target school vector based on the diagnostic data batch to obtain a training school vector. The school of thought transfer module 104 is used to perform school of thought transfer using the trained school of thought vector and the diagnostic text to obtain the school of thought transfer certificate.

[0044] In detail, the modules in the rapid transfer device 100 for realizing multi-school syndrome differentiation and diagnosis in traditional Chinese medicine described in this embodiment of the invention employ the same methods as described above during use. Figure 1 The method described herein is the same as the rapid transfer method for syndrome differentiation and diagnosis of multiple schools of TCM, and can produce the same technical effect, so it will not be elaborated here.

[0045] like Figure 3 The diagram shown is a schematic diagram of the electronic device 1 used in the present invention to realize a rapid transfer method for syndrome differentiation and diagnosis of multiple schools of traditional Chinese medicine.

[0046] The electronic device 1 may include a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a rapid migration program for diagnosis based on multiple schools of thought in traditional Chinese medicine.

[0047] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device 1, connecting various components of the electronic device 1 through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., executing a rapid migration program for multi-school syndrome differentiation diagnosis in traditional Chinese medicine), and calls data stored in the memory 11 to perform various functions and process data of the electronic device 1.

[0048] The memory 11 includes at least one type of readable storage medium, including flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as the portable hard drive of the electronic device 1. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 can include both internal and external storage units of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code for a rapid migration program for multi-school diagnostic methods in traditional Chinese medicine, but also to temporarily store data that has been output or will be output.

[0049] The communication bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable communication between the memory 11 and at least one processor 10, etc.

[0050] The communication interface 13 is used for communication between the aforementioned electronic device 1 and other devices, including a network interface and an employee interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, Bluetooth interface, etc.), typically used to establish a communication connection between the electronic device 1 and other electronic devices 1. The employee interface may be a display, an input unit (such as a keyboard), or optionally, a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual employee interface.

[0051] Figure 3 Only electronic device 1 with components is shown; 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, or combine certain components, or have different component arrangements.

[0052] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0053] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in terms of the scope of the patent invention.

[0054] The fast migration program for TCM multi-school syndrome differentiation diagnosis stored in the memory 11 of the electronic device 1 is a combination of multiple computer programs. When run in the processor 10, it can achieve the following: Obtain case samples, query candidate syndrome type labels of the case samples, characterize the dialectical node state of the candidate syndrome type labels, and construct a ring network of the candidate syndrome type labels based on the dialectical node state. Obtain the school of thought vector in the dialectical node state and the diagnostic text of the case sample, perform vector encoding on the diagnostic text to obtain the diagnostic vector, concatenate the diagnostic vector and the school of thought vector to obtain the concatenated vector, and reconstruct the school of thought vector based on the ring network to obtain the reconstructed school of thought vector. Construct a diagnostic data batch of the diagnostic text, select a target school vector from the reconstructed school vector using the diagnostic data batch, and perform comparative learning training on the target school vector based on the diagnostic data batch to obtain a training school vector; The training school vectors and the diagnostic text are used to perform school transfer of syndrome types to obtain school transfer syndrome types.

[0055] Specifically, the processor 10's implementation method of the above-mentioned computer program can be found in [reference needed]. Figure 1 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0056] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0057] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device 1, can perform the following: Obtain case samples, query candidate syndrome type labels of the case samples, characterize the dialectical node state of the candidate syndrome type labels, and construct a ring network of the candidate syndrome type labels based on the dialectical node state. Obtain the school of thought vector in the dialectical node state and the diagnostic text of the case sample, perform vector encoding on the diagnostic text to obtain the diagnostic vector, concatenate the diagnostic vector and the school of thought vector to obtain the concatenated vector, and reconstruct the school of thought vector based on the ring network to obtain the reconstructed school of thought vector. Construct a diagnostic data batch of the diagnostic text, select a target school vector from the reconstructed school vector using the diagnostic data batch, and perform comparative learning training on the target school vector based on the diagnostic data batch to obtain a training school vector; The training school vectors and the diagnostic text are used to perform school transfer of syndrome types to obtain school transfer syndrome types.

[0058] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0059] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0060] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0061] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0062] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0063] The embodiments of this invention can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0064] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a system claim may also be implemented by a single unit or device through software or hardware. The term "second class" is used to indicate names and does not indicate any specific order.

[0065] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for rapid transfer of diagnostic methods from multiple schools of thought in Traditional Chinese Medicine, characterized in that, The method includes: Obtain case samples, query candidate syndrome type labels of the case samples, characterize the dialectical node state of the candidate syndrome type labels, and construct a ring network of the candidate syndrome type labels based on the dialectical node state. Obtain the school of thought vector in the dialectical node state and the diagnostic text of the case sample, perform vector encoding on the diagnostic text to obtain the diagnostic vector, concatenate the diagnostic vector and the school of thought vector to obtain the concatenated vector, and reconstruct the school of thought vector based on the ring network to obtain the reconstructed school of thought vector. Construct a diagnostic data batch of the diagnostic text, select a target school vector from the reconstructed school vector using the diagnostic data batch, and perform comparative learning training on the target school vector based on the diagnostic data batch to obtain a training school vector; The training school vectors and the diagnostic text are used to perform school transfer of syndrome types to obtain school transfer syndrome types. The step of querying the candidate syndrome type labels of the case sample includes: querying the initial syndrome type label of the case sample, decomposing the initial syndrome type label into syndrome element labels, obtaining syndrome type samples within multiple dialectical systems corresponding to the case sample, identifying the character edit distance between the syndrome element label and the syndrome type sample, and when the character edit distance is less than a preset edit distance, using the syndrome type sample and the initial syndrome type label as the candidate syndrome type labels. The construction of the ring network for the candidate evidence type labels based on the dialectical node states includes: Obtain the network node corresponding to the dialectical node state, and determine the node path of the network node using the following format: in, This represents the node path of the network node. Nodes represent the syndrome type labels under different diagnostic systems within a specific case data set. This indicates switching from a certificate type under one school of thought to a certificate type under another school of thought; The network containing the dialectical node states and the node paths is defined as the ring network.

2. The method for rapid transfer of TCM multi-school syndrome differentiation and diagnosis as described in claim 1, characterized in that, The process of concatenating the diagnostic vector and the school of thought vector to obtain the concatenated vector includes: The school of thought vector is randomly initialized to obtain an initialization vector; The diagnostic vector and the school of thought vector are concatenated using the following format to obtain the concatenated vector: in, This represents the concatenated vector. This represents the initialization vector. Let i represent the diagnostic vector, where i represents the vector index.

3. The method for rapid transfer of TCM multi-school syndrome differentiation and diagnosis as described in claim 1, characterized in that, The process of reconstructing the school of thought vector based on the ring network using the spliced ​​vector to obtain the reconstructed school of thought vector includes: The concatenated vector is input into a preset certificate detection model; The certificate type detection model is used to output the certificate type label sequence of the concatenated vector; Query whether the switching relationship between the school vector in the spliced ​​vector and the school vector corresponding to the certificate type label sequence conforms to the switching relationship of the school vector in the ring network; When the switching relationship between the school vector in the spliced ​​vector and the school vector corresponding to the certificate type label sequence does not conform to the switching relationship of the school vector in the ring network, the target vector of the school vector in the spliced ​​vector is queried in the certificate type label sequence, and the target vector is replaced with the school vector to obtain the reconstructed school vector.

4. The method for rapid transfer of TCM multi-school syndrome differentiation and diagnosis as described in claim 1, characterized in that, The step of performing comparative learning training on the target school vector based on the diagnostic data batch to obtain the training school vector includes: The diagnostic data is batch-input into a preset contrastive learning model; Obtain data batch samples and their parent syndrome type samples from the diagnostic data batch; based on the parent syndrome type samples and the target school vector, query the diagnostic dissimilar samples and school identical samples in the data batch samples. The semantic distance between the diagnostic dissimilar samples is expanded using the contrastive learning model, resulting in an expanded semantic distance. The semantic distance of samples with the same genre is reduced by using the contrastive learning model, resulting in reduced semantic distance. The target genre vector, which includes the expanded semantic distance and the reduced semantic distance, is used as the training genre vector.

5. A device for rapid transfer of TCM multi-school syndrome differentiation diagnosis, used to implement the TCM multi-school syndrome differentiation diagnosis rapid transfer method as described in any one of claims 1 to 4, characterized in that, The device includes: The network construction module is used to acquire case samples, query candidate syndrome type labels of the case samples, characterize the dialectical node state of the candidate syndrome type labels, and construct a ring network of the candidate syndrome type labels based on the dialectical node state. The vector reconstruction module is used to obtain the school vector in the dialectical node state and the diagnostic text of the case sample, perform vector encoding on the diagnostic text to obtain the diagnostic vector, concatenate the diagnostic vector and the school vector to obtain the concatenated vector, and reconstruct the school vector based on the ring network to obtain the reconstructed school vector. The vector training module is used to construct a diagnostic data batch of the diagnostic text, select a target school vector from the reconstructed school vector using the diagnostic data batch, and perform comparative learning training on the target school vector based on the diagnostic data batch to obtain a training school vector. The school of thought transfer module is used to perform school of thought transfer using the trained school of thought vectors and the diagnostic text to obtain the school of thought transfer certificate.

6. 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 rapid migration method for TCM multi-school syndrome differentiation diagnosis as described in any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the rapid migration method for diagnosis of multiple schools of thought in traditional Chinese medicine as described in any one of claims 1 to 4.

Images