AI diagnosis method, device, storage medium and equipment based on medical inquiry request

Abstract

The present disclosure provides an AI diagnosis method, apparatus, storage medium and device based on a consultation request, by presenting questions to be asked and corresponding answer options on the client, when the patient / medical staff makes a choice from the answer options to be asked presented on the client, the server inputs the answers that the target patient has given in the past into a pre-trained disease diagnosis model to obtain the diagnosis result information. Then, determine whether to continue the consultation based on the diagnosis result information. If it is determined to continue the consultation, the current question information to be asked is determined from a preset consultation question information set based on the historical consultation reply information sequence, and the current question information to be asked is sent to the client. Subsequently, the client can continue to present the current question to be asked and the corresponding answer options for the patient / medical staff to answer. It can reduce the difficulty and complexity of the consultation, improve the consultation speed, and make the consultation process more targeted.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to an AI diagnosis method, apparatus, storage medium and device based on a medical inquiry request. Background Art

[0002] With the rapid popularization of smart terminals and the advancement of network technology, remote consultation services have emerged. Remote consultation refers to patients or medical staff in relatively underdeveloped areas using the client to input condition description information, and then the server generates diagnostic suggestion information based on the above condition description information and feeds it back to the client, so that patients can obtain diagnostic suggestions, reducing the time and economic cost of patients seeking medical treatment. Medical staff can also provide patients with diagnostic solutions based on the diagnostic suggestions, thereby improving the diagnostic level of medical staff.

[0003] In the existing remote consultation methods, users usually upload information describing their condition, such as text descriptions or scanned images of test results, etc. The system then stores the received information and professional doctors respond with diagnostic recommendations based on the information. Summary of the invention

[0004] The present disclosure proposes an AI diagnosis method, apparatus, storage medium and device based on a medical inquiry request.

[0005] In a first aspect, the present disclosure provides an AI diagnosis method based on a consultation request, which is applied to a server, and the method includes: in response to receiving a consultation reply request for a target patient identifier sent by a client, adding the consultation question and the corresponding selected answer options in the consultation reply request to a historical consultation reply information sequence corresponding to the target patient identifier; inputting the historical consultation reply information sequence into a pre-trained disease diagnosis model to obtain diagnosis result information; determining whether to continue the consultation based on the diagnosis result information; in response to determining to continue the consultation, determining the current question information to be consulted from the preset consultation question information set based on the historical consultation reply information sequence and the diagnosis result information, and sending the current question information to be consulted to the client, so that the client can present the current question information to be consulted, wherein the consultation question information includes the consultation question and corresponding at least two answer options.

[0006] In some optional embodiments, the method further comprises:

[0007] In response to receiving a consultation request for the target patient identifier sent by a client, determining a first question to be consulted from the preset consultation question information set, and sending the first question to be consulted to the client.

[0008] In some optional embodiments, the method further comprises:

[0009] In response to determining not to continue the consultation, the diagnosis result information is sent to the client so that the client can present the diagnosis result information.

[0010] In some optional implementations, each of the preset medical question information is associated with a question weight; and

[0011] The step of determining the first question information to be asked from the preset question information set includes:

[0012] The medical question information with the highest corresponding question weight in the preset medical question information set is determined as the first medical question information to be asked.

[0013] In some optional embodiments, the disease diagnosis model includes a confirmed diagnosis model; and the step of inputting the historical medical consultation answer information sequence into a pre-trained disease diagnosis model to obtain diagnosis result information includes:

[0014] Input the historical medical consultation answer information sequence into the confirmed diagnosis model to obtain a confirmed diagnosis probability value corresponding to each of the preset N disease identifiers;

[0015] For each of the preset N disease identifiers, the following diagnosis result generation operation is performed: in response to determining that the diagnosis probability value corresponding to the disease identifier is greater than a preset diagnosis probability threshold, generating a diagnosis result corresponding to the disease identifier using the diagnosis probability value corresponding to the disease identifier;

[0016] The generated diagnosis results are combined to obtain the diagnosis result information.

[0017] In some optional embodiments, the disease diagnosis model further includes a suspected diagnosis model; and

[0018] The step of inputting the historical medical consultation answer information sequence into a pre-trained disease diagnosis model to obtain diagnosis result information further includes:

[0019] Inputting the historical medical consultation answer information sequence into the suspected diagnosis model to obtain a suspected probability value corresponding to each of the preset N disease identifiers; and

[0020] The diagnostic result generating operation further includes:

[0021] In response to determining that the confirmed probability value corresponding to the disease identifier is not greater than the preset confirmed probability threshold, determining whether the suspected probability value corresponding to the disease identifier is greater than the preset suspected probability threshold;

[0022] In response to determining that it is greater than, a suspected diagnosis result corresponding to the disease identifier is generated using the suspected probability value corresponding to the disease identifier.

[0023] In some optional implementations, determining whether to continue the consultation according to the diagnosis result information includes:

[0024] Determine the number of different disease identifiers corresponding to each diagnosis result included in the diagnosis result information as the number of disease diagnosis results;

[0025] Determining whether the number of disease diagnosis results is less than a preset disease diagnosis result number threshold;

[0026] In response to determining that it is less than, determining to continue the consultation;

[0027] In response to determining not less than, it is determined not to continue the consultation.

[0028] In some optional implementations, sending the diagnosis result information to the client includes:

[0029] The first S diagnosis results with the highest corresponding confirmed probability values ​​and suspected probability values ​​among the confirmed diagnosis results and suspected diagnosis results in the diagnosis result information are sent to the client, where S is a positive integer.

[0030] In some optional implementations, the determining the current question information to be asked from the preset question information set based on the historical question answer information sequence and the diagnosis result information includes:

[0031] The historical medical consultation answer information sequence is input into a pre-trained medical question prediction model to obtain the current medical question information to be asked, wherein the medical question prediction model is used to characterize the correspondence between the medical consultation answer information sequence and the diagnosis result information and the medical question information in the preset medical question information set.

[0032] In a second aspect, the present disclosure provides an AI diagnosis method based on a consultation request, which is applied to a client, and the method includes: in response to receiving information about questions to be asked for a target patient identifier from a server, presenting the consultation questions and corresponding answer options in the information about questions to be asked; in response to detecting a selection operation for the presented answer options, generating a consultation reply request for the target patient identifier based on the target patient identifier, the information about questions to be asked, and the answer options targeted by the selection operation, and sending the consultation reply request to the server, so that the server determines whether to continue the consultation based on the consultation reply request, and in response to determining to continue the consultation, determining and returning the current information about questions to be asked from a preset consultation question information set.

[0033] In some optional embodiments, the method further comprises:

[0034] In response to detecting a consultation operation for the target patient identifier, a consultation request for the target patient identifier is generated and sent to the server.

[0035] In some optional embodiments, the method further comprises:

[0036] In response to receiving the diagnosis result information sent by the server, presenting the diagnosis result information.

[0037] In some optional implementations, the diagnosis result information includes a confirmed probability value or a suspected probability value corresponding to at least one disease marker among the preset N disease markers.

[0038] In a third aspect, the present disclosure provides an AI diagnostic device based on a consultation request, which is applied to a server, and the device includes: an answer option adding unit, which is configured to, in response to receiving a consultation reply request for a target patient identifier sent by a client, add the consultation question and the corresponding selected answer option in the consultation reply request to the historical consultation reply information sequence corresponding to the target patient identifier; a diagnosis unit, which is configured to input the historical consultation reply information sequence into a pre-trained disease diagnosis model to obtain diagnosis result information; a continue consultation determination unit, which is configured to determine whether to continue the consultation based on the diagnosis result information; a current question determination and sending unit, which is configured to, in response to determining to continue the consultation, determine the current question information to be consulted from the preset consultation question information set based on the historical consultation reply information sequence and the diagnosis result information, and send the current question information to the client, so that the client can present the current question information to be consulted, wherein the consultation question information includes the consultation question and corresponding at least two answer options.

[0039] In some optional embodiments, the device further comprises:

[0040] The first question determination and sending unit is configured to determine the first question information to be asked from the preset question information set in response to receiving a consultation request for the target patient identifier sent by the client, and send the first question information to be asked to the client.

[0041] In some optional embodiments, the device further comprises:

[0042] The diagnosis result sending unit is configured to send the diagnosis result information to the client in response to determining not to continue the consultation, so that the client can present the diagnosis result information.

[0043] In some optional implementations, each of the preset medical question information is associated with a question weight; and

[0044] The first problem determination and sending unit is further configured to:

[0045] The medical question information with the highest corresponding question weight in the preset medical question information set is determined as the first medical question information to be asked.

[0046] In some optional embodiments, the disease diagnosis model comprises a confirmed diagnosis model; and

[0047] The diagnostic unit is further configured to:

[0048] Input the historical medical consultation answer information sequence into the confirmed diagnosis model to obtain a confirmed diagnosis probability value corresponding to each of the preset N disease identifiers;

[0049] For each of the preset N disease identifiers, the following diagnosis result generation operation is performed: in response to determining that the diagnosis probability value corresponding to the disease identifier is greater than a preset diagnosis probability threshold, generating a diagnosis result corresponding to the disease identifier using the diagnosis probability value corresponding to the disease identifier;

[0050] The generated diagnosis results are combined to obtain the diagnosis result information.

[0051] In some optional embodiments, the disease diagnosis model further includes a suspected diagnosis model; and

[0052] The diagnostic unit is further configured to:

[0053] Inputting the historical medical consultation answer information sequence into the suspected diagnosis model to obtain a suspected probability value corresponding to each of the preset N disease identifiers; and

[0054] The diagnostic result generating operation further includes:

[0055] In response to determining that the confirmed probability value corresponding to the disease identifier is not greater than the preset confirmed probability threshold, determining whether the suspected probability value corresponding to the disease identifier is greater than the preset suspected probability threshold;

[0056] In response to determining that it is greater than, a suspected diagnosis result corresponding to the disease identifier is generated using the suspected probability value corresponding to the disease identifier.

[0057] In some optional embodiments, the continued medical consultation determination unit is further configured to:

[0058] Determine the number of different disease identifiers corresponding to each diagnosis result included in the diagnosis result information as the number of disease diagnosis results;

[0059] Determining whether the number of disease diagnosis results is less than a preset disease diagnosis result number threshold;

[0060] In response to determining that it is less than, determining to continue the consultation;

[0061] In response to determining not less than, it is determined not to continue the consultation.

[0062] In some optional implementations, the diagnosis result sending unit is further configured to:

[0063] The first S diagnosis results with the highest corresponding confirmed probability values ​​and suspected probability values ​​among the confirmed diagnosis results and suspected diagnosis results in the diagnosis result information are sent to the client, where S is a positive integer.

[0064] In some optional implementations, the current problem determination and sending unit is further configured to:

[0065] The historical medical consultation answer information sequence and the diagnosis result information are input into a pre-trained medical question prediction model to obtain the current medical question information to be asked, wherein the medical question prediction model is used to characterize the correspondence between the medical consultation answer information sequence and the diagnosis result information and the medical question information in the preset medical question information set.

[0066] In a fourth aspect, the present disclosure provides an AI diagnosis device based on a medical inquiry request, which is applied to a client, and the device includes:

[0067] A question and answer presenting unit, configured to present the question and corresponding answer options in the question information to be asked in response to receiving the question information to be asked for the target patient from the server;

[0068] The answer request sending unit is configured to generate a consultation answer request for the target patient identifier based on the target patient identifier, the question information to be asked and the answer option targeted by the selection operation in response to detecting a selection operation for the presented answer options, and to send the consultation answer request to the server so that the server determines whether to continue the consultation based on the consultation answer request, and in response to determining to continue the consultation, determines and returns the current question information to be asked from the preset consultation question information set.

[0069] In some optional embodiments, the device further comprises:

[0070] The consultation request sending unit is configured to generate and send a consultation request for the target patient identifier to the server in response to detecting a consultation operation for the target patient identifier.

[0071] In some optional embodiments, the device further comprises:

[0072] The diagnosis result presenting unit is configured to present the diagnosis result information in response to receiving the diagnosis result information sent by the server.

[0073] In some optional implementations, the diagnosis result information includes a confirmed probability value or a suspected probability value corresponding to at least one disease marker among the preset N disease markers.

[0074] In a fifth aspect, the present disclosure provides a server, comprising: one or more processors; a storage device, on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0075] In a sixth aspect, the present disclosure provides a client, comprising: one or more processors; a storage device, on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner in the second aspect.

[0076] In a seventh aspect, the present disclosure provides a medical consultation system, comprising a server as described in any implementation manner in the fifth aspect and at least one client as described in any implementation manner in the sixth aspect.

[0077] In an eighth aspect, the present disclosure provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by one or more processors, implements the method described in any implementation manner in the first aspect and / or the second aspect.

[0078] The AI ​​diagnosis method, device, storage medium and equipment based on the consultation request provided by the present disclosure, by presenting the questions to be asked and the corresponding answer options on the client, when the patient / medical staff makes a choice in the answer options for the questions to be asked presented on the client, the server then appends the answer options selected by the patient / medical staff on the client to the questions to be asked to the consultation to the consultation questions and the corresponding selected answers that the target patient has answered in history, that is, a historical consultation reply information sequence is obtained. Then the historical consultation reply information sequence is input into the pre-trained disease diagnosis model to obtain the diagnosis result information. Then, determine whether to continue the consultation based on the diagnosis result information. If it is determined to continue the consultation, the current question information to be asked is determined from the preset consultation question information set based on the historical consultation reply information sequence and the diagnosis result information, and the current question information to be asked is sent to the client. Then, the client can continue to present the current question to be asked and the corresponding answer options for the patient / medical staff to answer. The following technical effects can be achieved, including but not limited to:

[0079] First, during the consultation process, patients / medical staff only need to choose from the answer options corresponding to the questions provided without having to input any information, which reduces the difficulty and complexity of the operation for patients / medical staff to provide disease description information and makes it easier to use.

[0080] Second, during the consultation process, the server determines in real time whether the consultation needs to continue each time it receives the consultation reply information sent by the client. Only after it is determined to continue the consultation will it continue to provide consultation questions. There is no need to provide all the consultation questions in the preset consultation question information set to the patient / medical staff for answers, which can reduce the number of consultation questions answered by patients / medical staff and improve the consultation speed.

[0081] Third, the next question to be asked is dynamically determined based on the historical medical consultation answer information sequence that has been answered for the target patient's condition, rather than providing the question information in a preset fixed order. The historical medical consultation answer information sequence that has been answered for the target patient's condition can be dynamically adapted, making the consultation process more targeted. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Other features, objects and advantages of the present disclosure will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0083] Figure 1 is an exemplary system architecture diagram in which an embodiment of the present disclosure may be applied;

[0084] Figure 2A is a timing diagram of an embodiment of an AI diagnosis system based on a medical inquiry request according to the present disclosure;

[0085] Figure 2B is a decomposed flow chart of one embodiment of step 204 according to the present disclosure;

[0086] Figure 2C is a flow chart of one embodiment of a diagnostic result generating operation according to the present disclosure;

[0087] Figure 3 is a flow chart of one embodiment of the training steps according to the present disclosure;

[0088] Figure 4 is a flowchart of an embodiment of an AI diagnosis method based on a medical inquiry request applied to a server according to the present disclosure;

[0089] Figure 5 is a flowchart of an embodiment of an AI diagnosis method based on a medical inquiry request applied to a client according to the present disclosure;

[0090] Figure 6is a structural diagram of an embodiment of an AI diagnosis device based on a medical inquiry request and applied to a server according to the present disclosure;

[0091] Figure 7 is a structural diagram of an embodiment of an AI diagnosis device based on a medical inquiry request applied to a client according to the present disclosure;

[0092] Figure 8 It is a schematic diagram of the structure of a computer system of a client or server suitable for implementing the embodiments of the present disclosure. DETAILED DESCRIPTION

[0093] The present disclosure is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0094] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0095] Figure 1 An exemplary system architecture 100 is shown to which an embodiment of an AI diagnosis method based on a medical inquiry request or an AI diagnosis device based on a medical inquiry request of the present disclosure can be applied.

[0096] like Figure 1 As shown, the system architecture 100 may include clients 101, 102, 103, a network 104, and a server 105. The network 104 is used to provide a medium for communication links between the clients 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or optical fiber cables.

[0097] Users can use clients 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on clients 101, 102, 103, such as remote consultation applications, web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0098] Clients 101, 102, 103 can be hardware or software. When clients 101, 102, 103 are hardware, they can be various electronic devices with display screens, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, Moving Picture Experts Group Audio Layer 3), MP4 (Moving Picture Experts Group Audio Layer IV, Moving Picture Experts Group Audio Layer 4) players, laptop computers and desktop computers, etc. When clients 101, 102, 103 are software, they can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules (for example, to provide remote consultation services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0099] The server 105 may be a server that provides various services, such as a backend server that provides support for remote consultation applications or web pages that provide remote consultation services displayed on the clients 101, 102, and 103. The backend server may analyze and process the received consultation request and other data, and feed back the processing results (such as information on questions to be consulted) to the client.

[0100] It should be noted that the AI ​​diagnosis method based on the inquiry request applied to the server provided in the present disclosure is generally executed by the server 105, and accordingly, the AI ​​diagnosis device based on the inquiry request applied to the server is generally set in the server 105. The AI ​​diagnosis method based on the inquiry request applied to the client provided in the present disclosure is generally executed by the client 101, 102, 103, and accordingly, the AI ​​diagnosis device based on the inquiry request applied to the client is generally set in the client 101, 102, 103.

[0101] It should be noted that the server 105 can be hardware or software. When the server 105 is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server 105 is software, it can be implemented as multiple software or software modules (for example, for providing distributed services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0102] It should be understood that Figure 1 The number of clients, networks and servers in the embodiment is only for illustration. Any number of clients, networks and servers may be provided as required.

[0103] Continue to refer Figure 2A, which shows a timing sequence 200 of an embodiment of an AI diagnosis system based on a medical inquiry request according to the present disclosure. The AI ​​diagnosis system based on a medical inquiry request in the embodiment of the present disclosure may include a client and a server. The timing sequence 200 includes the following steps:

[0104] Step 201 , in response to receiving information of questions to be asked for a target patient identification from a server, the client presents the questions to be asked and corresponding answer options in the information of questions to be asked.

[0105] In this embodiment, the client (eg Figure 1 The client 101, 102, 103 shown in the figure can present the diagnosis questions and corresponding answer options in the diagnosis questions information when receiving the diagnosis questions information for the target patient identification from the server.

[0106] Here, the client may be installed with a remote consultation application, and the server may be a server that provides support for the remote consultation application. Alternatively, the client may be installed with a web browser application, and the user may use the web browser application to access a web page that provides a remote consultation service, and the server may be a server that provides support for the web page that provides a remote consultation service.

[0107] The target patient identifier can be a patient identifier that the user enters and verifies when logging in currently or logging in historically and choosing to remember the login information, using a remote consultation application installed on the client or accessing a remote consultation web page. The patient identifier is used to uniquely distinguish each registered or temporarily accessed patient. The patient identifier can take various forms, and the present disclosure does not specifically limit this. For example, it can include a combination of at least one of the following: text, numbers, and characters. For example, the patient identifier can be the patient's name, ID number, mobile phone number, or a combination of the above. The target patient identifier is used to indicate the target patient. Here, the target patient identifier only exemplifies a specific patient identifier and is not limited to a specified patient identifier.

[0108] The information of questions to be asked may include questions and corresponding answer options. Among them, the questions may be questions related to symptoms. For example: whether a specific disease symptom appears, the degree, time, frequency, etc. of a specific symptom, whether a specific examination or test has been done, and what is the value range of the specific test result of a specific test or examination. Correspondingly, the answer options corresponding to the questions are also related to the questions. For example, when the question is a yes or no question, the corresponding answer options may include: yes and no. When the question is about the degree, time, frequency, etc. of a specific symptom, the corresponding answer options may be preset degree descriptors, time length, frequency range, etc.

[0109] The client can also present the information of the question to be asked in various ways. The present disclosure does not make specific restrictions on this. For example, when the question is a yes or no question, the corresponding answer options of yes and no can be presented by radio buttons combined with corresponding answer option text or icons. When the question is about the degree, time, frequency, etc. of specific symptoms, and the number of corresponding answer options is greater than or equal to 3, check buttons can be used to combine the corresponding answer option text or icons to present.

[0110] Step 202, in response to detecting a selection operation for the presented answer options, the client generates a consultation answer request for the target patient identifier based on the target patient identifier, the information of the question to be asked, and the answer options for which the selection operation is performed, and sends the consultation answer request to the server.

[0111] As can be seen from the description of step 201, the client has presented the questions to be asked and the corresponding answer options. The user (for example, the patient / medical staff) can choose from the presented answer options based on the actual situation of the target patient for the current question to be asked, and the client can detect the selection operation after the user selects the answer option. The client can also detect the selection operation after the user selects the answer option and performs a first preset operation (for example, single-clicking, hovering for a preset time, dragging, double-clicking, etc.) on the first display object used to indicate the submission of the consultation reply information (for example, text, button or icon used to indicate the submission of the consultation reply information), and then generate a consultation reply request.

[0112] Here, the client can use various implementation methods to generate a consultation reply request. For example, the consultation reply request can carry the target patient identifier, the question information to be asked, and the answer options selected by the user in the client. For another example, the consultation reply request can also carry the target patient identifier, the consultation question identifier corresponding to the question information to be asked, and the answer options selected by the user in the client. Among them, the question identifier corresponding to the question information to be asked can be a part of the question information to be asked, and the question information to be asked sent by the server to the client already carries the question identifier, so the client can obtain the above-mentioned question identifier and carry the above-mentioned question identifier when generating a consultation reply request. The question identifier occupies less storage space than the question information to be asked, which can reduce the amount of data sent between the client and the server, reduce network bandwidth requirements, and increase the interaction speed between the client and the server.

[0113] Step 203, in response to receiving the consultation reply request for the target patient identifier sent by the client, the server adds the consultation question and the corresponding selected answer options in the consultation reply request to the historical consultation reply information sequence corresponding to the target patient identifier.

[0114] Here, the historical medical consultation answer information sequence corresponding to the target patient identification can be stored locally on the server, or stored in other electronic devices connected to the server network. The historical medical consultation answer information sequence corresponding to the target patient identification can be a medical consultation answer information sequence composed of medical consultation questions that have been answered / answered in the current medical consultation process according to the actual situation of the target patient and the corresponding selected answer options in the order of answering / answering. The medical consultation answer information includes the medical consultation questions and the corresponding selected answer options.

[0115] After receiving the consultation reply request for the target patient identifier sent by the client to the server in step 202, the server may receive the consultation reply request, parse the consultation reply request, obtain the consultation question and the corresponding selected answer option, and then add the parsed consultation question and the corresponding selected answer option to the historical consultation reply information sequence corresponding to the target patient identifier. Furthermore, the historical consultation reply information sequence corresponding to the target patient identifier stores the consultation reply information sequence composed of all the consultation questions and the corresponding selected answer options that have been answered / answered in the current consultation process according to the actual situation of the target patient until the current moment, in the order of answering / answering.

[0116] Step 204: The server inputs the historical medical consultation response information sequence into a pre-trained disease diagnosis model to obtain diagnosis result information.

[0117] In this embodiment, the server can input the historical medical consultation reply information sequence obtained in step 203 into the pre-trained disease diagnosis model to obtain the diagnosis result information for the target patient. Here, the disease diagnosis model can be used to characterize the correspondence between the medical consultation reply information sequence and the diagnosis result information. The diagnosis result information can be various forms of disease information used to characterize the patient and one or at least one disease. For example, it can be the corresponding probability values ​​of whether or not the patient has which diseases or which diseases have been diagnosed, whether or not the patient is suspected of having which diseases or which diseases are suspected of having which diseases, etc.

[0118] As an example, a disease diagnosis model can be a logical judgment rule or a calculation formula that performs corresponding logical judgments on the consultation answer information sequence to obtain the diagnosis result information, or a calculation formula that performs calculations to obtain the diagnosis result information, which is pre-established by a technician with medical knowledge based on a large amount of practice and statistical analysis of the sample consultation answer information sequence extracted from the patient's medical records and the corresponding annotated diagnosis result information of the patient.

[0119] In some optional implementations, the disease diagnosis model may also be pre-trained through a training step.

[0120] Please refer to Figure 3, which shows a process 300 of an embodiment of the training step according to the present disclosure, such as Figure 3 The training step 300 shown may include the following steps 301 to 304:

[0121] Step 301, determining the model structure of the initial disease diagnosis model and initializing the model parameters of the initial disease diagnosis model.

[0122] Here, the execution subject of the training step may be the above-mentioned server, or may be other electronic devices. If the execution subject of the training step is the above-mentioned server, the execution subject of the training step may store the model structure information of the trained disease diagnosis model and the parameter values ​​of the model parameters locally on the above-mentioned server after the disease diagnosis model is obtained through training. If the execution subject of the training step is not the above-mentioned server, the execution subject of the training step may send the model structure information of the trained disease diagnosis model and the parameter values ​​of the model parameters to the above-mentioned server after the disease diagnosis model is obtained through training.

[0123] Here, since the initial disease diagnosis model may include various types of computing models, the model structure information that needs to be determined for different types of computing models is also correspondingly different.

[0124] Optionally, the initial disease diagnosis model can be various machine learning models. Optionally, it can be an artificial neural network model, such as a deep neural network, or a convolutional neural network, or a recurrent neural network (RNN) suitable for processing time series data. For example, when the initial disease diagnosis model is a two-dimensional convolutional neural network, since the convolutional neural network is a multi-layer neural network, each layer is composed of multiple two-dimensional planes, and each plane is composed of multiple independent neurons, it is necessary to determine which layers the initial disease diagnosis model of the convolutional neural network type includes, such as the input layer, hidden layer and output layer, and which parameters each layer includes. For example, it can be determined which layers the hidden layer includes (for example, convolution layer, pooling layer, activation function layer, etc.), the connection order relationship between layers, and which parameters each layer includes (for example, weight weight, bias, convolution step length), etc. Among them, the convolution layer can be used to extract features. For each convolution layer, it is possible to determine how many convolution kernels there are, the size of each convolution kernel, the weights of each neuron in each convolution kernel, the bias item corresponding to each convolution kernel, the step size between two adjacent convolutions, whether padding is required, how many pixels to pad and the value used for padding (generally filled with 0), etc. The pooling layer can be used to downsample the input information to compress the amount of data and parameters and reduce overfitting. For each pooling layer, the pooling method of the pooling layer can be determined (for example, taking the regional average or taking the regional maximum). The excitation function layer is used to perform nonlinear calculations on the input information. For each excitation function layer, a specific excitation function can be determined. For example, the activation function can be ReLU and various variant activation functions of ReLU, Sigmoid function, Tanh (hyperbolic tangent) function, Maxout function, etc. For another example, it can be determined whether the output layer includes a classifier for implementing classification and a regressor for calculating values, the specific implementation algorithm and parameters of the classifier, and the specific implementation algorithm and parameters of the regressor.

[0125] Then, the model parameters of the initial disease diagnosis model can be initialized. In practice, the various model parameters of the initial disease diagnosis model can be initialized with some different small random numbers. "Small random numbers" are used to ensure that the model will not enter a saturated state due to excessive weights, thereby causing training failure, and "different" is used to ensure that the model can learn normally.

[0126] Step 302: determine a training sample set.

[0127] Here, the training samples in the training sample set may include a sample consultation answer information sequence extracted from a patient's medical record in practice and the corresponding annotated diagnosis result information of the patient.

[0128] It should be noted that the execution subject of the training step may first execute step 301 and then execute step 302, or may first execute step 302 and then execute step 301, and the present disclosure does not make any specific limitation on this.

[0129] In some implementations, the annotated diagnostic result information can be used to characterize, including but not limited to, whether the patient is diagnosed with a certain disease or a corresponding probability value, is excluded from being diagnosed with a certain disease or a corresponding probability value, is suspected of having a certain disease or a corresponding probability value, or is excluded from being suspected of having a certain disease or a corresponding probability value.

[0130] Step 303: Use the sample question answer information sequence in the training samples in the training sample set and the corresponding annotated diagnosis result information of the patient as the input and expected output of the initial disease diagnosis model respectively, and use the machine learning method to train the initial disease diagnosis model.

[0131] Specifically, the sample consultation reply information sequence in the training sample in the training sample set can first be input into the initial disease diagnosis model to obtain the corresponding diagnosis result information of the sample consultation reply information sequence. Then, the difference between the obtained diagnosis result information and the labeled diagnosis result information in the training sample can be calculated. Finally, based on the calculated difference, the model parameters of the initial disease diagnosis model can be adjusted, and the training can be terminated when the preset training termination conditions are met. For example, the preset training termination conditions here may include at least one of the following: the training time exceeds the preset duration, the number of training times exceeds the preset number of times, and the calculated difference is less than the preset difference threshold.

[0132] Here, various implementations may be used to adjust the model parameters of the initial disease diagnosis model based on the difference between the obtained diagnosis result information and the labeled diagnosis result information in the training sample. For example, Stochastic Gradient Descent (SGD), Newton's Method, Quasi-Newton Methods, Conjugate Gradient, heuristic optimization methods, and other optimization algorithms currently known or to be developed in the future may be used.

[0133] Step 304: determine the initial disease diagnosis model obtained through training as the pre-trained disease diagnosis model.

[0134] That is, the disease diagnosis model obtained according to the training steps from step 301 to step 304 can be based on a large number of training samples and use machine learning methods to automatically learn the rules in the question-answer information sequence and obtain the diagnosis results. If the training samples are generated based on diagnosis cases of doctors with a high level of practical Chinese medicine skills, the automatically learned disease diagnosis model will also have a correspondingly higher accuracy.

[0135] Step 205: The server determines whether to continue the consultation based on the diagnosis result information.

[0136] In this embodiment, the server can determine whether to continue the consultation according to the preset continued consultation determination rule and the diagnosis result information obtained in step 204. If it is determined to continue the consultation, the server goes to step 206 to continue the execution.

[0137] Here, the preset rules for determining continued consultation can be pre-formulated and stored in the server by the technician based on the various data included in the diagnosis result information and the meaning represented by the data. The rules for determining continued consultation are used to characterize whether a relatively definite diagnosis conclusion can be obtained based on the diagnosis result information, and there is no need to continue asking other consultation questions. If there is a relatively definite diagnosis conclusion, there is no need to continue the consultation; conversely, if there is no relatively definite diagnosis conclusion, it is necessary to continue asking other questions to achieve a relatively definite diagnosis conclusion. Therefore, according to the pre-set rules for determining continued consultation, it can be achieved that only when it is determined to continue the consultation, step 206 will be executed and the consultation question information will be further provided to the client, which can reduce the number of questions asked to the user.

[0138] Step 206 , in response to determining to continue the consultation, the server determines the current question information to be consulted from the preset consultation question information set based on the historical consultation answer information sequence and the diagnosis result information, and sends the current question information to be consulted to the client.

[0139] Here, the preset medical question information set includes medical questions and corresponding answer options. The preset medical question information set may be a medical question information set formulated by professional technicians with medical expertise and stored in a server.

[0140] In this embodiment, if the server determines to continue the consultation in step 205, then in step 206 the server can adopt various implementation methods to determine the current question information to be consulted from the preset consultation question information set based on the historical consultation reply information sequence corresponding to the target patient identifier and the diagnosis result information obtained in step 204, and send the determined current question information to be consulted to the client.

[0141] In some optional embodiments, the server may input the historical consultation reply information sequence corresponding to the target patient identifier and the diagnosis result information obtained in step 204 into a pre-trained consultation question prediction model to obtain the current question information to be asked. The consultation question prediction model is used to characterize the correspondence between the consultation reply information sequence and the diagnosis result information and the consultation question information in the preset consultation question information set. The consultation question prediction model may be a model trained based on machine learning, deep learning or transfer learning methods, and the corresponding training sample set may include a sample consultation reply information sequence extracted from the patient's medical records in practice and a corresponding annotated diagnosis result information obtained by a professional technician with professional medical knowledge according to clinical consultation practice based on the sample consultation reply information sequence, and the corresponding annotated predicted consultation question information for characterizing the next question to be asked. As an example, the consultation question prediction model may be a decision tree model. The decision tree model may include M nodes, and the nodes in the M nodes and the edges of the nodes pointing to other nodes correspond one-to-one to the consultation questions and corresponding answer options in each consultation question information in the preset consultation question information set. By adopting this optional implementation mode, a pre-trained medical question prediction model can be used to automatically determine the next question to be asked based on the historical medical question response information sequence and diagnosis result information of the target patient. In practice, the medical question prediction model can also be updated in real time to improve the intelligence level of determining the next question to be asked, and can be updated in real time based on practical medical experience.

[0142] It should be noted that, since the corresponding medical questions in the historical medical consultation response information sequence have been answered according to the actual situation of the target patient and the corresponding diagnosis result information has been obtained, the medical questions corresponding to the current question information to be asked determined in the preset medical consultation question information set should not belong to the medical questions included in the historical medical consultation response information sequence, that is, the medical questions that have been asked will no longer be asked.

[0143] After step 206, the client can receive the current question information to be asked for the target patient identification, and then continue to execute step 201, that is, the client responds to receiving the current question information to be asked for the target patient identification from the server, and presents the question to be asked and the corresponding answer options in the current question information to be asked.

[0144] Through step 201 to step 206, the following technical effects can be achieved, including but not limited to:

[0145] First, during the consultation process, patients / medical staff only need to choose from the answer options corresponding to the questions provided without having to input any information, which reduces the difficulty and complexity of the patient / medical staff in providing disease description information and makes it easier to use.

[0146] Second, during the consultation process, the server determines in real time whether the consultation needs to continue each time it receives the consultation reply information sent by the client. Only after it is determined to continue the consultation will it continue to provide consultation questions. There is no need to provide all the consultation questions in the preset consultation question information set to the patient / medical staff for answers, which can reduce the number of consultation questions answered by patients / medical staff and improve the consultation speed.

[0147] Third, the next question to be asked is dynamically determined based on the historical medical consultation answer information sequence that has been answered for the target patient's condition, rather than providing the question information in a preset fixed order. The historical medical consultation answer information sequence that has been answered for the target patient's condition can be dynamically adapted, making the consultation process more targeted.

[0148] In some cases, this embodiment may have the following optional implementations:

[0149] Optional implementation (I): Sequence 200 may further include the following steps 207 and 208:

[0150] Step 207 : In response to detecting the consultation operation for the target patient identifier, the client generates and sends a consultation request for the target patient identifier to the server.

[0151] Here, the consultation operation can be triggered in various situations, and the client can detect the consultation operation for the target patient identifier. For example, the user can use the remote consultation application installed on the client or visit the remote consultation web page, and enter the target patient identifier when logging in currently or logging in in the past and choosing to remember the login information, and after the verification is passed, the consultation operation can be triggered. For another example, after the above verification is passed, a second display object for indicating the submission of a consultation request (for example, text, button or icon for indicating the submission of a consultation request, etc.) can be presented on the client, and then the client can detect the consultation operation after detecting that the user has performed a second preset operation (for example, single-click, hover for a preset time, drag, double-click, etc.) on the above second display object, and then generate and send a consultation request for the target patient identifier to the server.

[0152] Step 208 , in response to receiving the consultation request for the target patient identifier sent by the client, the server determines the first question information to be consulted from the preset consultation question information set, and sends the first question information to be consulted to the client.

[0153] Here, the server may adopt various implementation methods to determine the first question information to be asked from the preset medical question information set, and send the determined first question information to be asked to the client.

[0154] For example, the server may randomly determine a piece of medical question information from a preset medical question information set as the first piece of medical question information to be asked.

[0155] For another example, the server may also determine a specified medical question information in the preset medical question information set as the first medical question information to be asked.

[0156] Optional implementation (two): Based on the above optional implementation (one), each preset medical question information in the preset medical question information set may be associated with a corresponding question weight. Accordingly, in step 208, determining the first question information to be asked from the preset medical question information set may be performed as follows: determining the medical question information with the highest corresponding question weight in the preset medical question information set as the first question information to be asked. The question weight associated with each preset medical question information in the preset medical question information set is used to characterize the importance of the medical questions in the medical question information in the process of diagnosing different diseases. As an example, the question weight associated with each preset medical question information in the preset medical question information set may be formulated by a professional technician with medical knowledge and stored in a server. For another example, the question weight associated with each preset medical question information in the preset medical question information set may also be obtained based on training of a machine learning model.

[0157] Specifically, for example, the model structure and model parameter information of the artificial neural network can be determined first, and the artificial neural network can include the first weight component weight coefficient, the second weight component weight coefficient, the third weight component weight coefficient and the fourth weight component weight coefficient of each preset medical question information. The first weight component weight coefficient, the second weight component weight coefficient, the third weight component weight coefficient and the fourth weight component weight coefficient correspond to the patient's medical record information, the knowledge and experience of medical experts, professional medical books and professional medical literature respectively.

[0158] Next, the weight coefficient of the first weight component, the weight coefficient of the second weight component, the weight coefficient of the third weight component and the weight coefficient of the fourth weight component may be initialized and set.

[0159] Then, the training sample is input into the artificial neural network to obtain the actual output. The training sample may include sample data and annotation information. The sample data may include patient medical records, medical expert knowledge and experience, professional medical books and professional medical literature, and the annotation information may include: the annotation weight coefficient of the patient medical records, medical expert knowledge and experience, professional medical books and professional medical literature on the information of the medical questions involved in the sample data. Based on the difference between the corresponding four weight coefficients of the actual output and the four corresponding weight coefficients included in the annotation information in the training sample, the network parameters of the artificial neural network are adjusted.

[0160] After the training is completed, the first weight component weight coefficient, the second weight component weight coefficient, the third weight component weight coefficient and the fourth weight component weight coefficient corresponding to each medical question in the preset medical question set in the above artificial neural network can be obtained.

[0161] Finally, for each medical question information in the preset medical question information set, the corresponding first weight component can be set manually according to a large amount of patient medical record information, the corresponding initial second weight component can be set according to the knowledge and experience of medical experts, the corresponding initial third weight component can be set according to the relevant content in professional medical books, and the corresponding initial fourth weight component can be set according to professional medical literature. Then, the first weight component, second weight component, third weight component and fourth weight component weight coefficient corresponding to each medical question can be weighted and summed to obtain the weight associated with each medical question information.

[0162] By adopting this optional implementation mode, when providing the first question to be asked to the client, the question with the highest question weight can be selected to achieve rapid diagnosis.

[0163] Optional implementation (three): Sequence 200 may further include the following steps 209 and 210:

[0164] Step 209: In response to determining not to continue the consultation, the server sends the diagnosis result information to the client.

[0165] Here, if the server determines in step 205 that it does not need to continue the consultation, it can indicate that a relatively certain diagnosis conclusion can be reached based on the diagnosis result information and there is no need to continue the consultation. Therefore, the diagnosis result information can be sent to the client, that is, there is no need to ask the clinic any questions, thereby reducing the number of questions asked to the target patient.

[0166] Step 210: the client presents the diagnosis result information in response to receiving the diagnosis result information sent by the server.

[0167] Here, the client may present the received diagnosis result information in various presentation modes, for example, the diagnosis result information may be presented in the form of text, pictures or voice.

[0168] After step 209 and step 210, it is possible to stop the questioning when the diagnosis result information can have a relatively certain diagnosis conclusion, and present the relatively certain diagnosis result information to the client for reference by the user (patient / medical staff). Compared with directly providing the diagnosis result information to the user for reference without making any judgment on the diagnosis result information, the availability and referenceability of the provided diagnosis result information can be improved.

[0169] Optional implementation (IV): The disease diagnosis model may include a confirmed diagnosis model. Accordingly, in step 204, the server inputs the historical consultation answer information sequence into the pre-trained disease diagnosis model to obtain the diagnosis result information, which may include: Figure 2B Steps 2041 to 2043 shown:

[0170] Step 2041, input the historical medical consultation reply information sequence into the diagnosis model to obtain the diagnosis probability value corresponding to each of the preset N disease identifiers.

[0171] The confirmed diagnosis model is used to characterize the corresponding relationship between the inquiry response information sequence and the disease confirmed probability value corresponding to each disease identifier in the preset N disease identifiers. Here, N is a positive integer.

[0172] As an example, a confirmed diagnosis model can be a calculation formula that is pre-established by a technician with medical knowledge based on a large amount of practice, based on statistical analysis of sample consultation answer information sequences extracted from patient medical records and the corresponding labeled confirmed probability values ​​of the patient being diagnosed with the disease indicated by each of the above-mentioned N preset disease identifiers. The formula is used to calculate the consultation answer information sequence and obtain the confirmed probability value.

[0173] As another example, the confirmed diagnosis model can also be a machine learning model obtained by training using a machine learning method based on a large number of sample consultation response information sequences extracted from patient medical records in practice and the corresponding labeled confirmed probability values ​​of the patient suffering from the disease indicated by each of the above-mentioned N preset disease identifiers.

[0174] Here, the diagnosis probability value corresponding to each disease identifier is used to represent the probability value of being diagnosed with the disease indicated by the disease identifier. The higher the diagnosis probability value corresponding to the disease identifier, the greater the possibility of being diagnosed with the disease indicated by the disease identifier.

[0175] Step 2042: For each of the preset N disease identifiers, a diagnosis result generation operation is performed.

[0176] Here, the diagnostic result generation operation may include: Figure 2C Step 20421 and step 20422 shown:

[0177] Step 20421, determine whether the diagnosis probability value corresponding to the disease identifier is greater than a preset diagnosis probability threshold.

[0178] If it is determined to be greater than, it indicates that the probability value of the target patient suffering from the disease indicated by the disease identifier is higher, that is, the diagnosis conclusion that the target patient suffers from the disease indicated by the disease identifier is relatively certain, and you can go to step 20422.

[0179] Here, the preset diagnosis probability threshold may be a different preset diagnosis probability threshold associated with each disease identifier, or may be a same preset diagnosis probability threshold associated with all disease identifiers in the preset N disease identifiers.

[0180] When the preset diagnosis probability threshold is associated with a different preset diagnosis probability threshold for each disease identifier, the preset diagnosis probability threshold can be a diagnosis probability threshold manually set by technical personnel based on practical experience for the disease identifier and can be updated at any time, or the preset diagnosis probability threshold can also be determined during the training process of the diagnosis model corresponding to the disease identifier.

[0181] When the preset diagnosis probability threshold is the same preset diagnosis probability threshold associated with all disease identifiers, the preset diagnosis probability threshold can be a diagnosis probability threshold manually set by technical personnel based on practical experience for all disease identifiers in the preset N disease identifiers and can be updated at any time, or the preset diagnosis probability threshold can also be determined during the training process of the diagnosis diagnosis model corresponding to each disease identifier in the preset N disease identifiers.

[0182] Step 20422, using the confirmed probability value corresponding to the disease identifier to generate a confirmed diagnosis result corresponding to the disease identifier.

[0183] Here, the server can generate a confirmed diagnosis result corresponding to the disease identifier using the confirmed probability value corresponding to the disease identifier when the probability value that the target patient suffers from the disease indicated by the disease identifier is high, that is, the diagnostic conclusion that the target patient suffers from the disease indicated by the disease identifier is relatively certain.

[0184] Step 2043, merging the generated diagnosis results to obtain diagnosis result information.

[0185] Here, after executing step 2042, the server may merge the various diagnosis results generated in step 2042 to obtain the diagnosis result information. It is understandable that in step 2042, a corresponding diagnosis result is not necessarily generated for each of the preset N disease identifiers. Only when a corresponding diagnosis result is generated for a certain disease identifier in step 2042, the diagnosis result corresponding to the disease identifier will be merged into the diagnosis result information.

[0186] By adopting this optional implementation, the design of N preset disease identifiers can be carried out in a targeted manner. For example, in practice, patients usually seek medical treatment only after they have corresponding symptoms, and they will choose the corresponding department when they seek medical treatment. Different N disease identifiers can be designed for different departments. For example, 9 disease identifiers for difficult and complicated diseases can be designed for the cardiovascular and cerebrovascular department, and 8 disease identifiers for difficult and complicated diseases can be designed for the obstetrics and gynecology department. Accordingly, when designing the confirmed diagnosis model corresponding to each disease identifier in the N disease identifiers, it can also be designed accordingly.

[0187] Optional implementation (V): The disease diagnosis model may also include a suspected diagnosis model. Accordingly, after executing step 2041, the server may also execute the following step 2041':

[0188] Step 2041', input the historical medical consultation reply information sequence into the suspected diagnosis model to obtain the suspected probability value corresponding to each disease identifier in the preset N disease identifiers.

[0189] Here, the suspected diagnosis model is used to characterize the correspondence between the inquiry response information sequence and the suspected probability value corresponding to each disease identifier in the preset N disease identifiers. Here, N is a positive integer. The suspected probability value corresponding to the disease identifier is used to characterize the probability value of suspected suffering from the disease indicated by the disease identifier.

[0190] As an example, a suspected diagnosis model can be a calculation formula that is pre-established by a technician with medical knowledge based on a large amount of practice, based on statistical analysis of sample consultation answer information sequences extracted from patient medical records and the corresponding labeled suspected probability values ​​of the patient suspected of suffering from the disease indicated by each of the above-mentioned N preset disease identifiers. The formula calculates the consultation answer information sequence and obtains the suspected probability value.

[0191] As another example, the suspected diagnosis model can also be a machine learning model obtained by training using a machine learning method based on a large number of sample consultation response information sequences extracted from patient medical records in practice and the corresponding labeled suspected probability values ​​of the patient suspected of suffering from the disease indicated by each of the above-mentioned N preset disease identifiers.

[0192] Here, the suspected probability value corresponding to each disease identifier is used to represent the probability value of suspected suffering from the disease indicated by the disease identifier. The higher the suspected probability value corresponding to the disease identifier, the greater the possibility of suspected suffering from the disease indicated by the disease identifier.

[0193] Based on the above step 2041', in the diagnosis result generation operation, if it is determined in step 20421 that the confirmed probability value corresponding to the disease identifier is not greater than the preset confirmed probability threshold, it indicates that the probability value of the target patient being diagnosed with the disease indicated by the disease identifier is low, that is, the diagnostic conclusion that the target patient is diagnosed with the disease indicated by the disease identifier is relatively uncertain, and the operation can proceed to step 20423 to execute and continue with other judgment logic.

[0194] Step 20423, determine whether the suspected probability value corresponding to the disease identifier is greater than a preset suspected probability threshold.

[0195] If it is determined to be greater than, it indicates that the probability value of the target patient suspected of suffering from the disease indicated by the disease identifier is higher, that is, the diagnostic conclusion that the target patient is suspected of suffering from the disease indicated by the disease identifier is relatively certain, and you can go to step 20424.

[0196] Here, the preset suspected probability threshold may be a different preset suspected probability threshold associated with each disease identifier, or may be a same preset suspected probability threshold associated with all disease identifiers in the preset N disease identifiers.

[0197] When the preset suspected probability threshold is a different preset suspected probability threshold associated with each disease identifier, the preset suspected probability threshold can be a suspected probability threshold manually set by technical personnel based on practical experience for the disease identifier and can be updated at any time, or the preset suspected probability threshold can also be determined during the training process of the suspected diagnostic model corresponding to the disease identifier.

[0198] When the preset suspected probability threshold is the same preset suspected probability threshold associated with all disease identifiers, the preset suspected probability threshold can be a suspected probability threshold manually set by technical personnel based on practical experience for all disease identifiers in the preset N disease identifiers and can be updated at any time, or the preset suspected probability threshold can also be determined during the training process of the suspected diagnostic model corresponding to each disease identifier in the preset N disease identifiers.

[0199] Step 20424, using the suspected probability value corresponding to the disease identifier to generate a suspected diagnosis result corresponding to the disease identifier.

[0200] Here, the server can generate a suspected diagnosis result corresponding to the disease identifier using the suspected probability value corresponding to the disease identifier when the probability value of the target patient suspected of having the disease indicated by the disease identifier is high, that is, the diagnosis conclusion that the target patient is suspected of having the disease indicated by the disease identifier is relatively certain. Furthermore, in step 2043, the suspected diagnosis result corresponding to the disease identifier can be merged into the diagnosis result information.

[0201] Through this optional implementation, when it is not certain that the target patient is confirmed to have a certain disease, it is possible to continue to determine whether the target patient is suspected of having the disease, and when it is determined with a high probability that the target patient is suspected of having the disease, a suspected confirmed diagnosis result is generated, rather than just generating a confirmed probability value. The suspected confirmed diagnosis result is also of reference value to patients / medical staff.

[0202] It should be noted that in the diagnosis result generation operation, when it is determined in step 20421 that the confirmed probability value corresponding to the disease identifier is not greater than the preset confirmed probability threshold, and when it is determined in step 20423 that the suspected probability value corresponding to the disease identifier is not greater than the preset suspected probability threshold, that is, the diagnostic conclusions for the target patient being confirmed to have the disease indicated by the disease identifier and the target patient being suspected to have the disease indicated by the disease identifier are both relatively uncertain, that is, as of now, no relatively definite conclusion can be reached based on the historical medical consultation response information sequence of the target patient, and it is also impossible to generate a corresponding diagnosis result for the disease identifier, and the diagnosis result generation operation can continue to be performed on other disease identifiers among the preset N disease identifiers for which the diagnosis result generation operation has not been performed.

[0203] Optional implementation (six): Based on the above optional implementation (fifth), the server in step 205 determines whether to continue the consultation according to the diagnosis result information as follows:

[0204] First, the number of different disease identifiers corresponding to each diagnosis result included in the diagnosis result information is determined as the number of disease diagnosis results.

[0205] If a confirmed diagnosis result or a suspected diagnosis result is generated for a disease identifier among the preset N disease identifiers in step 2042, the diagnosis result information includes the confirmed diagnosis result or the suspected diagnosis result corresponding to the disease identifier. Conversely, if no diagnosis result is generated for a disease identifier among the preset N disease identifiers in step 2042, the diagnosis result information also does not include any diagnosis result corresponding to the disease identifier. Therefore, the number of different disease identifiers corresponding to each diagnosis result included in the diagnosis result information is the number of disease identifiers among the preset N disease identifiers for which one of the confirmed diagnosis result and the suspected diagnosis result is generated.

[0206] Secondly, determine whether the number of disease diagnosis results is less than a preset disease diagnosis result number threshold.

[0207] Here, the preset disease diagnosis result number threshold may be a positive integer less than or equal to N.

[0208] If it is determined to be less than, it indicates that the number of diagnostic results obtained is too small to be suitable for presentation to the user, or there is no diagnostic result and further questioning is required. Therefore, it can be determined to continue the questioning.

[0209] In response to determining that it is not less than, it indicates that the number of obtained diagnosis results can be presented to the user and there is no need to further ask other questions, so it is determined not to continue the consultation.

[0210] With this optional implementation, when the number of diagnosis results is too small (for example, 0), the diagnosis results may not be presented to the user, but the consultation may continue. When it is determined that the number of diagnosis results is appropriate, the consultation may not be continued, and the diagnosis results may be provided to the user, so that an appropriate number of diagnosis results are provided to the user as a reference.

[0211] Optional implementation (seven): Based on the above optional implementation (five) or (six), the server sends the diagnosis result information to the client in step 209, which can be performed as follows:

[0212] The confirmed diagnosis results and suspected diagnosis results in the diagnosis result information and the first S diagnosis results with the highest corresponding confirmed probability values ​​or suspected probability values ​​among the excluded suspected diagnosis results are sent to the client. Here, S is a positive integer.

[0213] By adopting this optional implementation, the diagnosis results with higher relative probability values ​​among the confirmed diagnosis results and suspected diagnosis results can be sent to the client for presentation, rather than sending all the diagnosis results to the client for presentation, that is, the diagnosis results with more reference significance can be presented to the user.

[0214] The AI ​​diagnosis system based on the consultation request provided by the above-mentioned embodiment of the present disclosure presents the questions to be asked and the corresponding answer options on the client. When the patient / medical staff makes a choice from the answer options for the questions to be asked presented on the client, the server will append the answer options selected by the patient / medical staff on the client to the questions to be asked to the consultation questions and the corresponding selected answers that the target patient has answered in history, and obtain a historical consultation reply information sequence. Then, the historical consultation reply information sequence is input into the pre-trained disease diagnosis model to obtain the diagnosis result information. Then, it is determined whether to continue the consultation based on the diagnosis result information. If it is determined to continue the consultation, the current question information to be asked is determined from the preset consultation question information set based on the historical consultation reply information sequence and the diagnosis result information, and the current question information to be asked is sent to the client. Then, the client can continue to present the current question to be asked and the corresponding answer options for the patient / medical staff to answer. The system can achieve technical effects including but not limited to the following:

[0215] First, during the consultation process, patients / medical staff only need to choose from the answer options corresponding to the questions provided without having to input any information, which reduces the difficulty and complexity of the operation for patients / medical staff to provide disease description information and makes it easier to use.

[0216] Second, during the consultation process, the server determines in real time whether the consultation needs to continue each time it receives the consultation reply information sent by the client. Only after it is determined to continue the consultation will it continue to provide consultation questions. There is no need to provide all the consultation questions in the preset consultation question information set to the patient / medical staff for answers, which can reduce the number of consultation questions answered by patients / medical staff and improve the consultation speed.

[0217] Third, the next question to be asked is dynamically determined based on the historical medical consultation answer information sequence that has been answered for the target patient's condition, rather than providing the question information in a preset fixed order. The historical medical consultation answer information sequence that has been answered for the target patient's condition can be dynamically adapted, making the consultation process more targeted.

[0218] Continue to refer Figure 4 , which shows a process 400 of an embodiment of an AI diagnosis method based on a medical inquiry request according to the present disclosure. The AI ​​diagnosis method based on a medical inquiry request is applied to a server and includes the following steps:

[0219] Step 401, in response to receiving a consultation reply request for a target patient identifier sent by a client, adding the consultation question and the corresponding selected answer options in the consultation reply request to a historical consultation reply information sequence corresponding to the target patient identifier.

[0220] Step 402, input the historical medical consultation response information sequence into the pre-trained disease diagnosis model to obtain the diagnosis result information.

[0221] Step 403, determine whether to continue the consultation based on the diagnosis result information.

[0222] Step 404, in response to determining to continue the consultation, based on the historical consultation answer information sequence and the diagnosis result information, the current question information to be consulted is determined from the preset consultation question information set, and the current question information to be consulted is sent to the client.

[0223] In this embodiment, the specific operations of step 401, step 402, step 403 and step 404 and the technical effects produced are the same as those in Figure 2A In the illustrated embodiment, the operations and effects of step 203, step 204, step 205 and step 206 are substantially the same and will not be described in detail herein.

[0224] In some optional implementations, the process 400 may further include the following step 405:

[0225] Step 405, in response to receiving a consultation request for a target patient identifier sent by a client, determining a first question to be consulted from a preset consultation question information set, and sending the first question to be consulted to the client.

[0226] Here, the specific operation of step 405 and the technical effect produced are the same as Figure 2A The operation and effect of step 208 in the illustrated embodiment are substantially the same and will not be described in detail herein.

[0227] In some optional implementations, the process 400 may further include the following step 406:

[0228] Step 406: in response to determining not to continue the consultation, the diagnosis result information is sent to the client.

[0229] Here, the specific operation of step 406 and the technical effect produced are the same as Figure 2A The operation and effect of step 209 in the illustrated embodiment are substantially the same and will not be described in detail herein.

[0230] In some optional implementations, step 402, inputting the historical medical consultation answer information sequence into a pre-trained disease diagnosis model to obtain diagnosis result information, which may include: Figure 2B For the steps 2041 to 2043 shown, please refer to the relevant records of the optional implementation mode (four) and the optional implementation mode (five) in the above embodiments, which will not be repeated here.

[0231] In some optional implementations, each preset medical question information in the preset medical question information set may be associated with a corresponding question weight. Accordingly, in step 405, determining the first question information to be asked from the preset medical question information set may be performed as follows: determining the medical question information with the highest corresponding question weight in the preset medical question information set as the first question information to be asked. For details, please refer to the relevant records of the optional implementation method (II) in the above embodiment, which will not be repeated here.

[0232] In some optional implementations, determining whether to continue the consultation based on the diagnosis result information in step 403 may be performed as follows:

[0233] First, the number of different disease identifiers corresponding to each diagnosis result included in the diagnosis result information is determined as the number of disease diagnosis results.

[0234] Secondly, determine whether the number of disease diagnosis results is less than a preset disease diagnosis result number threshold.

[0235] If it is determined to be less than, it indicates that the number of diagnostic results obtained is too small to be suitable for presentation to the user, or there is no diagnostic result and further questioning is required. Therefore, it can be determined to continue the questioning.

[0236] In response to determining that it is not less than, it indicates that the number of obtained diagnosis results can be presented to the user and there is no need to further ask other questions, so it is determined not to continue the consultation.

[0237] For details, please refer to the relevant records about the optional implementation mode (six) in the above embodiment, which will not be repeated here.

[0238] In some optional implementations, sending the diagnosis result information to the client in step 406 may be performed as follows:

[0239] The first S diagnosis results with the highest corresponding confirmed probability values ​​or suspected probability values ​​among the confirmed diagnosis results and suspected diagnosis results in the diagnosis result information are sent to the client. Here, S is a positive integer.

[0240] For details, please refer to the relevant records about the optional implementation mode (seven) in the above embodiment, which will not be repeated here.

[0241] In some optional implementations, in step 404, based on the historical consultation answer information sequence, determining the current question information to be consulted from the preset consultation question information set may be performed as follows:

[0242] The historical consultation answer information sequence and diagnosis result information are input into the pre-trained consultation question prediction model to obtain the current question information to be consulted.

[0243] For details, please refer to the relevant records about step 206 in the above embodiment, which will not be repeated here.

[0244] The method provided by the above-mentioned embodiment of the present disclosure provides the diagnosis questions and the corresponding answer options on the server. During the diagnosis process, the patient / medical staff only needs to select from the answer options corresponding to the provided questions, without inputting, thereby reducing the difficulty and operation complexity of the patient / medical staff in providing the description information of the symptoms, and facilitating use. The server also determines in real time whether it is necessary to continue the diagnosis each time after receiving the diagnosis reply information sent by the client. Only after it is determined to continue the diagnosis, the diagnosis questions are continued to be provided. It is not necessary to provide all the diagnosis questions in the preset diagnosis question information set to the patient / medical staff for answering, which can reduce the number of diagnosis question answers of the patient / medical staff and improve the diagnosis speed. Finally, the server dynamically determines the next question information to be diagnosed based on the historical diagnosis reply information sequence that has been answered for the target patient's condition, instead of providing the question information to be diagnosed in a preset fixed order, and can dynamically adapt to the historical diagnosis reply information sequence that has been answered for the target patient's condition, so that the diagnosis process is more targeted.

[0245] Continue to refer Figure 5 , which shows a process 500 of an embodiment of an AI diagnosis method based on a medical inquiry request according to the present disclosure. The AI ​​diagnosis method based on a medical inquiry request, applied to a client, comprises the following steps:

[0246] Step 501, in response to receiving information of questions to be asked for a target patient identification from a server, presenting the questions to be asked and corresponding answer options in the information of questions to be asked.

[0247] Step 502, in response to detecting a selection operation for the presented answer options, generating a consultation answer request for the target patient identifier based on the target patient identifier, the information of the question to be asked, and the answer options for which the selection operation is performed, and sending the consultation answer request to the server.

[0248] In this embodiment, the specific operations of step 501 and step 502 and the technical effects produced are the same as those in Figure 2A In the illustrated embodiment, the operations and effects of step 201 and step 202 are substantially the same and will not be described in detail herein.

[0249] In some optional implementations, the above process 500 may further include the following steps 503:

[0250] Step 503: in response to detecting a consultation operation for the target patient identifier, generate and send a consultation request for the target patient identifier to the server.

[0251] In this embodiment, the specific operation of step 503 and the technical effect produced are the same as Figure 2A The operation and effect of step 207 in the illustrated embodiment are substantially the same and will not be described in detail herein.

[0252] In some optional implementations, the above process 500 may further include the following steps 504:

[0253] Step 504: the client presents the diagnosis result information in response to receiving the diagnosis result information sent by the server.

[0254] In this embodiment, the specific operation of step 504 and the technical effect produced are similar to Figure 2A The operation and effect of step 210 in the illustrated embodiment are substantially the same and will not be described in detail herein.

[0255] The AI ​​diagnosis method based on the consultation request provided by the above-mentioned embodiments of the present disclosure can realize that during the consultation process, the patient / medical staff only needs to select from the answer options corresponding to the provided questions without inputting, by presenting the questions to be asked and the corresponding answer options on the client, and presenting the diagnosis result information in some embodiments, thereby reducing the difficulty and operation complexity of the patient / medical staff in providing the symptom description information and facilitating use. In addition, it is not necessary to present all the consultation questions in the preset consultation question information set to the patient / medical staff for answering, which can reduce the number of answers to the patient / medical staff's consultation questions and improve the consultation speed. Finally, in some cases, the diagnosis result information is presented after the diagnosis result information is obtained, and the diagnosis result information reference can be provided to the user on the client.

[0256] Further references Figure 6 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of an AI diagnosis device based on a medical inquiry request. Figure 4 Corresponding to the method embodiment shown, the device can be specifically applied to various servers.

[0257] like Figure 6 As shown, the AI ​​diagnosis device 600 based on the consultation request of this embodiment includes: an answer option adding unit 601, a diagnosis unit 602, a continue consultation determination unit 603 and a current question determination and sending unit 604. Among them, the answer option adding unit 601 is configured to, in response to receiving a consultation reply request for a target patient identifier sent by a client, add the consultation question and the corresponding selected answer option in the consultation reply request to the historical consultation reply information sequence corresponding to the target patient identifier; the diagnosis unit 602 is configured to input the historical consultation reply information sequence into a pre-trained disease diagnosis model to obtain diagnosis result information; the continue consultation determination unit 603 is configured to determine whether to continue the consultation based on the diagnosis result information; the current question determination and sending unit 604 is configured to determine the current question information to be asked from the preset consultation question information set based on the historical consultation reply information sequence and the diagnosis result information in response to determining to continue the consultation, and send the current question information to be asked to the client, so that the client can present the current question information to be asked, wherein the consultation question information includes the consultation question and corresponding at least two answer options.

[0258] In this embodiment, the specific processing of the answer option adding unit 601, the diagnosis unit 602, the continued diagnosis determination unit 603, and the current question determination and sending unit 604 of the AI ​​diagnosis device 600 based on the consultation request and the technical effects thereof can be referred to respectively. Figure 4The relevant descriptions of step 401, step 402, step 403 and step 404 in the corresponding embodiment are not repeated here.

[0259] In some optional implementations, the device 600 may further include:

[0260] The first question determination and sending unit 605 is configured to determine the first question information to be asked from the preset question information set in response to receiving a consultation request for the target patient identifier sent by the client, and send the first question information to be asked to the client.

[0261] In some optional implementations, the device 600 may further include:

[0262] The diagnosis result sending unit 606 is configured to send the diagnosis result information to the client in response to determining not to continue the consultation, so that the client can present the diagnosis result information.

[0263] In some optional implementations, each of the preset medical question information may be associated with a corresponding question weight; and

[0264] The first question determination and sending unit 605 may be further configured to:

[0265] The medical question information with the highest corresponding question weight in the preset medical question information set is determined as the first medical question information to be asked.

[0266] In some optional embodiments, the disease diagnosis model may include a confirmed diagnosis model; and

[0267] The diagnosis unit 602 may be further configured to:

[0268] Input the historical medical consultation answer information sequence into the confirmed diagnosis model to obtain a confirmed diagnosis probability value corresponding to each of the preset N disease identifiers;

[0269] For each of the preset N disease identifiers, the following diagnosis result generation operation is performed: in response to determining that the diagnosis probability value corresponding to the disease identifier is greater than a preset diagnosis probability threshold, generating a diagnosis result corresponding to the disease identifier using the diagnosis probability value corresponding to the disease identifier;

[0270] The generated diagnosis results are combined to obtain the diagnosis result information.

[0271] In some optional embodiments, the disease diagnosis model may also include a suspected diagnosis model; and

[0272] The diagnosis unit 602 may be further configured to:

[0273] Inputting the historical medical consultation answer information sequence into the suspected diagnosis model to obtain a suspected probability value corresponding to each of the preset N disease identifiers; and

[0274] The diagnostic result generating operation may further include:

[0275] In response to determining that the confirmed probability value corresponding to the disease identifier is not greater than the preset confirmed probability threshold, determining whether the suspected probability value corresponding to the disease identifier is greater than the preset suspected probability threshold;

[0276] In response to determining that it is greater than, a suspected diagnosis result corresponding to the disease identifier is generated using the suspected probability value corresponding to the disease identifier.

[0277] In some optional implementations, the continued medical consultation determination unit 603 may be further configured to:

[0278] Determine the number of different disease identifiers corresponding to each diagnosis result included in the diagnosis result information as the number of disease diagnosis results;

[0279] Determining whether the number of disease diagnosis results is less than a preset disease diagnosis result number threshold;

[0280] In response to determining that it is less than, determining to continue the consultation;

[0281] In response to determining not less than, it is determined not to continue the consultation.

[0282] In some optional implementations, the diagnosis result sending unit 606 may be further configured to:

[0283] The first S diagnosis results with the highest corresponding confirmed probability values ​​and suspected probability values ​​among the confirmed diagnosis results and suspected diagnosis results in the diagnosis result information are sent to the client, where S is a positive integer.

[0284] In some optional implementations, the current problem determining and sending unit 604 may be further configured to:

[0285] The historical medical consultation answer information sequence and the diagnosis result information are input into a pre-trained medical question prediction model to obtain the current medical question information to be asked, wherein the medical question prediction model is used to characterize the correspondence between the medical consultation answer information sequence and the diagnosis result information and the medical question information in the preset medical question information set.

[0286] It should be noted that the implementation details and technical effects of each unit in the AI ​​diagnostic device based on medical inquiry request provided by the present disclosure can be referred to the description of other embodiments in the present disclosure and will not be repeated here.

[0287] Further references Figure 7 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of an AI diagnosis device based on a medical inquiry request. Figure 5 Corresponding to the method embodiment shown, the device can be specifically applied to various clients.

[0288] like Figure 7 As shown, the AI ​​diagnosis device 700 based on the consultation request of this embodiment includes: a question and answer presentation unit 701 and a reply request sending unit 702. Among them, the question and answer presentation unit 701 is configured to present the consultation questions and corresponding answer options in the consultation questions information in response to receiving the consultation questions information for the target patient identification from the server; the reply request sending unit 702 is configured to generate a consultation reply request for the target patient identification based on the target patient identification, the consultation questions information and the answer options for the selection operation in response to detecting the selection operation for the presented answer options, and send the consultation reply request to the server, so that the server determines whether to continue the consultation based on the consultation reply request, and determines and returns the current consultation questions information from the preset consultation question information set in response to determining to continue the consultation.

[0289] In this embodiment, the specific processing of the question and answer presentation unit 701 and the reply request sending unit 702 of the AI ​​diagnosis device 700 based on the medical consultation request and the technical effects thereof can be referred to respectively. Figure 5 The relevant descriptions of step 501 and step 502 in the corresponding embodiment are not repeated here.

[0290] In some optional implementations, the device 700 may further include:

[0291] The consultation request sending unit 703 is configured to generate and send a consultation request for the target patient identifier to the server in response to detecting a consultation operation for the target patient identifier.

[0292] In some optional implementations, the device 700 may further include:

[0293] The diagnosis result presenting unit 704 is configured to present the diagnosis result information in response to receiving the diagnosis result information sent by the server.

[0294] In some optional implementations, the diagnosis result information may include a confirmed probability value or a suspected probability value corresponding to at least one disease marker among the preset N disease markers.

[0295] It should be noted that the implementation details and technical effects of each unit in the AI ​​diagnostic device based on medical inquiry request provided by the present disclosure can be referred to the description of other embodiments in the present disclosure and will not be repeated here.

[0296] Reference below Figure 8 , which shows a schematic diagram of the structure of a computer system 800 suitable for implementing a client or server of an embodiment of the present disclosure. Figure 8 The computer system 800 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0297] like Figure 8 As shown, the computer system 800 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage device 708 to a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the electronic device 800 are also stored. The processing device 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0298] Typically, the following devices may be connected to the I / O interface 805: input devices 806 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, etc.; output devices 807 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 808 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 809. The communication devices 809 may allow the computer system 800 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 8 The computer system 800 is shown with various devices, but it should be understood that it is not required to implement or have all the devices shown. More or fewer devices may be implemented or have instead.

[0299] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 809, or installed from a storage device 808, or installed from a ROM 802. When the computer program is executed by the processing device 801, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0300] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0301] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0302] The computer readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device implements the following Figure 4 The embodiment and its optional implementations shown in the AI ​​diagnosis method based on the consultation request, and / or, as Figure 5 The illustrated embodiment and its optional implementations illustrate an AI diagnosis method based on a medical inquiry request.

[0303] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0304] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0305] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of the unit does not limit the unit itself in some cases. For example, the question and answer presentation unit may also be described as "a unit that presents the questions and corresponding answer options in the question information to be asked in response to receiving the question information to be asked for the target patient identification from the server."

[0306] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

Claims

1. An AI diagnosis method based on a medical inquiry request, applied to a server, the method comprising: In response to receiving a consultation reply request for a target patient identifier sent by a client, adding the consultation question and the corresponding selected answer option in the consultation reply request to a historical consultation reply information sequence corresponding to the target patient identifier, wherein the historical consultation reply information sequence corresponding to the target patient identifier is composed of the consultation questions and the corresponding selected answers that have been answered / replied to the actual situation of the target patient during the current consultation process in the order of answering / replying; Inputting the historical medical consultation answer information sequence into a pre-trained disease diagnosis model to obtain diagnosis result information; Determining whether to continue the consultation based on the diagnosis result information; In response to determining to continue the consultation, based on the historical consultation answer information sequence and the diagnosis result information, current question information to be asked is determined from a preset consultation question information set, and the current question information to be asked is sent to the client, so that the client presents the current question information to be asked, wherein the consultation question information includes a consultation question and corresponding at least two answer options; Wherein, the method further comprises: In response to receiving a consultation request for the target patient identifier sent by a client, determining a first piece of question information to be asked from the preset consultation question information set, and sending the first piece of question information to be asked to the client; Each preset medical question information is associated with a question weight, and Determining the first question information to be asked from the preset question information set includes: The medical question information with the highest corresponding question weight in the preset medical question information set is determined as the first question information to be asked, wherein, for each medical question information in the preset medical question information set, an initial first weight component is set according to a large amount of patient medical record information, an initial second weight component is set according to the knowledge and experience of medical experts, an initial third weight component is set according to relevant content in professional medical books, and an initial fourth weight component is set according to professional medical literature, and then the first weight component, second weight component, third weight component and fourth weight component corresponding to each medical question are weightedly summed according to the weight coefficient of the first weight component, the weight coefficient of the second weight component, the weight coefficient of the third weight component and the weight coefficient of the fourth weight component corresponding to the medical question to obtain the weight associated with each medical question information.

2. The method according to claim 1, wherein: The method further comprises: In response to determining not to continue the consultation, the diagnosis result information is sent to the client so that the client can present the diagnosis result information.

3. The method according to claim 1, wherein: The disease diagnosis model includes a confirmed diagnosis model; and The step of inputting the historical medical consultation answer information sequence into a pre-trained disease diagnosis model to obtain diagnosis result information includes: Input the historical medical consultation answer information sequence into the confirmed diagnosis model to obtain a confirmed diagnosis probability value corresponding to each of the preset N disease identifiers; For each of the preset N disease identifiers, the following diagnosis result generation operation is performed: in response to determining that the diagnosis probability value corresponding to the disease identifier is greater than a preset diagnosis probability threshold, generating a diagnosis result corresponding to the disease identifier using the diagnosis probability value corresponding to the disease identifier; The generated diagnosis results are combined to obtain the diagnosis result information.

4. The method according to claim 3, wherein: The disease diagnosis model also includes a suspected diagnosis model; and The step of inputting the historical medical consultation answer information sequence into a pre-trained disease diagnosis model to obtain diagnosis result information further includes: Inputting the historical medical consultation answer information sequence into the suspected diagnosis model to obtain a suspected probability value corresponding to each of the preset N disease identifiers; and The diagnostic result generation operation further includes: In response to determining that the confirmed probability value corresponding to the disease identifier is not greater than the preset confirmed probability threshold, determining whether the suspected probability value corresponding to the disease identifier is greater than the preset suspected probability threshold; In response to determining that it is greater than, a suspected diagnosis result corresponding to the disease identifier is generated using the suspected probability value corresponding to the disease identifier.

5. The method according to any one of claims 3 to 4, wherein: The determining whether to continue the consultation according to the diagnosis result information includes: Determine the number of different disease identifiers corresponding to each diagnosis result included in the diagnosis result information as the number of disease diagnosis results; Determining whether the number of disease diagnosis results is less than a preset disease diagnosis result number threshold; In response to determining that it is less than, determining to continue the consultation; In response to determining not less than, it is determined not to continue the consultation.

6. The method according to claim 5, wherein: The sending the diagnosis result information to the client includes: The first S diagnosis results with the highest corresponding confirmed probability values ​​and suspected probability values ​​among the confirmed diagnosis results and suspected diagnosis results in the diagnosis result information are sent to the client, where S is a positive integer.

7. The method according to claim 1, wherein: The step of determining the current question information to be asked from the preset question information set based on the historical question answer information sequence and the diagnosis result information includes: The historical medical consultation answer information sequence and the diagnosis result information are input into a pre-trained medical question prediction model to obtain the current medical question information to be asked, wherein the medical question prediction model is used to characterize the correspondence between the medical consultation answer information sequence and the diagnosis result information and the medical question information in the preset medical question information set.

8. An AI diagnosis method based on a medical inquiry request, applied to a client, the method comprising: In response to receiving the information of questions to be asked for the target patient identification from the server, presenting the questions to be asked and the corresponding answer options in the information of questions to be asked; In response to detecting a selection operation for the presented answer options, generating a consultation answer request for the target patient identifier based on the target patient identifier, the question information to be consulted, and the answer options for which the selection operation is performed, and sending the consultation answer request to the server, so that the server determines whether to continue the consultation based on the consultation answer request, and in response to determining to continue the consultation, determining and returning the current question information to be consulted from the preset consultation question information set; Wherein, the method further comprises: In response to receiving the first question-to-be-asked information for the target patient identification from the server, presenting the question-to-be-asked information and the corresponding answer options; Each preset medical question information is associated with a question weight, and the first question information to be asked is determined from the preset medical question information set, including: The medical question information with the highest corresponding question weight in the preset medical question information set is determined as the first question information to be asked, wherein, for each medical question information in the preset medical question information set, an initial first weight component is set according to a large amount of patient medical record information, an initial second weight component is set according to the knowledge and experience of medical experts, an initial third weight component is set according to relevant content in professional medical books, and an initial fourth weight component is set according to professional medical literature, and then the first weight component, second weight component, third weight component and fourth weight component corresponding to each medical question are weightedly summed according to the weight coefficient of the first weight component, the weight coefficient of the second weight component, the weight coefficient of the third weight component and the weight coefficient of the fourth weight component corresponding to the medical question to obtain the weight associated with each medical question information.

9. The method according to claim 8, wherein: The method further comprises: In response to detecting a consultation operation for the target patient identifier, a consultation request for the target patient identifier is generated and sent to the server.

10. The method according to claim 8, wherein: The method further comprises: In response to receiving the diagnosis result information sent by the server, presenting the diagnosis result information.

11. The method according to claim 8, wherein: The diagnosis result information includes a confirmed probability value or a suspected probability value corresponding to at least one disease marker among the preset N disease markers.

12. An AI diagnostic device based on a medical inquiry request, applied to a server, the device comprising: The answer option adding unit is configured to, in response to receiving a consultation reply request for a target patient identifier sent by a client, add the consultation question and the corresponding selected answer option in the consultation reply request to a historical consultation reply information sequence corresponding to the target patient identifier, wherein the historical consultation reply information sequence corresponding to the target patient identifier is a consultation reply information sequence consisting of the consultation questions and the corresponding selected answers that have been answered / answered in the current consultation process according to the actual situation of the target patient in the order of answering / answering; A diagnosis unit configured to input the historical medical consultation answer information sequence into a pre-trained disease diagnosis model to obtain diagnosis result information; a continue consultation determination unit, configured to determine whether to continue the consultation according to the diagnosis result information; a current question determination and sending unit, configured to, in response to determining to continue the consultation, determine the current question information to be asked from a preset consultation question information set based on the historical consultation answer information sequence and the diagnosis result information, and send the current question information to be asked to the client, so that the client presents the current question information to be asked, wherein the consultation question information includes a consultation question and corresponding at least two answer options; Wherein, the device further comprises: a first question determination and sending unit, configured to, in response to receiving a consultation request for the target patient identifier sent by a client, determine a first question to be asked from the preset consultation question information set, and send the first question to be asked to the client; Each preset medical question information is associated with a question weight; and The first problem determination and sending unit is further configured to: The medical question information with the highest corresponding question weight in the preset medical question information set is determined as the first question information to be asked, wherein, for each medical question information in the preset medical question information set, an initial first weight component is set according to a large amount of patient medical record information, an initial second weight component is set according to the knowledge and experience of medical experts, an initial third weight component is set according to relevant content in professional medical books, and an initial fourth weight component is set according to professional medical literature, and then the first weight component, second weight component, third weight component and fourth weight component corresponding to each medical question are weightedly summed according to the weight coefficient of the first weight component, the weight coefficient of the second weight component, the weight coefficient of the third weight component and the weight coefficient of the fourth weight component corresponding to the medical question to obtain the weight associated with each medical question information.

13. The device according to claim 12, wherein: The device also includes: The diagnosis result sending unit is configured to send the diagnosis result information to the client in response to determining not to continue the consultation, so that the client can present the diagnosis result information.

14. The device according to claim 12, wherein: The disease diagnosis model includes a confirmed diagnosis model; and The diagnostic unit is further configured to: Input the historical medical consultation answer information sequence into the confirmed diagnosis model to obtain a confirmed diagnosis probability value corresponding to each of the preset N disease identifiers; For each of the preset N disease identifiers, the following diagnosis result generation operation is performed: in response to determining that the diagnosis probability value corresponding to the disease identifier is greater than a preset diagnosis probability threshold, generating a diagnosis result corresponding to the disease identifier using the diagnosis probability value corresponding to the disease identifier; The generated diagnosis results are combined to obtain the diagnosis result information.

15. The device according to claim 14, wherein: The disease diagnosis model also includes a suspected diagnosis model; and The diagnostic unit is further configured to: Inputting the historical medical consultation answer information sequence into the suspected diagnosis model to obtain a suspected probability value corresponding to each of the preset N disease identifiers; as well as The diagnostic result generation operation further includes: In response to determining that the confirmed probability value corresponding to the disease identifier is not greater than the preset confirmed probability threshold, determining whether the suspected probability value corresponding to the disease identifier is greater than the preset suspected probability threshold; In response to determining that it is greater than, a suspected diagnosis result corresponding to the disease identifier is generated using the suspected probability value corresponding to the disease identifier.

16. The device according to any one of claims 14-15, wherein: The continued consultation determination unit is further configured to: Determine the number of different disease identifiers corresponding to each diagnosis result included in the diagnosis result information as the number of disease diagnosis results; Determining whether the number of disease diagnosis results is less than a preset disease diagnosis result number threshold; In response to determining that it is less than, determining to continue the consultation; In response to determining not less than, it is determined not to continue the consultation.

17. The device according to claim 16, wherein: The diagnosis result sending unit is further configured to: The first S diagnosis results with the highest corresponding confirmed probability values ​​and suspected probability values ​​among the confirmed diagnosis results and suspected diagnosis results in the diagnosis result information are sent to the client, where S is a positive integer.

18. The device according to claim 12, wherein: The current problem determination and sending unit is further configured to: The historical medical consultation answer information sequence and the diagnosis result information are input into a pre-trained medical question prediction model to obtain the current medical question information to be asked, wherein the medical question prediction model is used to characterize the correspondence between the medical consultation answer information sequence and the diagnosis result information and the medical question information in the preset medical question information set.

19. A device for presenting medical question information, applied to a client, comprising: A question and answer presenting unit, configured to present the question and corresponding answer options in the question information to be asked in response to receiving the question information to be asked for the target patient from the server; a reply request sending unit, configured to, in response to detecting a selection operation for a presented answer option, generate a consultation reply request for the target patient identifier based on the target patient identifier, the question information to be asked, and the answer option for which the selection operation is performed, and send the consultation reply request to the server, so that the server determines whether to continue the consultation based on the consultation reply request, and in response to determining to continue the consultation, determine and return the current question information to be asked from the preset consultation question information set; Wherein, the device further comprises: In response to receiving the first question-to-be-asked information for the target patient identification from the server, presenting the question-to-be-asked information and the corresponding answer options; Each preset medical question information is associated with a question weight, and the first question information to be asked is determined from the preset medical question information set, including: The medical question information with the highest corresponding question weight in the preset medical question information set is determined as the first question information to be asked, wherein, for each medical question information in the preset medical question information set, an initial first weight component is set according to a large amount of patient medical record information, an initial second weight component is set according to the knowledge and experience of medical experts, an initial third weight component is set according to relevant content in professional medical books, and an initial fourth weight component is set according to professional medical literature, and then the first weight component, second weight component, third weight component and fourth weight component corresponding to each medical question are weightedly summed according to the weight coefficient of the first weight component, the weight coefficient of the second weight component, the weight coefficient of the third weight component and the weight coefficient of the fourth weight component corresponding to the medical question to obtain the weight associated with each medical question information.

20. The device according to claim 19, wherein The device also includes: The consultation request sending unit is configured to generate and send a consultation request for the target patient identifier to the server in response to detecting a consultation operation for the target patient identifier.

21. The device according to claim 19, wherein The device also includes: The diagnosis result presenting unit is configured to present the diagnosis result information in response to receiving the diagnosis result information sent by the server.

22. The device according to claim 19, wherein The diagnosis result information includes a confirmed probability value or a suspected probability value corresponding to at least one disease marker among the preset N disease markers.

23. A server, comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method according to any one of claims 1 to 7.

24. A client, comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method according to any one of claims 8 to 11.

25. A medical consultation system, comprising the server according to claim 23 and at least one client according to claim 24.

26. A computer-readable storage medium having a computer program stored thereon, wherein: When the computer program is executed by one or more processors, the method according to any one of claims 1 to 11 is implemented.

Images