CN109243600B

CN109243600B - Method and apparatus for outputting information

Info

Publication number: CN109243600B
Application number: CN201810884772.4A
Authority: CN
Inventors: 杨叶辉; 夏源; 罗程亮; 范斌; 王华南
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2018-08-06
Filing date: 2018-08-06
Publication date: 2020-11-20
Anticipated expiration: 2038-08-06
Also published as: CN109243600A

Abstract

The embodiment of the application discloses a method and a device for outputting information. One embodiment of the method comprises: acquiring user input information; in response to the user input information including an image of the patient, determining a first set of probabilities based on the image, wherein the first probability is a probability that the patient has a disease of a predetermined set of diseases determined based on the image; determining a disease from the disease set as a target disease based on the first set of probabilities; and outputting the disease information of the target disease. The embodiment enriches the generation mode of the information, improves the diagnosis accuracy of the electronic equipment and is beneficial to saving the diagnosis time of medical staff.

Description

Method and apparatus for outputting information

Technical Field

The embodiment of the application relates to the technical field of computers, in particular to a method and a device for outputting information.

Background

At present, gaps exist between medical resources and the hospitalization requirements of patients, and the medical resources and the hospitalization requirements of the patients are limited by the current situations of long culture period, uneven levels and unbalanced distribution of doctors. The contradiction between the doctor-patient supply-demand relationship is difficult to automatically reconcile in a short time and tends to be more and more drastic.

Among the numerous medical needs, there are a great number of needs that can be solved by a doctor who gives a disease or a treatment plan to be diagnosed after the doctor simply makes an inquiry to a patient.

Disclosure of Invention

The embodiment of the application provides a method and a device for outputting information.

In a first aspect, an embodiment of the present application provides a method for outputting information, where the method includes: acquiring user input information; in response to the user input information including an image of the patient, determining a first set of probabilities based on the image, wherein the first probability is a probability that the patient has a disease of a predetermined set of diseases determined based on the image; determining a disease from the disease set as a target disease based on the first set of probabilities; and outputting the disease information of the target disease.

In some embodiments, determining the first set of probabilities based on the image includes: and inputting the image into a pre-trained convolutional neural network to obtain a first probability set.

In some embodiments, prior to determining the disease from the set of diseases as the target disease based on the first set of probabilities, the method further comprises: in response to the user input information not including the image of the patient, a first set of probabilities is determined based on the user input information.

In some embodiments, determining, from the set of diseases, a disease from which the patient suffers as the target disease based on the first set of probabilities includes: generating a problem for the patient based on the first set of probabilities; acquiring feedback information aiming at a problem; based on the feedback information and the user input information, a disease from which the patient suffers is determined from the set of diseases as a target disease.

In some embodiments, determining a disease from the set of diseases as the target disease based on the feedback information and the user input information comprises: acquiring prior probability of diseases in a disease set; determining a second set of probabilities based on the feedback information and the user input information, wherein the second probability is a probability that the patient has a disease of the predetermined set of diseases determined based on the feedback information and the user input information; determining a score for the patient having a disease of the set of diseases based on the first set of probabilities, the second set of probabilities, and the obtained prior probabilities, wherein the score characterizes a magnitude of likelihood that the patient has a disease of the set of diseases; determining whether to continue generating questions based on the determined scores; in response to determining not to continue generating the question, determining a disease from the set of diseases that the patient has as the target disease.

In some embodiments, generating a question for the patient based on the first set of probabilities includes: determining disease identities of the diseases in the set of diseases that the patient has based on the first set of probabilities; based on the disease identification and the pre-trained bayesian network, a question and an output for the patient are generated.

In some embodiments, the bayesian network has a structure of directed bipartite graphs, the start points of which are used to characterize the disease signatures, the end points of which are used to characterize the symptom information, and the conditional probabilities in the bayesian network are used to characterize the probability that a patient of the disease characterized by the disease signature produces the symptom characterized by the symptom information.

In some embodiments, the method further comprises: and continuing to train the Bayesian network based on the feedback information.

In a second aspect, an embodiment of the present application provides an apparatus for outputting information, including: an acquisition unit configured to acquire user input information; a first determining unit configured to determine a first set of probabilities based on the image in response to the user input information including an image of the patient, wherein the first probability is a probability that the patient has a disease of a predetermined set of diseases determined based on the image; a second determination unit configured to determine, as a target disease, a disease from which the patient has, from the set of diseases, based on the first set of probabilities; an output unit configured to output disease information of a target disease.

In some embodiments, the first determining unit is further configured to: and inputting the image into a pre-trained convolutional neural network to obtain a first probability set.

In some embodiments, the apparatus further comprises: a third determination unit configured to determine the first set of probabilities based on the user input information in response to the user input information not including the image of the patient.

In some embodiments, the second determining unit is further configured to: generating a problem for the patient based on the first set of probabilities; acquiring feedback information aiming at a problem; based on the feedback information and the user input information, a disease from which the patient suffers is determined from the set of diseases as a target disease.

In some embodiments, the second determining unit is further configured to: acquiring prior probability of diseases in a disease set; determining a second set of probabilities based on the feedback information and the user input information, wherein the second probability is a probability that the patient has a disease of the predetermined set of diseases determined based on the feedback information and the user input information; determining a score for the patient having a disease of the set of diseases based on the first set of probabilities, the second set of probabilities, and the obtained prior probabilities, wherein the score characterizes a magnitude of likelihood that the patient has a disease of the set of diseases; determining whether to continue generating questions based on the determined scores; in response to determining not to continue generating the question, determining a disease from the set of diseases that the patient has as the target disease.

In some embodiments, the second determining unit is further configured to: determining disease identities of the diseases in the set of diseases that the patient has based on the first set of probabilities; based on the disease identification and the pre-trained bayesian network, a question and an output for the patient are generated.

In some embodiments, the apparatus further comprises: a training unit configured to continue training the Bayesian network based on the feedback information.

In a third aspect, an embodiment of the present application provides an electronic device for outputting information, including: one or more processors; a storage device, on which one or more programs are stored, which, when executed by the one or more processors, cause the one or more processors to implement the method of any of the embodiments of the method for outputting information as described above.

In a fourth aspect, the present application provides a computer-readable medium for outputting information, on which a computer program is stored, which when executed by a processor implements the method of any one of the embodiments of the method for outputting information as described above.

According to the method and the device for outputting the information, the user input information is acquired, then the first probability set is determined based on the image under the condition that the user input information comprises the image of the patient, the first probability is determined based on the image, the probability that the patient suffers from the diseases in the predetermined disease set is determined, then the disease that the patient suffers from is determined from the disease set as the target disease based on the first probability set, and finally the disease information of the target disease is output, so that the generation mode of the information is enriched, the diagnosis accuracy of the electronic equipment is improved, and the diagnosis time of medical staff is saved.

Drawings

Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:

FIG. 1 is an exemplary system architecture diagram in which one embodiment of the present application may be applied;

FIG. 2 is a flow diagram for one embodiment of a method for outputting information, in accordance with the present application;

3A-3B are schematic diagrams of an application scenario of a method for outputting information according to an embodiment of the present application;

FIG. 4 is a flow diagram of yet another embodiment of a method for outputting information according to the present application;

FIG. 5 is a flow diagram of yet another embodiment of a method for outputting information according to the present application;

FIG. 6 is a schematic diagram of a Bayesian network in accordance with one embodiment of a method for outputting information of the present application;

FIG. 7 is a schematic block diagram illustrating one embodiment of an apparatus for outputting information according to the present application;

FIG. 8 is a block diagram of a computer system suitable for use in implementing the electronic device of an embodiment of the present application.

Detailed Description

The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.

It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.

Fig. 1 shows an exemplary system architecture 100 to which embodiments of a method for outputting information or an apparatus for outputting information of embodiments of the present application may be applied.

As shown in fig. 1, the system architecture 100 may include

terminal devices

101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the

terminal devices

101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

A user may use the

terminal devices

101, 102, 103 to interact with the server 105 over the network 104 to receive or send messages (e.g., send user input information or receive disease information), etc. The

terminal devices

101, 102, 103 may have various communication client applications installed thereon, such as a medical inquiry application, a web browser application, a shopping application, a search application, an instant messaging tool, a mailbox client, social platform software, and the like.

The

terminal apparatuses

101, 102, and 103 may be hardware or software. When the

terminal devices

101, 102, 103 are hardware, they may be various electronic devices having a display screen and supporting information input, reception and transmission, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like. When the

terminal apparatuses

101, 102, 103 are software, they can be installed in the electronic apparatuses listed above. It may be implemented as multiple pieces of software or software modules (e.g., software or software modules used to provide distributed services) or as a single piece of software or software module. And is not particularly limited herein.

The server 105 may be a server that provides various services, such as a background image processing server that provides support for images sent by the

terminal devices

101, 102, 103. The background image processing server may perform processing such as feature extraction on the received data such as the image, and feed back a processing result (e.g., probability that the patient has a disease in a predetermined set of diseases) to the terminal device.

It should be noted that the method for outputting information provided in the embodiment of the present application may be executed by the server 105, or may be executed by the

terminal devices

101, 102, and 103, and accordingly, the apparatus for outputting information may be provided in the server 105, or may be provided in the

terminal devices

101, 102, and 103.

The server may be hardware or software. When the server is hardware, it may be implemented as a distributed server cluster formed by multiple servers, or may be implemented as a single server. When the server is software, it may be implemented as multiple pieces of software or software modules (e.g., software or software modules used to provide distributed services), or as a single piece of software or software module. And is not particularly limited herein.

It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation. When the execution subject of the information processing method does not need data transmission with other electronic devices, the system architecture may not include a network and other electronic devices other than the execution subject.

With continued reference to FIG. 2, a flow 200 of one embodiment of a method for outputting information in accordance with the present application is shown. The method for outputting information comprises the following steps:

step 201, obtaining user input information.

In this embodiment, an execution subject of the method for outputting information (e.g., a server or a terminal device shown in fig. 1) may acquire user input information from other electronic devices, or locally, by a wired connection manner or a wireless connection manner. The user input information may be various information input by the user. The user input information may include, but is not limited to, at least one of: age information, marriage and childbirth status information, age information, disease symptom information, examination index information, and the like. The presentation form of the user input information may include at least one of: text, audio, image. The image may be a picture or a video. The image may include, but is not limited to, at least one of: computed Tomography (CT) images, Computed Radiography (CR) images, Magnetic Resonance examination (MR) images, direct Digital flat panel X-ray imaging (DR) images, and the like.

In some cases, the user input information may include an image of the patient (e.g., a lesion image of the patient), and may also include information describing symptoms of a disease suffered by the patient (e.g., skin itch), age information, sex information, and the like that cannot be fed back through the image.

As an example, when the execution subject is a server, the execution subject may acquire user input information from a terminal device (e.g., a mobile phone, a computer, etc.) used by the user. When the execution main body is a terminal device, the execution main body can directly acquire user input information input by a user.

In response to the user input information including an image of the patient, a first set of probabilities is determined based on the image, step 202.

In this embodiment, in a case where the user input information includes an image of the patient, the execution subject may determine the first set of probabilities based on the image. Wherein the first probability is a probability determined based on the image that the patient has a disease of a predetermined set of diseases. The number of first probabilities in the first set of probabilities may be the same as the number of diseases in the set of diseases. The first probability may be in a one-to-one correspondence with the disease. The image of the patient included in the user input information may be an image captured of the patient (e.g., a lesion image of the skin of a dermatologic patient). The set of diseases may be a set of all diseases; or a set of all diseases of one or more departments, for example, the set of diseases may be a set of all diseases of dermatology; the disease in the set of diseases can also be a disease with a characteristic, e.g., the disease in the set of diseases can be a disease that can characterize a focus of the patient by an image (e.g., fracture, cervical spondylosis, lumbar disc herniation).

Here, the patient and the user who inputs the user input information may be the same person or different persons. The patient may be a patient corresponding to the image comprised by the user input information mentioned above, e.g. the image may be acquired from the patient.

In some optional implementations of this embodiment, in the case that the user input information includes an image of the patient, the step 202 may be implemented as follows: and inputting the image into a pre-trained convolutional neural network to obtain a first probability set. The convolutional neural network may be used, among other things, to characterize the correspondence between the images and the probability of the patient having each disease in the set of diseases.

In practice, the executing entity or other electronic device may train the convolutional neural network according to the following steps:

first, a set of training samples is obtained. The training sample comprises a sample image and a disease identifier corresponding to the sample image included in the training sample, wherein the disease identifier may include but is not limited to one of the following: disease name, disease code.

Then, using a machine learning algorithm, the initial convolutional neural network is trained with the images included in the training samples as inputs and the disease identifiers included in the training samples as expected outputs, thereby training a convolutional neural network (i.e., a trained convolutional neural network) for determining the probability of the patient having a disease in the predetermined set of diseases. The initial convolutional neural network may be an untrained convolutional neural network, or may be a convolutional neural network that is not trained to be expected (for example, the obtained disease probability error is greater than a threshold or the loss function value is greater than a threshold).

Optionally, the executing entity or other electronic device may also train the convolutional neural network according to the following steps:

first, a set of training samples is obtained. The training sample comprises a sample image and a disease identifier corresponding to the image included in the training sample. Then, by using a machine learning algorithm, for each training sample in the training sample set, taking an image included in the training sample as an input, taking a disease identifier included in the training sample as an expected output, and training the initial convolutional neural network, thereby obtaining the convolutional neural network through training.

And then, acquiring a test sample set, wherein the test sample comprises a test image and a disease identifier corresponding to the test image included in the test sample.

Then, for each test sample in the test sample set, the test image included in the test sample is input to the convolutional neural network obtained by training, so as to obtain the actually output disease identifier.

Finally, comparing each actually output disease identification with each disease identification included in the test sample set to determine the accuracy of the convolutional neural network obtained through training. And if the accuracy is greater than or equal to a preset accuracy threshold, using the trained convolutional neural network as a convolutional neural network for determining the probability that the patient has the diseases in the predetermined disease set (namely, the trained convolutional neural network). And if the accuracy is smaller than the preset accuracy threshold, adjusting the parameters of the convolutional neural network, and continuing to train the convolutional neural network obtained by training.

It is understood that the convolutional neural network may include convolutional layers, pooling layers, fully-connected layers, dropout layers, and softmax layers. The softmax layer is the last layer of the convolutional neural network, and the number of nodes may be the number of diseases in the disease set. Thus, the probability of a patient having each disease in the set of diseases can be derived by the convolutional neural network.

As another implementation, in the case that the user input information includes an image of the patient, the step 202 may also be implemented as the following steps:

first, the executing body may send the image to an electronic device (e.g., a mobile phone, a computer, etc.) used by a doctor or other person with diagnosis capability.

The executive agent may then receive probabilities of the patient having a disease of a predetermined set of diseases returned by an electronic device used by a doctor or other diagnostically-capable person, and use the received set of probabilities as a first set of probabilities.

As yet another implementation, in a case where the user input information includes an image of the patient, the execution subject may further determine the first set of probabilities based on the user input information (including the image of the patient and other information in the user input information other than the image of the patient). For example, the execution body may send the user input information to an electronic device (e.g., a mobile phone, a computer, etc.) used by a doctor or other person with diagnosis capability. The executive agent may then receive probabilities of the patient having a disease of a predetermined set of diseases returned by an electronic device used by a doctor or other diagnostically-capable person, and use the received set of probabilities as a first set of probabilities.

Step 203, determining the disease suffered by the patient from the disease set as the target disease based on the first probability set.

In this embodiment, the execution subject may determine, as the target disease, a disease from which the patient has, from the disease set, based on the first probability set. The target disease may be one disease or a plurality of diseases. The number of target diseases may be set by the user as described above, or may be set by a technician.

For example, the execution agent may determine, as the target disease, a disease corresponding to a first probability with the highest probability in the first probability set, or may determine, as the target disease, three diseases corresponding to 3 (or another number of) first probabilities with the highest probability in the first probability set.

In some optional implementations of this embodiment, the executing main body may further execute the step 203 according to the following steps:

first, based on a first set of probabilities, a question for a patient is generated.

Then, feedback information for the problem is acquired.

Finally, based on the feedback information and the user input information, a disease from which the patient suffers is determined from the set of diseases as a target disease.

In some optional implementations of this embodiment, the executing entity may further continue to train the convolutional neural network based on the feedback information.

As an example, the above-described problem generated by the executing body may be a problem for symptoms that may be produced by a patient having a target disease, for example, when the executing body determines that the patient may have urticaria, a problem "whether red swelling is present", "whether itching is present", or the like may be generated. When the generated problem is that whether the user has a rhinorrhea, the feedback information is that the user has no, and the target disease is that the user has urticaria, it may be that the probability (i.e., the first probability) obtained by the convolutional neural network is not accurate, and in this scenario, the feedback information (e.g., "yes" or "no") may be used as feedback to determine whether the probability obtained by the convolutional neural network is accurate, so as to continue training the convolutional neural network, thereby improving the accuracy of the probability obtained by the convolutional neural network.

Step 204, outputting the disease information of the target disease.

In this embodiment, the execution subject may output disease information of the target disease. The disease information may be a one-to-one correspondence relationship with a disease, and the disease information may be introduction information and treatment information of the disease, or may be a disease identifier for identifying the disease, where the disease identifier may include, but is not limited to, one of the following: disease name, disease code.

With continuing reference to fig. 3A and 3B, fig. 3A and 3B are schematic diagrams of an application scenario of a method for outputting information according to the present embodiment. In fig. 3A, a user inputs user input information to a terminal device (i.e., the execution main body described above), wherein the user input information includes text information 301 and an image 302. After the terminal device obtains the user input information, it is determined that the user input information includes an image of the patient. The images 302 are then input to a pre-trained convolutional neural network, resulting in a probability that the patient has each disease in a predetermined set of diseases. Then, the terminal device determines a disease in a disease set corresponding to the highest probability among the obtained probabilities as a target disease. Finally, referring to fig. 3B, the terminal device outputs disease information of the target disease, "the patient having the above symptoms may have urticaria according to the information input by you".

The method provided by the above embodiment of the application enriches the generation ways of the information, improves the accuracy of the diagnosis of the electronic device, and helps to save the diagnosis time of medical staff by acquiring the user input information, then determining the first probability set based on the image when the user input information includes the image of the patient, wherein the first probability is the probability that the patient has the disease in the predetermined disease set determined based on the image, then determining the disease that the patient has as the target disease from the disease set based on the first probability set, and finally outputting the disease information of the target disease.

With further reference to fig. 4, a flow 400 of yet another embodiment of a method for outputting information is shown. The process 400 of the method for outputting information includes the steps of:

step 401, obtaining user input information. Thereafter, step 402 is performed.

In this embodiment, step 401 is substantially the same as step 201 in the corresponding embodiment of fig. 2, and is not described here again.

At step 402, it is determined whether the user input information includes an image of the patient. Then, if yes, go to step 404; if not, go to step 403.

In this embodiment, the execution subject may determine whether the user input information includes an image of the patient. As an example, when the user input information obtained by the execution main body does not include an image, the user input information may not include an image of the patient by identification (e.g., null or the like). Optionally, whether the user input information includes the image of the patient may be determined through a programming language such as javascript. The execution subject may determine all images included in the user input information as images of the patient, or may determine whether the images included in the user input information are images of the patient by using techniques such as image recognition. For example, when the execution subject or other electronic device determines that the images included in the user input information are lesion images, the execution subject may determine that all the images included in the user input information are images of the patient.

Based on the user input information, a first set of probabilities is determined, step 403. Thereafter, step 405 is performed.

In this embodiment, the execution subject may determine the first set of probabilities based on user input information. Wherein the first probability is a probability determined based on the image that the patient has a disease of a predetermined set of diseases.

As an implementation manner, the execution main body may send the user input information to an electronic device (e.g., a mobile phone, a computer, etc.) used by a doctor or other persons with diagnosis capability. The performing agent may then receive probabilities returned by an electronic device used by a doctor or other diagnostically-capable person that a patient has a disease of a predetermined set of diseases, and use the received set of probabilities as a first set of probabilities.

As another implementation, in a case where the user input information does not include an image of the patient, the execution subject may determine each of the first probabilities in the first set of probabilities as an equal one of the numerical values. For example, each first probability of the first set of probabilities may be determined to be 0.

Based on the image, a first set of probabilities is determined, step 404. Thereafter, step 405 is performed.

In this embodiment, step 404 is substantially the same as step 202 in the corresponding embodiment of fig. 2, and is not described herein again.

Step 405, determining the disease from the disease set as the target disease based on the first probability set. Thereafter, step 406 is performed.

In this embodiment, step 405 is substantially the same as step 203 in the corresponding embodiment of fig. 2, and is not described herein again.

Step 406, outputting disease information of the target disease.

In this embodiment, step 406 is substantially the same as step 204 in the corresponding embodiment of fig. 2, and is not described herein again.

As can be seen from fig. 4, compared to the embodiment corresponding to fig. 2, the flow 400 of the method for outputting information in the present embodiment highlights the step of determining the first set of probabilities in case the user input information does not comprise an image of the patient. Therefore, the scheme described in the embodiment can determine the mode of the first probability set under more conditions, so that the disease suffered by the patient can be diagnosed through the text information and the image of the patient, and the flexibility and the accuracy of diagnosis are improved.

Turning next to fig. 5, a flow 500 of yet another embodiment of a method for outputting information is shown. The process 500 of the method for outputting information includes the following steps:

step 501, obtaining user input information. Thereafter, step 502 is performed.

In this embodiment, step 501 is substantially the same as step 401 in the embodiment corresponding to fig. 4, and is not described here again.

Step 502, determine if the user input information includes an image of the patient. Then, if yes, go to step 504, otherwise go to step 503

In this embodiment, step 502 is substantially the same as step 402 in the embodiment corresponding to fig. 4, and is not described here again.

Based on the user input information, a first set of probabilities is determined, step 503. Thereafter, step 505 is performed.

In this embodiment, step 503 is substantially the same as step 403 in the embodiment corresponding to fig. 4, and is not described here again.

Based on the image, a first set of probabilities is determined, step 504. Thereafter, step 505 is performed.

In this embodiment, step 504 is substantially the same as step 404 in the embodiment corresponding to fig. 4, and is not described here again.

Based on the first set of probabilities, a question for the patient is generated, step 505. Thereafter, step 506 is performed.

In this embodiment, the execution subject may further generate a question for the patient based on the first set of probabilities. The problem for the patient may be a disease-related problem determined by the performing subject and possibly suffered by the patient.

As an example, the above-described problem generated by the executing body may be a problem for symptoms that may be produced by a patient having a target disease, for example, when the executing body determines that the patient may have urticaria, a problem "whether red swelling is present", "whether itching is present", or the like may be generated. When the generated problem is that whether the user has a rhinorrhea, the feedback information is no, and the target disease is urticaria, it may be stated that the probability obtained by the convolutional neural network is inaccurate, and in this scenario, the feedback information (for example, "yes" or "no") may be used as feedback to determine whether the probability obtained by the convolutional neural network is accurate, so as to continue training the convolutional neural network.

It is understood that the execution subject may generate a problem for information that cannot be represented by the image based on the first set of probabilities. Based on the generated questions, the performing subject can make a diagnosis (i.e., determine the target disease) more accurate.

For example, when the executing subject determines that the disease suffered by the patient may be "skin allergy" (i.e., the disease in the disease set corresponding to the first probability with the largest value in the first probability set is "skin allergy"), the questions generated by the executing subject for the patient may be "whether the skin is red", "whether the skin is itchy", and the like.

Here, the problem of the execution subject generation may be associated with a symptom possibly accompanied by a disease corresponding to the maximum 1 or more first probabilities in the first probability set, may be a problem related to age, sex, fertility status, and the like, or may be a problem for distinguishing two or more diseases.

In some optional implementations of this embodiment, the executing main body may execute the step 505 as follows:

first, based on the first set of probabilities, disease identities of diseases in the set of diseases that the patient has are determined. Here, the execution subject may determine, as the disease markers of the diseases in the disease set that the patient has, the disease markers of the disease corresponding to the one or more first probabilities with the largest numerical value in the first probability set.

Then, through the disease identification and the pre-trained bayesian network, a question and an output for the patient are generated based on the information gain maximization or the weight sampling (for example, selecting the question or symptom information corresponding to the disease identification and corresponding to the one or more weights with the highest weight). The Bayesian network can be used for representing the corresponding relation between the disease identification and the problem and can also be used for representing the corresponding relation between the disease identification and the symptom. When a bayesian network is used to characterize the correspondence of a disease identity to a symptom, the executing agent may generate a question for the symptom.

As an example, the bayesian network may be obtained by training the executing entity or other electronic device according to the following steps:

first, a set of training samples is obtained. The training sample includes images of the patient, information corresponding to the images included in the training sample (including feedback information and/or user input information), and disease identification corresponding to the images included in the training sample.

Then, by using a machine learning algorithm (for example, a maximum expectation algorithm, a gradient ascent method, or the like), for each training sample in the training sample set, the images and information included in the training sample are used as inputs, the disease markers included in the training sample are used as expected outputs, and the initial bayesian network is trained to obtain a trained bayesian network.

In some optional implementations of this embodiment, the directed acyclic graph of the bayesian network is a directed bipartite graph, a starting point of the directed bipartite graph is used to characterize the disease identification, an end point of the directed bipartite graph is used to characterize the symptom information, and the conditional probability in the bayesian network is used to characterize the probability that a patient of the disease characterized by the disease identification produces the symptom characterized by the symptom information.

As an example, referring to fig. 6, as shown in fig. 6, the structure of the bayesian network is a directed bipartite graph, the

disease identifiers

601, 602, 603 are the starting points of the nodes of the directed bipartite graph, the

symptom information

604, 605, 606, 607 are the end points of the nodes of the directed bipartite graph, and the edges of the directed bipartite graph (i.e., the conditional probabilities in the bayesian network) are used to characterize the probability that the patient of the disease characterized by the disease identifier produces the symptom characterized by the symptom information.

Turning now to fig. 5.

Step 506, feedback information for the problem is obtained. Thereafter, step 507 is performed.

In this embodiment, the execution body may acquire feedback information for the problem. Wherein, the feedback information may be information replied by the user for the question. For example, when the question is "ask how much your body temperature is", the feedback information may be "36.9"; when the question is "whether fever is present", the feedback information may be "yes".

It will be appreciated that questioning the patient, and obtaining feedback information regarding the questions, helps the performing agent to further improve the accuracy of the diagnosis.

In some optional implementations of this embodiment, the executing body may further continue to train the bayesian network based on the feedback information.

first, a set of training samples is obtained. The training sample comprises an image of a patient, feedback information corresponding to the image comprised by the training sample and a disease identifier corresponding to the image comprised by the training sample. Wherein the feedback information may be information obtained for one or more questions and fed back by the user.

Then, by using a machine learning algorithm (for example, a maximum expectation algorithm, a gradient ascent method, or the like), for each training sample in the training sample set, the image and the feedback information included in the training sample are used as inputs, the disease identifier included in the training sample is used as an expected output, and the initial bayesian network is trained to obtain a trained bayesian network.

Step 507, obtaining the prior probability of the diseases in the disease set. Thereafter, step 508 is performed.

In this embodiment, the executing entity may obtain the prior probability of the diseases in the disease set from other electronic devices communicatively connected thereto or locally. The prior probability may be manually given by a professional (e.g., medical staff) and stored via the execution subject or the electronic device communicatively connected thereto, or may be obtained by counting diseases in a disease set and prevalence distribution of the diseases.

Based on the feedback information and the user input information, a second set of probabilities is determined, step 508. Thereafter, step 509 is performed.

In this embodiment, the execution subject may determine the second probability set based on the feedback information and the user input information. Wherein the second probability may be a probability that the patient has a disease of a predetermined set of diseases.

As an example, the executing entity may determine the second probability set as follows:

first, the execution body may send the feedback information and the user input information to an electronic device (e.g., a mobile phone, a computer, etc.) used by a doctor or other person with diagnosis capability.

The performing agent may then receive a probability returned by an electronic device used by a doctor or other diagnostically-capable person that the patient has a disease of a predetermined set of diseases, and use the received set of probabilities as a second set of probabilities.

As yet another example, the executing entity may determine the second probability set as follows:

first, a set of training samples is obtained. The training sample includes images of the patient, user information (including feedback information and user input information) corresponding to the images included in the training sample, and disease signatures corresponding to the images included in the training sample.

Then, by using a machine learning algorithm (for example, a maximum expectation algorithm, a gradient ascent method, or the like), for each training sample in the training sample set, the image and the inverse user information included in the training sample are used as input, the disease identifier included in the training sample is used as an expected output, and the initial bayesian network is trained to obtain a trained bayesian network.

And finally, taking the set of the conditional probabilities of the trained Bayesian network as the second probability set.

Step 509, determining a score for the patient having a disease of the set of diseases based on the first set of probabilities, the second set of probabilities, and the obtained prior probabilities. Thereafter, step 510 is performed.

In this embodiment, the executing entity may determine a score for the patient having a disease in the set of diseases based on the first probability set, the second probability set, and the obtained prior probabilities. Wherein the score characterizes a magnitude of likelihood that the patient has a disease of the set of diseases. Each disease in the set of diseases corresponds to a first probability, a second probability and a prior probability.

As an example, the executive agent may determine the score of a disease in a set of diseases according to the following formula:

S＝p1×(p2+λ×p3)

wherein S is the score, p1 is the second probability, p2 is the prior probability, p3 is the first probability, λ is the weight set by the technician, and the value of λ may be a number greater than or equal to 0.

As another example, when the executing main body sequentially generates a plurality of questions and sequentially acquires feedback information of the user for the questions, a score of a patient having a disease in a disease set after each acquisition of the feedback information may be distinguished by a flag, and thus, the executing main body may determine the score of the disease in the disease set according to the following formula:

S(i,t)＝p(t)×(p’+λ×f_i)

wherein i is a disease marker, and t isTime, S (i, t) is the score of the disease with disease identification i at time t, p (t) is the second probability at time t, p' is the prior probability, f_iFor a first probability of a disease identified as i, λ is a weight set by the technician, and the value of λ may be a number greater than or equal to 0.

Based on the determined scores, it is determined whether to continue generating questions, step 510. If not, go to step 511; if yes, go to step 505.

In this embodiment, the execution subject may determine whether to continue generating the question based on the determined score.

As an example, the executing entity may arrange the obtained scores in order from big to small, and then, the executing entity may determine whether to continue generating the question according to the following formula:

where r is the disease number (i.e., rank), the letter l is the number of another disease, i is a disease, i_rFor characterizing diseases of sequence number r, i_lFor characterizing diseases of sequence number l, k₁Number of target diseases, k, to be determined, set for the user or the technician₂Numerical values set for the technician, typically k₂When the size is larger, the determined target disease is more accurate, and if n is the number of diseases in the disease set, k is₁、k₂N generally satisfies: 0<k₁<k₂N, t is time, score (i)_rT) score of disease with index r at time t, score (i)_lAnd t) is the score of the disease with the serial number l at the t-th time and is a preset threshold value larger than 0.

According to the formula, k is before ranking₁Sum of scores for individual diseases minus rank k₁+1 to k₂The executive agent may not continue to generate questions when the difference in the sum of disease scores of (a) is greater than a threshold; k when ranking₁Score sum of individual diseases minus rank k₁+1 to k₂And the difference is less than or equal to the threshold, the executive may continue to generate questions.

As yet another example, the execution subject may not continue to generate a question when the ratio of the score with the largest determined score median to the sum of the determined scores is greater than a preset ratio threshold (e.g., 50%, 60%, etc.); when the ratio of the score with the maximum score to the sum of the scores is less than or equal to the preset ratio threshold, the executing body can continue to generate the problem.

Step 511, determining the disease from the disease set, wherein the patient has the disease as the target disease. Thereafter, step 512 is performed.

In this embodiment, the execution subject may determine a disease from the disease set, which the patient has, as the target disease.

As an example, the performing subject may determine one or more diseases corresponding to one or more scores with the largest score among the determined scores as diseases (i.e., target diseases) suffered by the patient.

Step 512, outputting the disease information of the target disease.

In this embodiment, step 512 is substantially the same as step 406 in the embodiment corresponding to fig. 4, and is not described herein again.

As can be seen from fig. 5, compared to the embodiment corresponding to fig. 2, the flow 500 of the method for outputting information in the present embodiment highlights the step of determining the target disease from the set of diseases based on the first set of probabilities, and the step of determining the first set of probabilities in case the user input information does not comprise an image of the patient. Therefore, the scheme described in the embodiment can introduce more ways of determining the target disease and ways of determining the first probability set in more cases, thereby enriching the diagnosis means of the disease and contributing to improving the accuracy of diagnosis.

With further reference to fig. 7, as an implementation of the method shown in the above figures, the present application provides an embodiment of an apparatus for outputting information, which corresponds to the embodiment of the method shown in fig. 2, and which may include the same or corresponding features or effects as the embodiment of the method shown in fig. 2, in addition to the features described below. The device can be applied to various electronic equipment.

As shown in fig. 7, the apparatus 700 for outputting information of the present embodiment includes: an acquisition unit 701, a first determination unit 702, a second determination unit 703, and an output unit 704. Wherein the obtaining unit 701 is configured to obtain user input information; the first determining unit 702 is configured to determine, based on the image, a first set of probabilities in response to the user input information comprising the image of the patient, wherein the first probability is a probability determined based on the image that the patient has a disease of a predetermined set of diseases; the second determining unit 703 is configured to determine, based on the first set of probabilities, a disease from the set of diseases as the target disease; the output unit 704 is configured to output disease information of the target disease.

In this embodiment, the obtaining unit 701 of the apparatus 700 for outputting information may obtain the user input information from other electronic devices, or locally, through a wired connection manner or a wireless connection manner. The user input information may be various information input by the user. The user input information may include, but is not limited to, at least one of: age information, marriage and childbirth status information, age information, disease symptom information, examination index information, and the like. The presentation form of the user input information may include at least one of: text, audio, image. The image may be a picture or a video. The image may include, but is not limited to, at least one of: computed Tomography (CT) images, Computed Radiography (CR) images, Magnetic Resonance examination (MR) images, direct Digital flat panel X-ray imaging (DR) images, and the like.

In this embodiment, in a case where the user input information includes an image of a patient, the above-described first determination unit 702 may determine the first probability set based on the image. Wherein the first probability is a probability that the patient has a disease in a predetermined set of diseases determined based on the image, a number of the first probabilities in the first set of probabilities may be the same as a number of the diseases in the set of diseases, and the first probabilities and the diseases may be in a one-to-one correspondence relationship. The image included in the input information may be an image of a lesion of the patient. The set of diseases may be a set of all diseases; or a set of all diseases of one or more departments, for example, the set of diseases may be a set of all diseases of dermatology; the disease in the set of diseases can also be a disease with a characteristic, e.g., the disease in the set of diseases can be a disease that can characterize a focus of the patient by an image (e.g., fracture, cervical spondylosis, lumbar disc herniation).

In this embodiment, the second determining unit 703 may determine, as the target disease, a disease from the disease set, which the patient has, based on the first probability set. The target disease may be one disease or a plurality of diseases. The number of target diseases may be set by the user as described above, or may be set by a technician.

In this embodiment, the output unit 704 may output disease information of the target disease. The disease information may be a one-to-one correspondence relationship with a disease, and the disease information may be introduction information and treatment information of the disease, or may be a disease identifier for identifying the disease, where the disease identifier may include, but is not limited to, one of the following: disease name, disease code.

In some optional implementations of the present embodiment, the first determining unit 702 is further configured to: and inputting the image into a pre-trained convolutional neural network to obtain a first probability set. The convolutional neural network may be used, among other things, to characterize the correspondence between the images and the probability of the patient having each disease in the set of diseases.

In some optional implementations of this embodiment, the apparatus 700 further includes: a third determination unit (not shown in the figures) is configured to determine the first set of probabilities based on the user input information in response to the user input information not comprising an image of the patient. Wherein the first probability is a probability determined based on the image that the patient has a disease of a predetermined set of diseases.

In some optional implementations of this embodiment, the second determining unit 703 is further configured to: generating a problem for the patient based on the first set of probabilities; acquiring feedback information aiming at a problem; based on the feedback information and the user input information, a disease from which the patient suffers is determined from the set of diseases as a target disease. The problem for the patient may be a disease-related problem that the patient may have, as determined by the apparatus 700 described above. The feedback information may be information that the user replied to for the question. For example, when the question is "ask how much your body temperature is", the feedback information may be "36.9"; when the question is "whether fever is present", the feedback information may be "yes".

In some optional implementations of this embodiment, the second determining unit 703 is further configured to: acquiring prior probability of diseases in a disease set; determining a second set of probabilities based on the feedback information and the user input information, wherein the second probability is a probability that the patient has a disease of the predetermined set of diseases determined based on the feedback information and the user input information; determining a score for the patient having a disease of the set of diseases based on the first set of probabilities, the second set of probabilities, and the obtained prior probabilities, wherein the score characterizes a magnitude of likelihood that the patient has a disease of the set of diseases; determining whether to continue generating questions based on the determined scores; in response to determining not to continue generating the question, determining a disease from the set of diseases that the patient has as the target disease. The prior probability may be manually given by a professional (e.g., medical staff) and stored via the apparatus 700 or the electronic device communicatively connected thereto, or may be obtained by counting diseases in a disease set and prevalence distribution of the diseases. The second probability may be a probability that the patient has a disease of a predetermined set of diseases. Each disease in the set of diseases corresponds to a first probability, a second probability and a prior probability.

In some optional implementations of this embodiment, the bayesian network has a structure of a directed bipartite graph, where a start point of the directed bipartite graph is used to characterize the disease identification, an end point of the directed bipartite graph is used to characterize the symptom information, and a conditional probability in the bayesian network is used to characterize a probability that a patient of the disease characterized by the disease identification produces the symptom characterized by the symptom information.

In some optional implementations of this embodiment, the apparatus 700 further includes: the training unit (not shown in the figure) is configured to continue training the bayesian network based on the feedback information.

In some optional implementations of this embodiment, the second determining unit 703 is further configured to: determining disease identities of the diseases in the set of diseases that the patient has based on the first set of probabilities; based on the disease identification and the pre-trained bayesian network, a question and an output for the patient are generated. The Bayesian network can be used for representing the corresponding relation between the disease identification and the problem and can also be used for representing the corresponding relation between the disease identification and the symptom. When a bayesian network is used to characterize the correspondence of a disease identity to a symptom, the apparatus 700 described above can generate a question for the symptom.

The apparatus provided by the above embodiment of the present application acquires the user input information through the acquisition unit 701, then the first determination unit 702 determines a first probability set based on the image in response to the user input information including the image of the patient, where the first probability is a probability that the patient has a disease in a predetermined disease set determined based on the image, then the second determination unit 703 determines the disease that the patient has as a target disease from the disease set based on the first probability set, and finally the output unit 704 outputs the disease information of the target disease, thereby enriching the generation manner of the information, improving the accuracy of the diagnosis of the electronic device, and contributing to saving the diagnosis time of the medical staff.

Referring now to FIG. 8, a block diagram of a computer system 800 suitable for use in implementing the control device of an embodiment of the present application is shown. The control device shown in fig. 8 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.

As shown in fig. 8, the computer system 800 includes a Central Processing Unit (CPU)801 that can perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)802 or a program loaded from a storage section 808 into a Random Access Memory (RAM) 803. In the RAM 803, various programs and data necessary for the operation of the system 800 are also stored. The CPU 801, ROM 802, and RAM 803 are connected to each other via a bus 804. An input/output (I/O) interface 805 is also connected to bus 804.

The following components are connected to the I/O interface 805: an input portion 806 including a keyboard, a mouse, and the like; an output section 807 including a signal such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage portion 808 including a hard disk and the like; and a communication section 809 including a network interface card such as a LAN card, a modem, or the like. The communication section 809 performs communication processing via a network such as the internet. A drive 810 is also connected to the I/O interface 805 as necessary. A removable medium 811 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 810 as necessary, so that a computer program read out therefrom is mounted on the storage section 808 as necessary.

In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication section 809 and/or installed from the removable medium 811. The computer program performs the above-described functions defined in the method of the present application when executed by the Central Processing Unit (CPU) 801.

It should be noted that the computer readable medium described herein can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In this application, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations for aspects of the present application may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of 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 the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described in the embodiments of the present application may be implemented by software or hardware. The described units may also be provided in a processor, and may be described as: a processor includes an acquisition unit, a first determination unit, a second determination unit, and an output unit. The names of these units do not in some cases constitute a limitation on the unit itself, and for example, the acquisition unit may also be described as a "unit that acquires user input information".

As another aspect, the present application also provides a computer-readable medium, which may be contained in the electronic device described in the above embodiments; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: acquiring user input information; in response to the user input information including an image of the patient, determining a first set of probabilities based on the image, wherein the first probability is a probability that the patient has a disease of a predetermined set of diseases determined based on the image; determining a disease from the disease set as a target disease based on the first set of probabilities; and outputting the disease information of the target disease.

The above description is only a preferred embodiment of the application and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention herein disclosed is not limited to the particular combination of features described above, but also encompasses other arrangements formed by any combination of the above features or their equivalents without departing from the spirit of the invention. For example, the above features may be replaced with (but not limited to) features having similar functions disclosed in the present application.

Claims

1. An apparatus for outputting information, comprising:

an acquisition unit configured to acquire user input information;

a first determining unit configured to determine a first set of probabilities based on an image of a patient in response to the user input information comprising the image, wherein a first probability is a probability that the patient has a disease of a predetermined set of diseases determined based on the image;

a second determination unit configured to determine, as a target disease, a disease from which the patient suffers from the set of diseases based on the first set of probabilities;

wherein the second determination unit is further configured to:

generating a question for the patient for information not embodied by the images based on the first set of probabilities; acquiring feedback information aiming at the problem; determining a second set of probabilities based on the feedback information and the user input information, wherein a second probability is a probability that the patient has a disease of a predetermined set of diseases determined based on feedback information and user input information; determining a score for the patient having a disease in the set of diseases based on the first set of probabilities, the second set of probabilities, and a prior probability of the disease in the set of diseases; determining whether to continue generating questions based on the determined scores;

an output unit configured to output disease information of the target disease;

the second determination unit is further configured to:

determining, based on the first set of probabilities, disease identities of diseases in the set of diseases that the patient has; generating a question and an output for the patient based on a disease identification and a pre-trained Bayesian network;

the Bayesian network has a structure of a directed bipartite graph, wherein the starting point of the directed bipartite graph is used for representing disease identification, the end point of the directed bipartite graph is used for representing symptom information, and the conditional probability in the Bayesian network is used for representing the probability that a patient with a disease represented by the disease identification generates symptoms represented by the symptom information;

the second determination unit is further configured to perform the determining whether to continue generating the question based on the determined score as follows:

arranging the obtained scores in descending order, and k is arranged before ranking₁Score sum of individual diseases minus rank k₁+1 to k₂When the difference value of the disease score sum is larger than the threshold value, the problem is not generated continuously; k when ranking₁Score sum of individual diseases minus rank k₁+1 to k₂If the difference between the disease scores of (1) is less than or equal to the threshold value, continuing to generate the question.

2. The apparatus of claim 1, wherein the first determining unit is further configured to:

and inputting the image into a pre-trained convolutional neural network to obtain a first probability set.

3. The apparatus of claim 2, wherein the apparatus further comprises:

a third determination unit configured to determine a first set of probabilities based on the user input information in response to the user input information not including an image of a patient.

4. The apparatus of claim 1, wherein the score characterizes a magnitude of a likelihood that the patient has a disease of a set of diseases;

the second determination unit is further configured to:

obtaining a prior probability of a disease in the set of diseases;

determining whether to continue generating questions based on the determined scores;

in response to determining not to continue generating questions, determining a disease from the set of diseases that the patient has as a target disease.

5. The apparatus of claim 1, wherein the apparatus further comprises:

a training unit configured to continue training the Bayesian network based on the feedback information.

6. An electronic device, comprising:

one or more processors;

a storage device having one or more programs stored thereon,

when executed by the one or more processors, cause the one or more processors to perform operations comprising:

acquiring user input information;

in response to the user input information including an image of a patient, determining a first set of probabilities based on the image, wherein a first probability is a probability that the patient has a disease of a predetermined set of diseases determined based on the image;

determining, from the set of diseases, a disease from which the patient suffers as a target disease based on the first set of probabilities;

wherein the determining, from the set of diseases, a disease from which the patient has as a target disease based on the first set of probabilities comprises: generating a question for the patient based on the first set of probabilities; acquiring feedback information aiming at the problem; determining a second set of probabilities based on the feedback information and the user input information, wherein a second probability is a probability that the patient has a disease of a predetermined set of diseases determined based on feedback information and user input information; determining a score for the patient having a disease in the set of diseases based on the first set of probabilities, the second set of probabilities, and a prior probability of the disease in the set of diseases; determining whether to continue generating questions based on the determined scores;

outputting disease information of the target disease;

the generating, based on the first set of probabilities, a question for the patient for information not embodied by the images, comprising:

the determining whether to continue generating questions based on the determined scores includes: arranging the obtained scores in descending order, and k is arranged before ranking₁Score sum of individual diseases minus rank k₁+1 to k₂When the difference value of the disease score sum is larger than the threshold value, the problem is not generated continuously; k when ranking₁Score sum of individual diseases minus rank k₁+1 to k₂Is less than or equal to a threshold valueThe question continues to be generated.

7. A computer-readable medium, on which a computer program is stored, wherein the program, when executed by a processor, performs the operations of:

acquiring user input information;

outputting disease information of the target disease;

the determining whether to continue generating questions based on the determined scores includes: