CN111477310A - Triage data processing method and device, computer equipment and storage medium - Google Patents
Triage data processing method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN111477310A CN111477310A CN202010142969.8A CN202010142969A CN111477310A CN 111477310 A CN111477310 A CN 111477310A CN 202010142969 A CN202010142969 A CN 202010142969A CN 111477310 A CN111477310 A CN 111477310A
- Authority
- CN
- China
- Prior art keywords
- symptom
- result
- model
- information
- triage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- General Business, Economics & Management (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Computational Biology (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Epidemiology (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a diagnosis data processing method, a diagnosis data processing device, computer equipment and a storage medium, wherein the method comprises the following steps: receiving a triage request, and acquiring patient information; acquiring symptom information from the patient information by a maximum word matching method; inputting the symptom information into a combined prediction model, and performing prediction processing on the symptom information through the combined prediction model to obtain a first symptom department set; inputting symptom information into a reinforcement learning triage model, and acquiring a first triage result which is output after a first action is executed and contains a first symptom result and a first state result; determining a first symptom result in the first diagnosis result as a final symptom result when the first state result is the first state; the final symptom result is the department at which the patient visits the clinic. The invention realizes the preprocessing of dimensionality reduction through the probability model and the deep neural network model and the automatic triage based on the reinforcement learning method so as to determine the department of the patient with respect to the disease, improve the accuracy of the diagnosis and improve the experience of the patient.
Description
Technical Field
The present invention relates to the field of data processing, and in particular, to a method and an apparatus for processing triage data, a computer device, and a storage medium.
Background
At present, when a patient goes to a hospital for a doctor, the patient firstly needs to go to a diagnosis separating table for manual diagnosis, a large amount of queuing time is consumed by the patient in the process, and higher requirements are provided for the depth and the breadth of professional knowledge of service personnel of the diagnosis separating table.
Disclosure of Invention
The invention provides a diagnosis data processing method, a diagnosis data processing device, computer equipment and a storage medium, which realize the dimensionality reduction preprocessing through a probability model and a deep neural network model and automatic diagnosis based on a reinforcement learning method, can quickly and accurately determine departments of patients needing to be affected, improve the accuracy of diagnosis and improve the experience of the patients.
A triage data processing method, comprising:
receiving a triage request, and acquiring patient information;
acquiring symptom information from the patient information by a maximum word matching method;
inputting the symptom information into a combined prediction model, and performing prediction processing on the symptom information through the combined prediction model to obtain a first symptom department set output by the combined prediction model;
inputting the symptom information into a reinforcement learning triage model, and acquiring a first triage result output by the reinforcement learning triage model after executing a first action; the first action is selected from the first action space set after the reinforcement learning triage model analyzes and processes the input symptom information, and the first action space set is a preset action space total set which is activated by the first symptom department set and then output; the first triage result comprises a first symptom result and a first status result;
determining the first symptom result in the first diagnosis result as a final symptom result when the first state result is a first state; the final symptom result is the department at which the patient is attending.
A triage data processing apparatus comprising:
the receiving module is used for receiving the triage request and acquiring the patient information;
the acquisition module is used for acquiring symptom information from the patient information by a maximum word matching method;
the prediction module is used for inputting the symptom information into a combined prediction model, performing prediction processing on the symptom information through the combined prediction model, and acquiring a first symptom department set output by the combined prediction model;
the activation module is used for inputting the symptom information into the reinforcement learning triage model and acquiring a first triage result output by the reinforcement learning triage model after executing a first action; the first action is selected from the first action space set after the reinforcement learning triage model analyzes and processes the input symptom information, and the first action space set is a preset action space total set which is activated by the first symptom department set and then output; the first triage result comprises a first symptom result and a first status result;
the output module is used for determining the first symptom result in the first diagnosis result as a final symptom result when the first state result is in a first state; the final symptom result is the department at which the patient is attending.
A computer device comprising a memory, a processor and a computer program stored in said memory and executable on said processor, said processor implementing the steps of the triage data processing method described above when executing said computer program.
A computer-readable storage medium, which stores a computer program that, when executed by a processor, implements the steps of the triage data processing method described above.
The triage data processing method, the triage data processing device, the computer equipment and the storage medium provided by the invention can acquire accurate symptom information from the patient information by a maximum word matching method, input the symptom information into a combined prediction model formed by combining a trained symptom prediction probability model and a trained department deep convolutional neural network model, acquire the first symptom department set, activate and output the action space total set through the first symptom department set, realize that the action space total set is used as the first action space set in the reinforcement learning triage model after being subjected to dimension reduction processing, input the symptom information into the reinforcement learning triage model, acquire the first triage result output after executing the first action, and when the first state result in the triage result is in a first state, and determining the first symptom result in the triage results as a final symptom result. The method realizes the preprocessing of reducing the dimension through the probability model and the deep neural network model, and carries out automatic triage based on the reinforcement learning method, can quickly and accurately determine the department of the patient who needs to be sick, saves the time of the patient, improves the accuracy of seeing a doctor, and improves the experience of the patient.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
FIG. 1 is a schematic diagram of an application environment of a triage data processing method according to an embodiment of the present invention;
FIG. 2 is a flow chart of a triage data processing method according to an embodiment of the present invention;
FIG. 3 is a flow chart of a triage data processing method according to another embodiment of the present invention;
FIG. 4 is a flowchart of step S10 of the triage data processing method according to an embodiment of the present invention;
FIG. 5 is a flowchart illustrating step S20 of the triage data processing method according to an embodiment of the present invention;
FIG. 6 is a flowchart of step S30 of the triage data processing method according to an embodiment of the present invention;
FIG. 7 is a flowchart of step S301 of the triage data processing method according to an embodiment of the present invention;
FIG. 8 is a flowchart illustrating the step S302 of the triage data processing method according to an embodiment of the present invention;
FIG. 9 is a functional block diagram of a triage data processing apparatus in an embodiment of the present invention;
FIG. 10 is a schematic diagram of a computer device in an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The triage data processing method provided by the invention can be applied to the application environment shown in fig. 1, wherein a client (computer device) communicates with a server through a network. The client (computer device) includes, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, cameras, and portable wearable devices. The server may be implemented as a stand-alone server or as a server cluster consisting of a plurality of servers.
In an embodiment, as shown in fig. 2, a method for processing triage data is provided, which mainly includes the following steps S10-S50:
and S10, receiving the triage request and acquiring the patient information.
Understandably, after receiving the triage request, the patient information is acquired, where the triage request is a request triggered after selecting and confirming the patient information that needs to be triaged, and the triggering manner may be set according to a requirement, for example, a trigger button that can be triggered by clicking, sliding, or the like is provided on an application platform interface, and the trigger is automatically triggered after a preset program is executed, or the like.
Wherein the patient information inputs information for the patient relating to the patient's symptoms.
In an embodiment, as shown in fig. 4, before the step S10, that is, before the receiving the triage request and acquiring the patient information, the method includes:
and S101, receiving a patient input instruction and acquiring patient input information.
Understandably, the patient input instruction is received, the patient input information is acquired, the patient input instruction is an instruction triggered after the patient input information is input on a display interface of an application program, the patient input information is acquired after the patient input instruction is received, and the acquisition mode of the patient input information can be set as required, for example, the acquisition mode can be that the patient input information is acquired through the patient input instruction, the patient input information is acquired according to a storage path of the patient input information contained in the patient input instruction, and the like.
S102, inputting the patient input information into a preset preprocessing model, and recognizing the patient input information by the preprocessing model to obtain a recognition result; wherein the recognition result includes text, speech, and images.
Understandably, the pre-processing model is a preset model for recognizing the patient input information, the patient input information is input to the pre-processing model, the pre-processing model can determine the recognition result according to the format of the patient input information, and the recognition result comprises text, voice and images.
And S103, acquiring a conversion model corresponding to the identification result.
Understandably, determining a conversion model corresponding to the recognition result according to the recognition result, wherein the conversion model comprises a text conversion model, a voice conversion model and an image conversion model, namely, if the recognition result is a text, obtaining the text conversion model, if the recognition result is a voice, obtaining the voice conversion model, if the recognition result is an image, obtaining the image conversion model, and the conversion model is a trained neural network model, so that obtaining the more targeted conversion model can improve the conversion efficiency and the accuracy.
And S104, inputting the patient input information into the conversion model, and performing text conversion on the patient input information by the conversion model to output a conversion result.
Understandably, the patient input information is input to a conversion model corresponding to the recognition result, and conversion into the conversion result is performed by the conversion model.
And S105, determining the conversion result as the patient information.
Therefore, through the recognition of texts, voices and images carried out on the patient input information input by the patient, different conversion models are corresponding to different recognition results, the patient information is obtained from the patient input information, various input channels are provided for the patient, and the patient experience is improved.
And S20, obtaining symptom information from the patient information through a maximum word matching method.
Understandably, splitting a plurality of single texts from the patient information by the maximum word matching method, performing maximum lexization on the plurality of single texts, namely combining the single texts with the corresponding front and rear single texts to generate a front text, a rear text and a full text, acquiring the single texts, the front text, the rear text and the full text with the highest matching value in a preset symptom word bank, determining the text with the highest matching value as the maximum word group corresponding to the single texts, performing zero clearing on all the maximum word groups, namely removing the maximum word group with the matching value of zero, and determining all the maximum word groups after the zero clearing as the symptom information.
In one embodiment, as shown in fig. 5, the step S20 of obtaining symptom information in the patient information by the maximum word matching method includes:
s201, acquiring a preset symptom word bank; the symptom lexicon comprises a plurality of symptom words.
Understandably, the symptom word bank is a bank for storing all symptom words, that is, the symptom word bank contains a plurality of symptom words, the symptom word bank can add the symptom words according to the requirement, and the symptom word bank can add the symptom words with the change of time, and the symptom words are words given a symptom name, for example: the symptom thesaurus can screen symptom words of common symptoms in symcat public data.
S202, splitting the patient information into a plurality of single texts.
Understandably, the text can be a single character or a single phrase, for example: the patient information is "yesterday to today coughs all the time", and then split into "yesterday", "to", "today", "always", "coughs".
S203, obtaining the starting position and the ending position of the single text, combining the previous single text of the starting position with the single text to generate a front text, combining the next single text of the ending position with the single text to generate a back text, and combining the previous single text of the starting position, the single text and the next single text of the ending position to generate a full text.
Understandably, the starting position may be set according to requirements, for example, the starting position may be a number of starting digits of the single text from the first position from left to right in the patient information, and then the number of the starting digits may reach the single text; the ending position can also be set according to requirements, for example, the ending position can be obtained by increasing the number of digits of the single text after the starting position; combining the single text before the start position with the single text to generate a front text, combining the single text after the end position with the single text to generate a back text, and combining the single text before the start position, the single text and the single text after the end position to generate a full text, for example: the patient information is that the patient has epigastric pain for three days, the plurality of separated single texts are that the patient has epigastric pain for three days, the abdominal pain is used as a single text, the front text is that the patient has epigastric pain, the rear text is that the patient has abdominal pain for three days, and the full text is that the patient has epigastric pain for three days.
And S204, obtaining matching values of the single text, the preposed text, the postposed text and the full-positioned text with texts in the symptom word stock, and determining the text with the highest matching value as the maximum word group corresponding to the single text.
Understandably, the matching value is a number of characters matching consistently, the single text and the text in the symptom lexicon are searched and matched to obtain a matching value corresponding to the single text, the pre-text and the text in the symptom lexicon are searched and matched to obtain a matching value corresponding to the pre-text, the post-text and the text in the symptom lexicon are searched and matched to obtain a matching value corresponding to the post-text, the full text and the text in the symptom lexicon are searched and matched to obtain a matching value corresponding to the full text, for example: in the above example, the matching value for abdominal pain is 2, the matching value for epigastric pain is 3, the matching value for three days of abdominal pain is 0, and the matching value for three days of epigastric pain is 0.
And S205, performing zero clearing processing on all the maximum phrases corresponding to the single texts in the patient information, and determining all the maximum phrases after the zero clearing processing as the symptom information.
Understandably, the maximum phrases with the matching value of zero in all the maximum phrases corresponding to each single text are removed, and all the removed maximum phrases are determined as the symptom information.
Therefore, the accuracy of symptom acquisition can be ensured through the maximum word matching method, and the accuracy and the correctness of symptom acquisition are improved.
And S30, inputting the symptom information into a combined prediction model, and performing prediction processing on the symptom information through the combined prediction model to obtain a first symptom department set output by the combined prediction model.
Understandably, the combined prediction model is a model formed by combining a trained symptom prediction probability model and a trained department deep convolutional neural network model, the symptom prediction probability model is obtained by inputting a first symptom sample into a bayesian probability model for training, the department deep convolutional neural network model is obtained by inputting a second symptom sample into a deep neural network model for training, the first symptom department set is obtained by splicing and normalizing a prediction probability distribution result output by the symptom prediction probability model and a department prediction distribution result output by the department deep convolutional neural network model, the prediction probability distribution result is a distribution diagram of probability values corresponding to all symptom words, the department prediction distribution result is a distribution diagram of probability values corresponding to all departments, and the department is a department name set according to requirements, such as surgery, medicine, orthopedics, etc., the predicted probability distribution is a set of data, the outcome of the department predicted distribution is a set of data, and the set of first symptom departments is a set of arrays processed from the predicted probability distribution and the outcome of the department predicted distribution.
In an embodiment, as shown in fig. 6, in step S30, the inputting the symptom information into a combined prediction model, and performing prediction processing on the symptom information through the combined prediction model to obtain a first set of symptom departments output by the combined prediction model includes:
s301, inputting the symptom information into a trained symptom prediction probability model, predicting the symptom information through the symptom prediction probability model, and obtaining a prediction probability distribution result output by the symptom prediction probability model; wherein the predicted probability distribution result characterizes a matching probability distribution of symptoms associated with the symptom information in a symptom set.
Understandably, the symptom prediction probability model is a trained bayesian probability model, and the symptom information is input into the symptom prediction probability model to be subjected to prior distribution processing, so as to obtain the prediction probability distribution result, wherein the prediction probability distribution result is a matching probability distribution of symptoms related to the symptom information in the symptom set, the prediction probability distribution result is an array composed of percentage values ranging from 0% to 100%, and the symptom set is a total set of all symptom words.
In an embodiment, as shown in fig. 7, before the step S301, that is, before the step S301 inputs the symptom information into the trained symptom prediction probability model, the method includes:
s3011, obtaining a first symptom sample; wherein each of the first symptom samples is associated with a symptom category label.
The first symptom samples are collected symptom information, each first symptom sample is manually determined and is associated with a symptom category label, namely, each first symptom sample is labeled, and the symptom category label is a symptom word in the symptom set.
S3012, inputting the first symptom sample into a Bayesian probability model containing first initial parameters.
Understandably, the bayesian probability model is a model of a neural network structure formed based on a bayesian algorithm, and the bayesian probability model includes the first initial parameter, where the first initial parameter may be set according to requirements, such as setting the first initial parameter to a default preset value, or a random parameter value, and so on.
S3013, carrying out prior distribution processing on the first symptom sample through the Bayesian probability model.
Understandably, the prior distribution processing is probability distribution processing in the bayesian algorithm, and the prior distribution processing is performed on the first symptom sample.
S3014, obtaining a distribution result output by the Bayesian probability model, and determining a first loss value according to the distribution result and the matching degree of the symptom category label.
Understandably, performing probability distribution after prior distribution processing according to the bayesian probability model, wherein the probability distribution is a distribution diagram of probability values corresponding to all symptom words in the symptom set, obtaining a distribution result of the bayesian probability model, the distribution result is a distribution diagram of probability values corresponding to all the symptom words, obtaining the symptom words corresponding to all the probability values in the distribution result, and determining a loss value corresponding to the probability values by comparing the probability values corresponding to all the symptom words with symptom category labels of the first symptom sample, namely calculating the first loss value through a loss function of the bayesian probability model. S3015, when the first loss value reaches a preset first convergence condition, recording the converged Bayesian probability model as a trained symptom prediction probability model.
The preset first convergence condition may be a condition that the loss value is small and does not decrease again after 500 times of calculation, that is, when the loss value is small and does not decrease again after 500 times of calculation, the training is stopped, and the converged bayesian probability model is recorded as a trained symptom prediction probability model; the preset first convergence condition may also be a condition that the loss value is smaller than a set threshold, that is, when the loss value is smaller than the set threshold, the training is stopped, and the converged bayesian probability model is recorded as a trained symptom prediction probability model.
Therefore, the Bayesian probability model is trained by inputting the first symptom sample, and the accuracy and reliability of the distribution result can be improved.
In an embodiment, after the step S3014, that is, after the obtaining of the distribution result output by the bayesian probability model and determining the loss value according to the matching degree between the distribution result and the symptom category label, the method includes:
s3016, when the first loss value does not reach a preset first convergence condition, iteratively updating a first initial parameter of the Bayes probability model, and when the first loss value reaches the preset first convergence condition, recording the converged Bayes probability model as a trained symptom prediction probability model.
Therefore, when the first loss value does not reach the preset first convergence condition, the first initial parameter of the Bayesian probability model is continuously updated and iterated, so that the accurate distribution result can be continuously drawn close, and the accuracy of the distribution result is higher and higher.
S302, inputting the symptom information into a trained department deep convolution neural network model, and performing character feature extraction on the symptom information through the department deep convolution neural network model to obtain a department prediction distribution result output by the department deep convolution neural network model; wherein the department prediction distribution result characterizes a matching probability distribution of departments in the set of departments associated with the symptom information.
Understandably, the department deep convolutional neural network model is a trained deep neural network model, and text feature extraction is performed by inputting the symptom information into the deep convolutional neural network model, so that a department prediction distribution result can be obtained, the department prediction distribution result is a matching probability distribution of departments related to the symptom information in the department set, the department set is a total set of all departments of a hospital, and the department prediction distribution result is an array consisting of percentage values in a range of 0% to 100%.
In an embodiment, as shown in fig. 8, before the step S302, that is, before the step S302 of inputting the symptom information into the trained department deep convolutional neural network model, the method includes:
s3021, obtaining a second symptom sample; wherein each of the second symptom samples is associated with a department label.
The second symptom samples are collected symptom information, each second symptom sample is manually determined and is associated with a department label, namely, each second symptom sample is labeled, and the department labels are names corresponding to all departments.
And S3022, inputting the second symptom sample into a deep neural network model containing a second initial parameter.
Understandably, the deep neural network model is a model of a neural network structure based on multi-class recognition, the network structure of the deep neural network model can be set according to requirements, for example, the neural network structure can be selected as VGG, Goog L eNet, and the like, and the deep neural network model includes the second initial parameter, wherein the second initial parameter can be set according to requirements, for example, the second initial parameter is set to default to a preset value, or a random parameter value, and the like.
S3023, extracting character features in the symptom sample through the deep neural network model.
Understandably, the character features are text data converted into feature vectors, a more common method for extracting character features is extraction by a bag-of-words method, and the feature vectors corresponding to the characters are obtained according to the occurrence frequency.
And S3024, acquiring the recognition result output by the deep neural network model according to the character features, and determining a second loss value according to the matching degree of the recognition result and the department label.
Understandably, acquiring an identification result of the deep neural network model according to the character features extracted by the deep neural network model, wherein the identification result is a distribution graph of probability values corresponding to all departments, acquiring all probability values in the identification result, and determining a loss value corresponding to the probability values by comparing the probability values corresponding to all departments with department labels of the second symptom sample, namely calculating the second loss value through a loss function of the deep neural network model.
And S3025, when the second loss value reaches a preset second convergence condition, recording the converged deep neural network model as a trained department deep convolutional neural network model.
The preset second convergence condition may be a condition that the second loss value is small and does not decrease again after being calculated for 5000 times, that is, when the second loss value is small and does not decrease again after being calculated for 5000 times, the training is stopped, and the converged deep neural network model is recorded as a trained department deep convolutional neural network model; the preset second convergence condition may also be a condition that the second loss value is smaller than a set threshold, that is, when the second loss value is smaller than the set threshold, the training is stopped, and the converged deep neural network model is recorded as a trained department deep convolutional neural network model.
Therefore, the initial neural network model is trained according to the certificate type label value in the training image sample, and the identification can be carried out by extracting texture features, and the accuracy and reliability of the identification result are improved.
In an embodiment, after the step S3024, the method includes:
and S3026, when the second loss value does not reach a preset second convergence condition, iteratively updating a second initial parameter of the deep neural network model until the second loss value reaches the preset second convergence condition, and recording the converged deep neural network model as a trained department deep convolutional neural network model.
Therefore, when the second loss value does not reach the preset second convergence condition, the second initial parameter of the iterative deep neural network model is continuously updated, so that the iterative deep neural network model can be continuously drawn close to an accurate recognition result, and the accuracy of the recognition result is higher and higher.
And S303, splicing and normalizing the prediction probability distribution result and the department prediction distribution result to obtain the first symptom department set.
Understandably, the prediction probability distribution result and the department prediction distribution result are spliced, that is, an array corresponding to the prediction probability distribution result and an array corresponding to the department prediction distribution result are spliced to obtain a spliced array with the same dimension as the action space total set, the spliced array is normalized, that is, a percentage value which is greater than or equal to a preset percentage threshold value in the spliced array is normalized to 1, a percentage value which is smaller than the preset percentage threshold value in the spliced array is normalized to 0, and the normalized spliced array is determined as the first symptom department set.
Therefore, the symptom information is predicted through the symptom prediction probability model, the prediction probability distribution result of the matching probability distribution of the symptom relevant to the symptom information is obtained, the symptom information is predicted through the department depth convolution neural network model, the department prediction distribution result of the matching probability distribution of the department relevant to the symptom information is obtained, the prediction probability distribution result and the department prediction distribution result are spliced and normalized, the first symptom department set is obtained, reasonable prediction of the symptom information is achieved, and the activation factor is provided for the action space total set.
S40, inputting the symptom information into a reinforcement learning triage model, and acquiring a first triage result output by the reinforcement learning triage model after executing a first action; the first action is selected from the first action space set after the reinforcement learning triage model analyzes and processes the input symptom information, and the first action space set is a preset action space total set which is activated by the first symptom department set and then output; the first triage result includes a first symptom result and a first status result.
Understandably, the reinforcement learning triage model may set a learning method according to a requirement, and preferably, the reinforcement learning triage model is set as a DQN learning method, the symptom information is used as an Agent (Agent) of the reinforcement learning triage model, the reinforcement learning triage model guides the input symptom information, a first action is selected from the first action space set, the first action is used as an action to be performed on the symptom information, the action space total set is a set including all actions, the action space total set is a group of arrays, the action space total set has the same dimension as the array of the first symptom department set, the action space total set is activated by the first symptom department set and then outputs the first action space set, the activation is obtained by matching the action space total set with the first symptom department set according to an activation principle, the activation principle mode is that two arrays of array contents which correspond to each other one by one are matched, reserved or deleted, namely the array contents of the action space total set are judged to be reserved or deleted according to corresponding values in the arrays of the first symptom department set, if the value of the first symptom department set is 1, the array contents of the action space total set which correspond to the value of the symptom department set are reserved, if the value of the first symptom department set is 0, the array contents of the action space total set which correspond to the value of the symptom department set are deleted, the finally obtained first action space set is obtained after the preset action space total set is activated by the first symptom department set, the first action space set provides a set of all guided actions, and the reinforcement learning diagnosis division model outputs the first action after the symptom information is executed by the first action And (5) dividing diagnosis results.
Therefore, the action space total set is activated through the first symptom department set to output the first action space set, the action space total set is subjected to dimension reduction processing and then is used as the first action space set in the reinforcement learning triage model, the learning effect of the reinforcement learning triage model can be improved, and the accuracy of the reinforcement learning triage model is improved.
S50, determining the first symptom result in the first diagnosis result as a final symptom result when the first state result is the first state; the final symptom result is the department at which the patient is attending.
Understandably, the first state may be set as a termination state, that is, a final state is reached after interactive dialogue with the reinforcement learning triage model, and when the first state results in the first state, the first symptom result is determined as a final symptom result, and the symptom result is a department that needs to visit the patient.
According to the invention, accurate symptom information is obtained from the patient information through a maximum word matching method, the symptom information is input into the combined prediction model (a model formed by combining a trained symptom prediction probability model and a trained department deep convolutional neural network model), the first symptom department set is obtained, the action space total set is activated through the first symptom department set to output the first action space set, the action space total set is subjected to dimension reduction processing and then is used as the first action space set in the reinforcement learning triage model (the learning effect of the reinforcement learning triage model can be improved, and the accuracy of the reinforcement learning triage model is improved), and the symptom information is input into the reinforcement learning triage model to obtain the first action (selected from the first action space set after analysis processing of the symptom information) output after the first action is executed (the analysis processing of the symptom information) And determining the first symptom result in the triage results as a final symptom result (department in which the patient is present) if the first state result in the triage results is the first state (termination state).
Therefore, the invention realizes the pretreatment of dimension reduction through the probability model and the deep neural network model and the automatic triage based on the reinforcement learning method, can quickly and accurately determine the department of the patient who needs to be in the disease, saves the time of the patient, improves the accuracy of the diagnosis and improves the experience of the patient.
In one embodiment, the first triage result further comprises a first reward result; as shown in fig. 3, after the step S40, that is, after the obtaining of the first diagnosis result output by the reinforcement learning diagnosis model after the first action is performed, the method includes:
s60, when the first status result is the second status, using the first symptom result in the first diagnosis result as the next symptom information, and associating the first reward result with the next symptom information.
Understandably, the second state can be a non-terminated state, i.e. the final state is not reached after the interactive dialogue with the reinforcement learning triage model, and at this time, taking the first symptom result as the input of the next interactive dialogue with the reinforcement learning diagnosis model, using the first symptom result as next symptom information, the first triage result further comprising the first reward result, the first reward result is a reward value given to the first action after the reinforcement learning triage model executes the first action, the reward value can be a value assigned each time, or can be a value accumulated according to the number of interactive conversations, the reinforcement learning triage model can be provided with the reward value to approach to an accurate clinic (direction of maximizing the reward value), and the first reward result is related to the next symptom information.
And S70, inputting the next symptom information into the combined prediction model, and performing prediction processing on the symptom information through the combined prediction model to obtain a second symptom department set output by the combined prediction model.
Understandably, the second set of symptom departments is derived by inputting the next symptom information into the combined predictive model, the second set of symptom departments also being a set of arrays, the second set of symptom departments having the same dimensions as the first set of symptom departments.
S80, inputting the next symptom information and the first reward result associated with the next symptom information into the reinforcement learning triage model, and obtaining a second diagnosis result output by the reinforcement learning triage model after executing a second action; the second action is selected from the second action space set after the reinforcement learning triage model analyzes and processes the input first symptom information and the first reward result, and the second action space set is output after the action space total set is activated by the second symptom department set; the second diagnosis result includes a second symptom result and a second status result.
Understandably, the reinforcement learning triage model analyzes the input next symptom information and the first reward result associated with the next symptom information, so as to guide the next symptom information, selects a second action from the second action space set, wherein the second action is an action to be performed on the symptom information, the action space set has the same array dimension as the second symptom department set, the action space set is output after being activated by the second symptom department set, the second action space set provides a set of all guided actions, and the reinforcement learning triage model passes the second triage result output after performing the second action on the symptom information.
S90, determining the second symptom result in the second diagnosis result as a final symptom result when the second state result is the first state; the final symptom result is the department at which the patient is attending.
Therefore, the reinforcement learning triage model learns from the interactive dialogue and guides the direction of the optimal strategy to be close, when the reinforcement learning triage model is in the second state (non-termination state), the reinforcement learning triage model continuously learns and interacts with the dialogue until the reinforcement learning triage model is in the first state (termination state), so that the final symptom result is output, and the diagnosis accuracy is improved.
In an embodiment, a triage data processing apparatus is provided, and the triage data processing apparatus corresponds to the triage data processing method in the above embodiment one to one. As shown in fig. 9, the triage data processing apparatus includes a receiving module 11, an obtaining module 12, a predicting module 13, an activating module 14, and an outputting module 15.
The functional modules are explained in detail as follows:
the receiving module 11 is configured to receive a triage request and acquire patient information;
an obtaining module 12, configured to obtain symptom information from the patient information by a maximum word matching method;
the prediction module 13 is configured to input the symptom information into a combined prediction model, perform prediction processing on the symptom information through the combined prediction model, and obtain a first symptom department set output by the combined prediction model;
the activation module 14 is configured to input the symptom information into a reinforcement learning triage model, and obtain a first triage result output by the reinforcement learning triage model after a first action is executed; the first action is selected from the first action space set after the reinforcement learning triage model analyzes and processes the input symptom information, and the first action space set is a preset action space total set which is activated by the first symptom department set and then output; the first triage result comprises a first symptom result and a first status result;
an output module 15, configured to determine the first symptom result in the first diagnosis result as a final symptom result when the first state result is a first state; the final symptom result is the department at which the patient is attending.
In one embodiment, the activation module 14 includes:
a reward unit, configured to, when the first state result is a second state, take the first symptom result in the first diagnosis result as next symptom information, and associate the first reward result with the next symptom information;
the input unit is used for inputting the next symptom information into the combined prediction model, performing prediction processing on the symptom information through the combined prediction model, and acquiring a second symptom department set output by the combined prediction model;
the activation unit is used for inputting the next symptom information and the first reward result associated with the next symptom information into the reinforcement learning triage model and acquiring a second diagnosis result output by the reinforcement learning triage model after a second action is executed; the second action is selected from the second action space set after the reinforcement learning triage model analyzes and processes the input first symptom information and the first reward result, and the second action space set is output after the action space total set is activated by the second symptom department set; the second diagnosis result comprises a second symptom result and a second state result;
a determination unit, configured to determine the second symptom result in the second diagnosis result as a final symptom result when the second state result is the first state; the final symptom result is the department at which the patient is attending.
In one embodiment, the receiving module 11 includes:
the receiving unit is used for receiving a patient input instruction and acquiring patient input information;
the identification unit is used for inputting the patient input information into a preset preprocessing model, and the preprocessing model identifies the patient input information to obtain an identification result; wherein the recognition result comprises text, speech and images;
the selection unit is used for acquiring a conversion model corresponding to the identification result;
the conversion unit is used for inputting the patient input information into the conversion model, and the conversion model performs text conversion on the patient input information and outputs a conversion result;
a result output unit for determining the conversion result as the patient information.
In one embodiment, the obtaining module 12 includes:
the first acquisition unit is used for acquiring a preset symptom word bank; the symptom word bank comprises a plurality of symptom words;
the splitting unit is used for splitting the patient information into a plurality of single texts;
a second obtaining unit, configured to obtain a start position and an end position of the single text, combine a single text before the start position with the single text to generate a pre-positioned text, combine a single text after the end position with the single text to generate a post-positioned text, and combine the single text before the start position, the single text, and the single text after the end position to generate a full-positioned text;
the matching unit is used for acquiring matching values of the single text, the preposed text, the postpositive text and the full-position text with texts in the symptom word stock, and determining the text with the highest matching value as a maximum word group corresponding to the single text;
and the zero clearing unit is used for carrying out zero clearing treatment on all the maximum phrases corresponding to the single texts in the patient information and determining all the maximum phrases after the zero clearing treatment as the symptom information.
In one embodiment, the prediction module 13 comprises:
the first model unit is used for inputting the symptom information into a trained symptom prediction probability model, predicting the symptom information through the symptom prediction probability model and acquiring a prediction probability distribution result output by the symptom prediction probability model; wherein the predicted probability distribution result characterizes a matching probability distribution of symptoms associated with the symptom information in a symptom set;
the second model unit is used for inputting the symptom information into a trained department deep convolution neural network model, performing character feature extraction on the symptom information through the department deep convolution neural network model, and acquiring a department prediction distribution result output by the department deep convolution neural network model; wherein the department prediction distribution result characterizes a matching probability distribution of departments in a set of departments associated with the symptom information;
and the splicing unit is used for splicing and normalizing the prediction probability distribution result and the department prediction distribution result to obtain the first symptom department set.
In one embodiment, the first model element comprises:
a first obtaining subunit, configured to obtain a first symptom sample; wherein each of the first symptom samples is associated with a symptom category label;
the first input subunit is used for inputting the first symptom sample into a Bayesian probability model containing first initial parameters;
the first processing subunit is used for carrying out prior distribution processing on the first symptom sample through the Bayesian probability model;
the first output subunit is used for acquiring a distribution result output by the Bayesian probability model and determining a first loss value according to the matching degree of the distribution result and the symptom category label;
and the first convergence subunit is used for recording the converged Bayesian probability model as a trained symptom prediction probability model when the first loss value reaches a preset first convergence condition.
In one embodiment, the second model element comprises:
a second obtaining subunit, configured to obtain a second symptom sample; wherein each of the second symptom samples is associated with a department label;
a second input subunit, configured to input the second symptom sample into a deep neural network model including a second initial parameter;
the second processing subunit is used for extracting character features in the symptom sample through the deep neural network model;
the second output subunit is used for acquiring a recognition result output by the deep neural network model according to the character features and determining a second loss value according to the matching degree of the recognition result and the department label;
and the second convergence subunit is used for recording the converged deep neural network model as a trained department deep convolutional neural network model when the second loss value reaches a preset second convergence condition.
For specific limitations of the triage data processing apparatus, reference may be made to the above limitations on the triage data processing method, which are not described herein again. The modules in the triage data processing device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 10. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of triage data processing.
In one embodiment, a computer device is provided, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the triage data processing method in the above embodiments is implemented.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the triage data processing method in the above-described embodiments.
It will be understood by those of ordinary skill in the art that all or a portion of the processes of the methods of the embodiments described above may be implemented by a computer program that may be stored on a non-volatile computer-readable storage medium, which when executed, may include the processes of the embodiments of the methods described above, wherein any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (10)
1. A method for processing triage data, comprising:
receiving a triage request, and acquiring patient information;
acquiring symptom information from the patient information by a maximum word matching method;
inputting the symptom information into a combined prediction model, and performing prediction processing on the symptom information through the combined prediction model to obtain a first symptom department set output by the combined prediction model;
inputting the symptom information into a reinforcement learning triage model, and acquiring a first triage result output by the reinforcement learning triage model after executing a first action; the first action is selected from the first action space set after the reinforcement learning triage model analyzes and processes the input symptom information, and the first action space set is a preset action space total set which is activated by the first symptom department set and then output; the first triage result comprises a first symptom result and a first status result;
determining the first symptom result in the first diagnosis result as a final symptom result when the first state result is a first state; the final symptom result is the department at which the patient is attending.
2. The triage data processing method of claim 1, wherein the first triage outcome further includes a first reward outcome;
namely, after obtaining the first diagnosis result output by the reinforcement learning diagnosis model after executing the first action, the method includes:
when the first state result is a second state, taking the first symptom result in the first diagnosis result as next symptom information, and simultaneously associating the first reward result with the next symptom information;
inputting the next symptom information into the combined prediction model, and performing prediction processing on the symptom information through the combined prediction model to obtain a second symptom department set output by the combined prediction model;
inputting the next symptom information and the first reward result associated with the next symptom information into the reinforcement learning triage model, and acquiring a second diagnosis result output by the reinforcement learning triage model after a second action is executed; the second action is selected from the second action space set after the reinforcement learning triage model analyzes and processes the input first symptom information and the first reward result, and the second action space set is output after the action space total set is activated by the second symptom department set; the second diagnosis result comprises a second symptom result and a second state result;
determining the second symptom result in the second diagnosis result as a final symptom result when the second state result is the first state; the final symptom result is the department at which the patient is attending.
3. The triage data processing method of claim 1, wherein the step of receiving the triage request and obtaining the patient information comprises:
receiving a patient input instruction, and acquiring patient input information;
inputting the patient input information into a preset preprocessing model, and recognizing the patient input information by the preprocessing model to obtain a recognition result; wherein the recognition result comprises text, speech and images;
acquiring a conversion model corresponding to the recognition result;
inputting the patient input information into the conversion model, and performing text conversion on the patient input information by the conversion model to output a conversion result;
determining the conversion result as the patient information.
4. The triage data processing method of claim 1, wherein obtaining symptom information in the patient information by a maximum word matching method comprises:
acquiring a preset symptom word bank; the symptom word bank comprises a plurality of symptom words;
splitting the patient information into a plurality of single texts;
acquiring a starting position and an ending position of the single text, combining the previous single text of the starting position with the single text to generate a front text, combining the next single text of the ending position with the single text to generate a back text, and combining the previous single text of the starting position, the single text and the next single text of the ending position to generate a full text;
acquiring matching values of the single texts, the front texts, the rear texts and the full texts with texts in the symptom word stock, and determining the texts with the highest matching values as the maximum phrases corresponding to the single texts;
and performing zero clearing treatment on all the maximum phrases corresponding to the single texts in the patient information, and determining all the maximum phrases after zero clearing treatment as the symptom information.
5. The triage data processing method of claim 1, wherein the step of inputting the symptom information into a combined prediction model, and performing prediction processing on the symptom information through the combined prediction model to obtain a first set of symptom departments output by the combined prediction model comprises:
inputting the symptom information into a trained symptom prediction probability model, predicting the symptom information through the symptom prediction probability model, and obtaining a prediction probability distribution result output by the symptom prediction probability model; wherein the predicted probability distribution result characterizes a matching probability distribution of symptoms associated with the symptom information in a symptom set;
inputting the symptom information into a trained department deep convolution neural network model, and performing character feature extraction on the symptom information through the department deep convolution neural network model to obtain a department prediction distribution result output by the department deep convolution neural network model; wherein the department prediction distribution result characterizes a matching probability distribution of departments in a set of departments associated with the symptom information;
and splicing and normalizing the prediction probability distribution result and the department prediction distribution result to obtain the first symptom department set.
6. The triage data processing method of claim 5, wherein the entering the symptom information into a trained symptom prediction probability model comprises:
obtaining a first symptom sample; wherein each of the first symptom samples is associated with a symptom category label;
inputting the first symptom sample into a Bayesian probability model containing first initial parameters;
carrying out prior distribution processing on the first symptom sample through the Bayesian probability model;
obtaining a distribution result output by the Bayesian probability model, and determining a first loss value according to the distribution result and the matching degree of the symptom category label;
and when the first loss value reaches a preset first convergence condition, recording the converged Bayesian probability model as a trained symptom prediction probability model.
7. The triage data processing method of claim 5, wherein inputting the symptom information before the trained department deep convolutional neural network model comprises:
obtaining a second symptom sample; wherein each of the second symptom samples is associated with a department label;
inputting the second symptom sample into a deep neural network model containing second initial parameters;
extracting character features in the symptom sample through the deep neural network model;
acquiring a recognition result output by the deep neural network model according to the character features, and determining a second loss value according to the matching degree of the recognition result and the department label;
and when the second loss value reaches a preset second convergence condition, recording the converged deep neural network model as a trained department deep convolutional neural network model.
8. A triage data processing apparatus, comprising:
the receiving module is used for receiving the triage request and acquiring the patient information;
the acquisition module is used for acquiring symptom information from the patient information by a maximum word matching method;
the prediction module is used for inputting the symptom information into a combined prediction model, performing prediction processing on the symptom information through the combined prediction model, and acquiring a first symptom department set output by the combined prediction model;
the activation module is used for inputting the symptom information into the reinforcement learning triage model and acquiring a first triage result output by the reinforcement learning triage model after executing a first action; the first action is selected from the first action space set after the reinforcement learning triage model analyzes and processes the input symptom information, and the first action space set is a preset action space total set which is activated by the first symptom department set and then output; the first triage result comprises a first symptom result and a first status result;
the output module is used for determining the first symptom result in the first diagnosis result as a final symptom result when the first state result is in a first state; the final symptom result is the department at which the patient is attending.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the triage data processing method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the triage data processing method according to any one of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010142969.8A CN111477310A (en) | 2020-03-04 | 2020-03-04 | Triage data processing method and device, computer equipment and storage medium |
| PCT/CN2020/099063 WO2021174728A1 (en) | 2020-03-04 | 2020-06-30 | Triage data processing method and apparatus, computer device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010142969.8A CN111477310A (en) | 2020-03-04 | 2020-03-04 | Triage data processing method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111477310A true CN111477310A (en) | 2020-07-31 |
Family
ID=71747991
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010142969.8A Pending CN111477310A (en) | 2020-03-04 | 2020-03-04 | Triage data processing method and device, computer equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111477310A (en) |
| WO (1) | WO2021174728A1 (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111951943A (en) * | 2020-09-27 | 2020-11-17 | 平安科技（深圳）有限公司 | Intelligent triage method and device, electronic equipment and storage medium |
| CN112016295A (en) * | 2020-09-04 | 2020-12-01 | 平安科技（深圳）有限公司 | Symptom data processing method and device, computer equipment and storage medium |
| CN112017742A (en) * | 2020-09-08 | 2020-12-01 | 平安科技（深圳）有限公司 | Triage data processing method and device, computer equipment and storage medium |
| CN112164454A (en) * | 2020-10-10 | 2021-01-01 | 联仁健康医疗大数据科技股份有限公司 | Diagnosis prediction method and device and electronic equipment |
| CN112927787A (en) * | 2021-03-22 | 2021-06-08 | 塔米智能科技（北京）有限公司 | Diagnosis method, diagnosis device, diagnosis equipment and diagnosis storage medium |
| CN112951394A (en) * | 2021-03-10 | 2021-06-11 | 中电健康云科技有限公司 | Method for hospital triage and medical examination item prediction based on deep learning |
| WO2021151327A1 (en) * | 2020-09-09 | 2021-08-05 | 平安科技（深圳）有限公司 | Triage data processing method and apparatus, and device and medium |
| WO2021164388A1 (en) * | 2020-09-25 | 2021-08-26 | 平安科技（深圳）有限公司 | Triage fusion model training method, triage method, apparatus, device, and medium |
| CN113409932A (en) * | 2021-07-21 | 2021-09-17 | 挂号网（杭州）科技有限公司 | Triage method, triage device, computer equipment and storage medium |
| WO2022227203A1 (en) * | 2021-04-30 | 2022-11-03 | 平安科技（深圳）有限公司 | Triage method, apparatus and device based on dialogue representation, and storage medium |
| CN116936080A (en) * | 2023-07-27 | 2023-10-24 | 中日友好医院(中日友好临床医学研究所) | Preliminary diagnosis guiding method and device based on dialogue and electronic medical record |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114724691B (en) * | 2022-04-25 | 2024-09-24 | 中国科学院自动化研究所 | Intelligent emergency diagnosis and sub-diagnosis system based on psychological model |
| CN115132303A (en) * | 2022-04-28 | 2022-09-30 | 腾讯科技（深圳）有限公司 | Physiological label prediction method, model training method, device, equipment and medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109829038A (en) * | 2018-12-11 | 2019-05-31 | 平安科技（深圳）有限公司 | Question and answer feedback method, device, equipment and storage medium based on deep learning |
| CN110085308B (en) * | 2019-04-23 | 2022-02-25 | 挂号网（杭州）科技有限公司 | Diagnosis and treatment department classification method based on fusion deep learning |
| CN110047584A (en) * | 2019-04-23 | 2019-07-23 | 清华大学 | Hospital distributing diagnosis method, system, device and medium based on deep learning |
| CN110675944A (en) * | 2019-09-20 | 2020-01-10 | 京东方科技集团股份有限公司 | Triage method and device, computer equipment and medium |
-
2020
- 2020-03-04 CN CN202010142969.8A patent/CN111477310A/en active Pending
- 2020-06-30 WO PCT/CN2020/099063 patent/WO2021174728A1/en active Application Filing
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021151328A1 (en) * | 2020-09-04 | 2021-08-05 | 平安科技（深圳）有限公司 | Symptom data processing method and apparatus, and computer device and storage medium |
| CN112016295A (en) * | 2020-09-04 | 2020-12-01 | 平安科技（深圳）有限公司 | Symptom data processing method and device, computer equipment and storage medium |
| CN112016295B (en) * | 2020-09-04 | 2024-02-23 | 平安科技（深圳）有限公司 | Symptom data processing method, symptom data processing device, computer equipment and storage medium |
| CN112017742A (en) * | 2020-09-08 | 2020-12-01 | 平安科技（深圳）有限公司 | Triage data processing method and device, computer equipment and storage medium |
| CN112017742B (en) * | 2020-09-08 | 2024-05-24 | 平安科技（深圳）有限公司 | Triage data processing method, triage data processing device, computer equipment and storage medium |
| WO2021151356A1 (en) * | 2020-09-08 | 2021-08-05 | 平安科技（深圳）有限公司 | Triage data processing method and apparatus, and computer device and storage medium |
| WO2021151327A1 (en) * | 2020-09-09 | 2021-08-05 | 平安科技（深圳）有限公司 | Triage data processing method and apparatus, and device and medium |
| WO2021164388A1 (en) * | 2020-09-25 | 2021-08-26 | 平安科技（深圳）有限公司 | Triage fusion model training method, triage method, apparatus, device, and medium |
| CN111951943B (en) * | 2020-09-27 | 2021-01-05 | 平安科技（深圳）有限公司 | Intelligent triage method and device, electronic equipment and storage medium |
| WO2021179625A1 (en) * | 2020-09-27 | 2021-09-16 | 平安科技（深圳）有限公司 | Smart diagnosis guide method and apparatus, electronic device, and storage medium |
| CN111951943A (en) * | 2020-09-27 | 2020-11-17 | 平安科技（深圳）有限公司 | Intelligent triage method and device, electronic equipment and storage medium |
| CN112164454A (en) * | 2020-10-10 | 2021-01-01 | 联仁健康医疗大数据科技股份有限公司 | Diagnosis prediction method and device and electronic equipment |
| CN112951394A (en) * | 2021-03-10 | 2021-06-11 | 中电健康云科技有限公司 | Method for hospital triage and medical examination item prediction based on deep learning |
| CN112927787A (en) * | 2021-03-22 | 2021-06-08 | 塔米智能科技（北京）有限公司 | Diagnosis method, diagnosis device, diagnosis equipment and diagnosis storage medium |
| WO2022227203A1 (en) * | 2021-04-30 | 2022-11-03 | 平安科技（深圳）有限公司 | Triage method, apparatus and device based on dialogue representation, and storage medium |
| CN113409932A (en) * | 2021-07-21 | 2021-09-17 | 挂号网（杭州）科技有限公司 | Triage method, triage device, computer equipment and storage medium |
| CN116936080A (en) * | 2023-07-27 | 2023-10-24 | 中日友好医院(中日友好临床医学研究所) | Preliminary diagnosis guiding method and device based on dialogue and electronic medical record |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021174728A1 (en) | 2021-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111477310A (en) | Triage data processing method and device, computer equipment and storage medium | |
| CN108986908B (en) | Method and device for processing inquiry data, computer equipment and storage medium | |
| KR102217457B1 (en) | A chat service providing system that can provide medical consultation according to customer's needs with a chat robot | |
| US20140122109A1 (en) | Clinical diagnosis objects interaction | |
| CN112016295B (en) | Symptom data processing method, symptom data processing device, computer equipment and storage medium | |
| US20130268203A1 (en) | System and method for disease diagnosis through iterative discovery of symptoms using matrix based correlation engine | |
| WO2018158385A1 (en) | Psychotherapy triage method | |
| CN112287068B (en) | Artificial intelligence-based inquiry dialogue data processing method and device | |
| US11742087B2 (en) | Processing clinical notes using recurrent neural networks | |
| US11907273B2 (en) | Augmenting user responses to queries | |
| CN111933291A (en) | Medical information recommendation device, method, system, equipment and readable storage medium | |
| WO2021151356A1 (en) | Triage data processing method and apparatus, and computer device and storage medium | |
| CN110931137A (en) | Machine-assisted dialog system, method and device | |
| CN112634889B (en) | Electronic case input method, device, terminal and medium based on artificial intelligence | |
| CN110223134B (en) | Product recommendation method based on voice recognition and related equipment | |
| CN114613523A (en) | Doctor allocation method, device, storage medium and equipment for on-line medical inquiry | |
| CN112349435A (en) | Online inquiry method, device, server, medium and system | |
| CN114628001A (en) | Prescription recommendation method, system, equipment and storage medium based on neural network | |
| CN113782163A (en) | Information pushing method and device and computer readable storage medium | |
| CN113012774A (en) | Automatic medical record encoding method and device, electronic equipment and storage medium | |
| CN112231556B (en) | User image drawing method, device, equipment and medium based on dialogue scene | |
| CN114743647A (en) | Medical data processing method, device, equipment and storage medium | |
| CN110931002A (en) | Human-computer interaction method and device, computer equipment and storage medium | |
| US11727685B2 (en) | System and method for generation of process graphs from multi-media narratives | |
| CN116719840A (en) | Medical information pushing method based on post-medical-record structured processing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220922 Address after: Room 2601 (Unit 07), Qianhai Free Trade Building, No. 3048, Xinghai Avenue, Nanshan Street, Qianhai Shenzhen-Hong Kong Cooperation Zone, Shenzhen, Guangdong 518000 Applicant after: Shenzhen Ping An Smart Healthcare Technology Co.,Ltd. Address before: 1-34 / F, Qianhai free trade building, 3048 Xinghai Avenue, Mawan, Qianhai Shenzhen Hong Kong cooperation zone, Shenzhen, Guangdong 518000 Applicant before: Ping An International Smart City Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right |