CN-116775832-B - Pre-consultation method, device, equipment and medium based on artificial intelligence

CN116775832BCN 116775832 BCN116775832 BCN 116775832BCN-116775832-B

Abstract

The application relates to the technical field of artificial intelligence and intelligent medical treatment and discloses an artificial intelligence-based pre-consultation method, device, equipment and medium, wherein the method comprises the steps of inputting an i-1 th round of consultation step and pre-consultation dialogue summarization data of the i-th round of the pre-consultation step prediction model corresponding to a target doctor to predict the i-th round of the consultation step, wherein the target doctor is a doctor which the target patient wants to visit, inputting the i-th round of the consultation step and the pre-consultation dialogue summarization data of the i-th round of the pre-consultation step prediction model corresponding to the target doctor to generate a reply text, obtaining a target reply text, and displaying the target reply text based on an virtual image corresponding to the target doctor. Therefore, the automatic pre-consultation is realized, the time for the doctor to take the consultation is shortened, the accuracy of the pre-consultation is improved, and the missed diagnosis and the misdiagnosis are avoided.

Inventors

GU YUE
YIN JINGYU
TIAN JINGTAO

Assignees

平安科技（深圳）有限公司

Dates

Publication Date: 20260512
Application Date: 20230620

Claims (9)

1. A pre-interrogation method based on artificial intelligence, the method comprising: Acquiring i-th round of pre-consultation dialogue summarized data, wherein the i-th round of pre-consultation dialogue summarized data comprises 1-th round to i-th round of single-round pre-consultation dialogue data, and the single-round pre-consultation dialogue data is dialogue data for performing one round of pre-consultation on a target patient; inputting the i-1 th round of inquiry steps and the pre-inquiry dialogue summarized data of the i-th round of inquiry steps into an inquiry step prediction model corresponding to a target doctor to predict the i-th round of inquiry steps, wherein the target doctor is a doctor which the target patient wants to visit; Inputting the inquiry step of the ith round and the pre-inquiry dialogue summarized data into a pre-inquiry reply generation model corresponding to the target doctor to generate a reply text, so as to obtain a target reply text; displaying the target reply text based on the virtual image corresponding to the target doctor; the step of inputting the i-1 th round of inquiry step and the i-th round of pre-inquiry dialogue summarized data into an inquiry step prediction model corresponding to a target doctor to predict the i-th round of inquiry step comprises the following steps: Adopting a preset first splicing symbol to splice the inquiry step of the ith wheel-1 and the pre-inquiry dialogue summary data of the ith wheel to obtain first splicing data; inputting the first spliced data into the inquiry step prediction model corresponding to the target doctor to predict the inquiry step of the ith round, wherein the inquiry step prediction model is a classification model obtained based on a Bert model and classification layer training.
2. The method for pre-consultation based on artificial intelligence according to claim 1, wherein the step of inputting the consultation step of the ith round and the pre-consultation dialogue summary data into a pre-consultation reply generation model corresponding to the target doctor to generate a reply text and obtain a target reply text comprises the following steps: Inputting the inquiry step of the ith round and the pre-inquiry dialogue summarized data into the pre-inquiry reply generation model corresponding to the target doctor to generate a reply text, so as to obtain an initial reply text; And inputting the initial reply text into a text style conversion model corresponding to the target doctor to perform text style conversion, so as to obtain the target reply text.
3. The method for pre-consultation based on artificial intelligence according to claim 1, wherein the step of inputting the consultation step of the ith round and the pre-consultation dialogue summary data into a pre-consultation reply generation model corresponding to the target doctor to generate a reply text and obtain a target reply text comprises the following steps: extracting symptom information of the ith round according to the pre-consultation dialogue summary data of the ith round; Adopting a preset second splicing symbol to splice the inquiry step of the ith round, the symptom information of the ith round and the pre-inquiry dialogue summary data of the ith round to obtain second splicing data; and inputting the second spliced data into the pre-consultation reply generation model corresponding to the target doctor to generate a reply text, so as to obtain the target reply text, wherein the pre-consultation reply generation model is a model obtained based on GPT training.
4. The artificial intelligence based pre-consultation method according to claim 1, wherein the step of displaying the target reply text based on the avatar corresponding to the target doctor includes: performing voice conversion on the target reply text based on the voice characteristics corresponding to the target doctor to obtain target voice; and displaying the target voice based on the virtual image corresponding to the target doctor, wherein the virtual image corresponding to the target doctor is controlled based on facial expression characteristics and action characteristics corresponding to the target doctor in the displaying process.
5. The artificial intelligence based pre-consultation method of claim 1 including: Acquiring a pre-consultation ending signal corresponding to the target patient; Responding to the pre-consultation ending signal, and acquiring each single-round pre-consultation dialogue data corresponding to the target patient as data to be analyzed; and inputting the data to be analyzed into a pre-consultation conclusion generation model corresponding to the target doctor to generate a pre-consultation conclusion, so as to obtain an initial pre-consultation conclusion.
6. The artificial intelligence based pre-consultation method according to claim 5, wherein the step of inputting the data to be analyzed into a pre-consultation conclusion generation model corresponding to the target doctor to generate a pre-consultation conclusion, after the step of obtaining an initial pre-consultation conclusion, further comprises: Acquiring a supplementary inquiry signal; Responding to the supplementary inquiry signal, and displaying the initial pre-inquiry conclusion; Acquiring inquiry supplemental information input by the target doctor based on the displayed initial pre-inquiry conclusion; acquiring a supplement completion signal; and responding to the supplement completion signal, inputting the inquiry supplementary information and the data to be analyzed into a pre-inquiry conclusion generation model corresponding to the target doctor to generate a pre-inquiry conclusion, so as to obtain a target pre-inquiry conclusion.
7. An artificial intelligence based pre-interrogation device, the device comprising: The data acquisition module is used for acquiring the i-th round of pre-consultation dialogue summarized data, wherein the i-th round of pre-consultation dialogue summarized data comprises all the single-round pre-consultation dialogue data from the 1 st round to the i-th round, and the single-round pre-consultation dialogue data is dialogue data for carrying out one round of pre-consultation on a target patient; the inquiry step prediction module is used for inputting the inquiry step of the ith round and the inquiry dialogue summarized data of the ith round into an inquiry step prediction model corresponding to a target doctor to predict the inquiry step of the ith round, wherein the target doctor is a doctor which the target patient wants to visit; The reply text generation module is used for inputting the inquiry step of the ith round and the pre-inquiry dialogue summarization data into a pre-inquiry reply generation model corresponding to the target doctor to generate a reply text so as to obtain a target reply text; The display module is used for displaying the target reply text based on the virtual image corresponding to the target doctor; the step of inputting the i-1 th round of inquiry step and the i-th round of pre-inquiry dialogue summarized data into an inquiry step prediction model corresponding to a target doctor to predict the i-th round of inquiry step comprises the following steps: Adopting a preset first splicing symbol to splice the inquiry step of the ith wheel-1 and the pre-inquiry dialogue summary data of the ith wheel to obtain first splicing data; inputting the first spliced data into the inquiry step prediction model corresponding to the target doctor to predict the inquiry step of the ith round, wherein the inquiry step prediction model is a classification model obtained based on a Bert model and classification layer training.
8. 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 steps of the artificial intelligence based pre-interrogation method according to any one of claims 1 to 6 when the computer program is executed.
9. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the steps of the artificial intelligence based pre-interrogation method of any of claims 1 to 6.

Description

Pre-consultation method, device, equipment and medium based on artificial intelligence Technical Field The invention relates to the technical field of artificial intelligence and intelligent medical treatment, in particular to an artificial intelligence-based pre-consultation method, device, equipment and medium. Background With the increasing population, medical resources are increasingly scarce. At present, a doctor can only ask step by step to obtain information of a patient by aiming at the inquiry process of the patient, the inquiry process is long, so that the diagnosis receiving efficiency of the doctor is difficult to improve, and the patient sometimes expresses the symptoms of the patient without knowing the patient, so that the inquiry process is further prolonged, and even missed diagnosis and misdiagnosis are possible. Disclosure of Invention Based on the above, it is necessary to provide a pre-consultation method, device, equipment and medium based on artificial intelligence aiming at the technical problems that the prior art is gradually inquired by doctors to obtain the information of patients, the consultation process is longer, and even missed diagnosis and misdiagnosis can occur. In a first aspect, there is provided an artificial intelligence based pre-interrogation method, the method comprising: Acquiring i-th round of pre-consultation dialogue summarized data, wherein the i-th round of pre-consultation dialogue summarized data comprises 1-th round to i-th round of single-round pre-consultation dialogue data, and the single-round pre-consultation dialogue data is dialogue data for performing one round of pre-consultation on a target patient; inputting the i-1 th round of inquiry steps and the pre-inquiry dialogue summarized data of the i-th round of inquiry steps into an inquiry step prediction model corresponding to a target doctor to predict the i-th round of inquiry steps, wherein the target doctor is a doctor which the target patient wants to visit; Inputting the inquiry step of the ith round and the pre-inquiry dialogue summarized data into a pre-inquiry reply generation model corresponding to the target doctor to generate a reply text, so as to obtain a target reply text; and displaying the target reply text based on the virtual image corresponding to the target doctor. In a second aspect, there is provided an artificial intelligence based pre-consultation apparatus, the apparatus comprising: The data acquisition module is used for acquiring the i-th round of pre-consultation dialogue summarized data, wherein the i-th round of pre-consultation dialogue summarized data comprises all the single-round pre-consultation dialogue data from the 1 st round to the i-th round, and the single-round pre-consultation dialogue data is dialogue data for carrying out one round of pre-consultation on a target patient; the inquiry step prediction module is used for inputting the inquiry step of the ith round and the inquiry dialogue summarized data of the ith round into an inquiry step prediction model corresponding to a target doctor to predict the inquiry step of the ith round, wherein the target doctor is a doctor which the target patient wants to visit; The reply text generation module is used for inputting the inquiry step of the ith round and the pre-inquiry dialogue summarization data into a pre-inquiry reply generation model corresponding to the target doctor to generate a reply text so as to obtain a target reply text; and the display module is used for displaying the target reply text based on the virtual image corresponding to the target doctor. In a third aspect, a computer device is provided, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the artificial intelligence based pre-interrogation method described above when the computer program is executed. In a fourth aspect, a computer readable storage medium is provided, the computer readable storage medium storing a computer program which, when executed by a processor, implements the steps of the artificial intelligence based pre-interrogation method described above. The artificial intelligence-based pre-consultation method comprises the steps of inputting the i-1 th round of consultation steps and the pre-consultation dialogue summarized data of the i round of consultation steps corresponding to a target doctor into a consultation step prediction model for predicting the i round of consultation steps, wherein the target doctor is a doctor which the target patient wants to visit, inputting the i round of consultation steps and the pre-consultation dialogue summarized data into a pre-consultation reply generation model corresponding to the target doctor for reply text generation to obtain a target reply text, and displaying the target reply text based on an virtual image corresponding to the target doctor. Thereby realizing automatic pre-co