阅读说明：本技术 一种医疗数据处理方法和系统 (Medical data processing method and system ) 是由 陈浩宇 于 2021-08-17 设计创作，主要内容包括：一种医疗数据处理方法和系统,其方法包括：通过预设的第一问答模式,获取患者的发病部位信息,并提取发病部位信息的关键字,通过预设的第二问答模式,获取患者的病症信息,并提取病症信息的关键字,根据发病部位信息的关键字和病症信息的关键字进行病例匹配,将匹配度最高的病例从病例库中取出,作为参考病例发送给医生,并自动为患者挂号；本申请通过获取患者的患病信息,自动获取病例库中的相似病例,以此作为参考病例发给医生,可节约大量的问诊时间,而且,医生通过参考病例可采取针对性治疗,使得患者治疗效率更高。(A medical data processing method and system, the method comprising: acquiring the information of the diseased part of the patient through a preset first question-answering mode, extracting keywords of the information of the diseased part, acquiring the disease information of the patient through a preset second question-answering mode, extracting the keywords of the disease information, performing case matching according to the keywords of the information of the diseased part and the keywords of the disease information, taking out the case with the highest matching degree from a case library, sending the case as a reference case to a doctor, and automatically registering the patient; according to the method and the system, the similar cases in the case base are automatically acquired by acquiring the illness information of the patient and are sent to the doctor as the reference case, so that a large amount of inquiry time can be saved, and the doctor can adopt targeted treatment through the reference case, so that the treatment efficiency of the patient is higher.)

1. A medical data processing method, comprising:

acquiring the information of the diseased part of the patient through a preset first question-answering mode, and extracting keywords of the information of the diseased part;

acquiring the disease information of the patient through a preset second question-answering mode, and extracting keywords of the disease information;

performing case matching according to the keywords of the disease part information and the keywords of the disease information;

and taking the case with the highest matching degree out of the case library, sending the case to a doctor as a reference case, and automatically registering the patient.

2. The medical data processing method according to claim 1, further comprising, after extracting the keyword of the diseased part information:

and acquiring the diseased part of the patient according to the keywords of the diseased part information.

3. The medical data processing method according to claim 2, wherein before acquiring the disease information of the patient through the preset second question-answering mode and extracting the keyword of the disease information, the method further comprises:

and selecting a corresponding second question-answering mode according to the diseased part of the patient.

4. The medical data processing method according to claim 1, wherein before the case matching based on the keyword of the disease site information and the keyword of the disease condition information, the method further comprises:

classifying the cases in the case base according to the disease parts, and extracting and storing the keywords of the disease symptoms as case labels.

5. The medical data processing method of claim 4, wherein the extracting the case with the highest matching degree from the case library comprises:

and matching the keywords of the disease information with the case labels, and taking out the case corresponding to the case label with the highest matching degree from the case library.

6. The medical data processing method according to claim 1, wherein the preset first question-answering mode and the preset second question-answering mode are question-answering rules made by a physician or an expert.

7. The medical data processing method according to claim 1, wherein the extracting of the keyword of the disease information includes:

converting the disease information into text information, cutting and dividing the text by taking sentences as units, cutting the text by using sentence separators, obtaining text vocabularies through a word segmentation device, marking the attributes of the vocabularies, removing the vocabularies with irrelevant semantics, and keeping the disease state vocabularies and the disease degree vocabularies.

8. A medical data processing system comprising:

the first question-answering module is used for acquiring the information of the morbidity part of the patient through a preset first question-answering mode and extracting keywords of the information of the morbidity part;

the second question-answer module is used for acquiring the disease information of the patient through a preset second question-answer mode and extracting keywords of the disease information;

the matching module is used for matching cases according to the keywords of the diseased parts and the keywords of the disease information;

and the case acquisition module is used for taking out the case with the highest matching degree from the case library, sending the case to a doctor as a reference case and automatically registering the patient.

9. A computer arrangement comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor, when executing the computer program, implements the medical data processing method according to any of claims 1-7.

10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the medical data processing method according to any one of claims 1-7.

Technical Field

The invention relates to the technical field of medical information processing, in particular to a medical data processing method and system.

Background

When a patient is in a visit, a long-time inquiry is usually needed, the efficiency is low, the current medical resources are very tight, the medical resources are wasted due to the long-time repeated inquiry, the same disease symptoms are mostly similar, if the symptoms of the patient can be obtained in advance, the closest disease information can be obtained according to the symptom characteristics and sent to a doctor, and meanwhile, the patient is automatically registered, so that a great amount of diagnosis time can be saved.

Therefore, how to provide a method for rapidly diagnosing diseases is a problem to be solved urgently by those skilled in the art.

Disclosure of Invention

The embodiment of the application provides a medical data processing method and system, and aims to solve the problem of overlong medical diagnosis time.

In a first aspect, the present application provides a medical data processing method, comprising:

acquiring the information of the diseased part of the patient through a preset first question-answering mode, and extracting keywords of the information of the diseased part;

acquiring the disease information of the patient through a preset second question-answering mode, and extracting keywords of the disease information;

performing case matching according to the keywords of the disease part information and the keywords of the disease information;

and taking the case with the highest matching degree out of the case library, sending the case to a doctor as a reference case, and automatically registering the patient.

In one embodiment, after extracting the keyword of the disease-onset part information, the method further includes:

and acquiring the diseased part of the patient according to the keywords of the diseased part information.

In one embodiment, before acquiring the disease information of the patient through a preset second question-answering mode and extracting keywords of the disease information, the method further includes:

and selecting a corresponding second question-answering mode according to the diseased part of the patient.

In one embodiment, before performing case matching based on the keyword of the disease site information and the keyword of the disease condition information, the method further includes:

classifying the cases in the case base according to the disease parts, and extracting and storing the keywords of the disease symptoms as case labels.

In one embodiment, the retrieving the case with the highest degree of matching from the case repository includes:

and matching the keywords of the disease information with the case labels, and taking out the case corresponding to the case label with the highest matching degree from the case library.

In one embodiment, the preset first question-answering mode and the preset second question-answering mode are question-answering rules made by a physician or a specialist.

In a second aspect, the present application also provides a medical data processing system comprising:

the first question-answering module is used for acquiring the information of the morbidity part of the patient through a preset first question-answering mode and extracting keywords of the information of the morbidity part;

the second question-answer module is used for acquiring the disease information of the patient through a preset second question-answer mode and extracting keywords of the disease information;

the matching module is used for matching cases according to the keywords of the diseased parts and the keywords of the disease information;

and the case acquisition module is used for taking out the case with the highest matching degree from the case library, sending the case to a doctor as a reference case and automatically registering the patient.

In a third aspect, the present application further provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the medical data processing method according to any one of the above first aspects.

In a fourth aspect, the present application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is adapted to perform the method of processing medical data according to any of the first aspect when executed by a processor.

According to the medical data processing method and system, the information of the diseased part of the patient is obtained through the preset first question-answer mode, the keywords of the information of the diseased part are extracted, the disease information of the patient is obtained through the preset second question-answer mode, the keywords of the disease information are extracted, case matching is carried out according to the keywords of the information of the diseased part and the keywords of the disease information, the case with the highest matching degree is taken out of the case library and sent to a doctor as a reference case, and the patient is automatically registered.

Drawings

For better clarity of the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a flowchart of a medical data processing method according to an embodiment of the present application.

Detailed Description

The following further describes embodiments of the present invention with reference to the drawings. It should be noted that the description of the embodiments is provided to help understanding of the present invention, but the present invention is not limited thereto. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

Referring to fig. 1, a flowchart of a medical data processing method is shown in an embodiment, and includes:

s101, acquiring the information of the diseased part of the patient through a preset first question answering mode, and extracting keywords of the information of the diseased part.

In an embodiment, after acquiring the information of the disease part of the patient through a preset first question-answering mode and extracting keywords of the information of the disease part, the method further includes:

and acquiring the diseased part of the patient according to the keywords of the diseased part information.

The first question-answering mode may be a question-answering rule set according to a doctor or an expert according to a question-answering experience.

S102, acquiring disease information of the patient through a preset second question-answer mode, and extracting keywords of the disease information.

In an embodiment, before acquiring the disease information of the patient through a preset second question-answering mode and extracting a keyword of the disease information, the method further includes:

and selecting a corresponding second question-answering mode according to the diseased part of the patient.

After the diseased part of the patient is obtained, the corresponding disease condition question-answer mode and the corresponding keyword extraction model can be selected according to the diseased part to obtain the corresponding symptom keywords, and each question-answer mode corresponds to one keyword extraction model, so that the accuracy of symptom information extraction is ensured.

Because the disease symptoms of different disease conditions are different, in order to obtain the disease condition of the patient in a targeted manner, it is necessary to determine the disease site of the patient first, and select a corresponding second question-and-answer mode according to the disease site, where the second question-and-answer mode may be one of preset question-and-answer modes, where the preset second question-and-answer mode may be a question-and-answer rule set according to a doctor or an expert according to a question-and-answer experience, for example:

if the disease part is the head, a head disease question-answer mode is automatically selected, and the specific questions comprise: disease condition, disease degree and disease frequency.

The disease condition of the patient is inquired through a preset question-answer mode, the disease type of the patient can be quickly determined, and the waste of medical resources and money caused by selecting a wrong department when the user registers can be avoided.

And S103, matching cases according to the keywords of the disease part information and the keywords of the disease information.

And classifying cases according to the acquired keywords of the information of the disease part, determining the disease part of the patient, and then matching the cases in the case library according to the symptom keywords.

In one embodiment, before case matching according to the condition keywords, the method further comprises:

cases in the existing case base are classified according to the disease site.

After classifying the cases in the existing case base according to the disease parts, the method further comprises the following steps:

and acquiring keywords of different cases, and storing the keywords as labels of the cases.

When the case matching is carried out according to the keywords of the disease part information and the keywords of the disease information, the matching result can be obtained only by inquiring the label information of the case, a large amount of time and resources are saved, and the case with the highest matching degree can be determined by matching the keywords with the most corresponding case.

In one embodiment, the keywords of the condition information include a condition status keyword and a condition degree keyword.

Wherein, extracting keywords in the disease information comprises:

a. converting the disease information into text information, taking sentences as units, and cutting the text once by using sentence separators, namely T ═ V₁，V₂，V₃，…，V_n]Where T is a set of text information, V₁-V_nThe sentence is obtained by cutting the text information set T once.

b. For each sentence V_iE.g. T, performing primary word segmentation and one-time part-of-speech tagging on the words, removing words and repeated words which are irrelevant to semantics and are subjected to the primary word segmentation and one-time part-of-speech tagging, and keeping disease state words, namely V_i＝[v_i1，v_i2，…,_in]Wherein v is_ijAs a sentence V_iT is a text information set, V_iIs the ith sentence in the text information set T.

In one embodiment, the removing repeated vocabulary after the primary word segmentation and the primary part-of-speech tagging comprises:

(1) and performing semantic division on the existing disease state words to obtain a word semantic set.

If the diseased part of the patient is the head, the words related to the disease state generally include "dizziness", "headache", etc., and "headache" or "dizziness" can be classified into these two forms although there are various expression modes, and the state between the two forms, i.e., slight headache or dizziness, or headache accompanied by dizziness, and the patient sometimes cannot confirm which symptom is, therefore, the disease state can be classified into "headache state", "dizziness state" and "intermediate state".

(2) Judging the similarity of each vocabulary in the sentence with other vocabularies, wherein the similarity calculation formula is as follows:

v_imfor the mth vocabulary in the sentence, v_inFor the nth word in the sentence, P is v_imAnd v_inDegree of association of (1), S_dIs v is_imAnd v_inDistance in lexical semantic collections, S_cTo adjust the parameters, S_minTo determine the minimum similarity of similarity between two words.

(3) And comparing the obtained similarity with a similarity threshold, if the similarity is smaller than the similarity threshold, indicating that the two vocabularies are not similar, and if the similarity is larger than the similarity threshold, indicating that the two vocabularies are similar, and rejecting one vocabulary.

The method comprises the steps of carrying out semantic division on the existing disease state words to obtain word semantic sets, judging the similarity of each word and other words in a sentence by utilizing the distance between each word and other words in the word semantic sets and the minimum similarity of the two words, and accordingly rejecting the words with high similarity_min1, thereby calculating the obtained Sim (v)_im，v_in) 0, reducing computational complexity.

c. Will be provided withConverting the disease information into text information, using sentence as unit, and utilizing sentence separator to make secondary cutting on the text, i.e. T ═ S₁，S₂，S₃，...，S_n]Where T is a set of text information, S₁-S_nThe sentence is obtained after the text information set T is cut for the second time.

d. For each sentence S_iE.g. T, performing secondary word segmentation and secondary part-of-speech tagging on the words, removing words and repeated words which are irrelevant to semantics and are subjected to the secondary word segmentation and secondary part-of-speech tagging, and keeping disease degree words, namely S_i＝[t_i1，t_i2，...，t_in]Wherein, t_ijAs a sentence S_iThe key word of the degree of disease, S_iIs the ith sentence in the text information set.

Because the acquired disease information is the same, the method and principle of performing primary cutting and secondary cutting on the text information are the same, but because the keywords corresponding to the disease state and the disease degree are different, the ways of performing word segmentation and part-of-speech tagging on sentences are different, for example: when the disease state is expressed, the words are complex, so that long bytes can be adopted for word segmentation once, and the part-of-speech tagging can be emphasized on the words related to time; when the degree of disease is expressed, the words are often short, and the words can be divided by using short bytes, and secondary part-of-speech tagging can be performed on the words related to the degree.

In an embodiment, acquiring keywords of different cases and storing the keywords as tags of the cases further includes:

the labels of the cases are sequentially stored according to the degree of the disorder.

In order to ensure the identification accuracy of the disease degree, a disease degree synonym library can be established, a plurality of synonyms for describing the same disease degree are used in a fixed description mode to deal with different expression of the disease degree of a patient, wherein words in the disease degree synonym library are consistent with words of a case label and are stored according to the sequence of the disease degree, and irrelevant words are words with different or different semantemes from words in the synonym library.

For example: the patient's degree of headache, a common indication of the degree of the condition, is generally: "severe," "intolerable," etc., may be used to indicate "severe," thereby reducing the amount of work required to classify or identify a case.

In one embodiment, the removing of the semanteme-independent vocabulary and the repeated vocabulary after the secondary word segmentation and the secondary part-of-speech tagging processing and the keeping of the disease degree vocabulary therein comprises:

(1) dividing each sentence into a plurality of vocabularies according to preset sentence separators, and judging the association degree of each vocabulary in the sentence and the vocabulary in the synonym library of the disease degree, wherein the association degree is calculated by adopting the following formula:

wherein, T (a)_i) Representing the degree of association of the ith word in the sentence, W_iRepresenting the degree of semantic similarity of the words in the ith word and case degree synonym library, W_ijThe position of the approximate vocabulary expressing the ith vocabulary in the synonym library is j, n is the number of synonyms of the vocabulary which is semantically similar to the ith vocabulary, and alpha is an adjusting parameter.

(2) Will T (a)_i) And comparing the vocabulary with a set relevance threshold, if the relevance is smaller than a relevance first threshold, the vocabulary is represented as a vocabulary without meaning, and then rejecting the vocabulary, and if the relevance is larger than a relevance second threshold, the vocabulary is represented as a repeated vocabulary, and the vocabulary is also rejected, and the vocabulary with the relevance between the relevance first threshold and the relevance second threshold is reserved.

The method compares the relevance degree of each vocabulary with the vocabulary of the synonym library, considers that the expression of a patient is unclear when describing the disease degree, judges the disease degree of the patient by establishing the synonym library in advance, since the words in the thesaurus are ordered according to the degree of disorder, the position of the disorder degree words in the thesaurus, and the number of synonyms, which also represent the severity of the degree of the condition, are fully contemplated by the present application, and then calculate the association degree between each vocabulary in the sentence and the vocabulary in the synonym library of the symptom degree, thereby reducing the probability of misjudgment of the state of illness caused by unclear expression of the patient in describing the state of illness, and in addition, the vocabulary with irrelevant semantics is removed through the vocabulary relevancy, the complexity of subsequent vocabulary screening is reduced, and the workload is reduced.

e. Will V_iAnd S_iAnd matching the disease state keywords and the disease degree keywords in the database in sequence to obtain a disease state and degree set, and then matching the disease state and degree set with existing case labels in a case library.

The traditional keyword screening technology is to screen out the words with the highest frequency as keywords, which is obviously not applicable to the scheme of the application, and if the disease state and the disease degree are screened at the same time, the condition of an illness may be judged unreasonably, for example: repeated description of the disease state or degree by the patient can cause the phenomenon of word confusion when the extracted keywords are recombined, so that the disease state keywords and the disease degree keywords are respectively extracted through converting the voice signals of the patient into text information and carrying out secondary cutting, and the disease state information and the degree information of the patient are obtained through recombination matching, so that the error of disease judgment can be reduced.

And S104, taking the case with the highest matching degree out of the case library, sending the case to a doctor as a reference case, and automatically registering the patient.

The case that will correspond with the label that patient's disease keyword matching degree is the highest is extracted from the case storehouse, send the doctor that corresponds the department as the reference case, and register for the patient automatically, the doctor diagnoses the patient according to referring to the case, can save a large amount of time, the doctor can also be according to the medicine and the treatment condition of referring to the case, the pertinence treats the patient, and, after sending the case for the doctor, register for the patient automatically, can avoid the patient because judge wrong state of an illness and hang the wrong number, cause the waste of money and time.

The embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. It will be apparent to those skilled in the art that various changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, and the scope of protection is still within the scope of the invention.