阅读说明：本技术 一种医疗诊断辅助系统的工作方法 (A kind of working method of medical diagnosis auxiliary system ) 是由 高瞻 金博 佟晓梅 王雷 金嵩 于 2019-08-26 设计创作，主要内容包括：本发明提供一种医疗诊断辅助系统的工作方法,包括以下步骤：将症状信息输入服务器中,对患者检查数据采样；服务器识别症状信息中的关键词和相关句,提取相应的关键词；根据从历史病历数据获得的样本训练和统计得到该疾病的详细特征,找出潜在的病患,并刻画出疾病模型；将刻画出的疾病模型与数据库中的诊断模型进行匹配,得出与患者症状信息最接近的诊断模型结果。本发明通过数据不断积累进一步提高数据分析的准确性,相比现有技术的互联网问诊准确性提高70％,本发明根据数据的逻辑运算不断扩充新的分析词汇和逻辑算法,通过辅助诊断提高医生诊断能力,降低误诊、漏诊率。医生根据实际情况及本发明的查询结果,对症下药,可达到100％的准确率。(The present invention provides a kind of working method of medical diagnosis auxiliary system, comprising the following steps: inputs in server symptom information, checks data sampling to patient；Server identifies keyword and related sentence in symptom information, extracts corresponding keyword；The detailed features of the disease are obtained according to the sample training and statistics that obtain from history medical record data, find out potential sufferer, and depict disease model；The disease model depicted is matched with the diagnostic model in database, is obtained and the immediate diagnostic model result of patient symptom information.The present invention constantly accumulates the accuracy for further increasing data analysis by data, internet interrogation accuracy compared with prior art improves 70%, the present invention constantly expands new analysis vocabulary and logical algorithm according to the logical operation of data, diagnosis ability is improved by auxiliary diagnosis, reduces mistaken diagnosis, rate of missed diagnosis.Doctor according to the actual situation and query result of the invention, suits the remedy to the case, can reach 100% accuracy rate.)

1. a kind of working method of medical diagnosis auxiliary system, comprising the following steps:

S1, disease retrieval

S11, the symptom information in main suit is inputted in server, data sampling, the inspection data sampling packet is checked to patient It includes and extracts this physical examination situation of patient and patient history's inspection situation；

S12, server carry out natural language processing to the symptom information, identify keyword and related sentence in symptom information, And extract corresponding keyword；

S2, machine learning

S21, pathology is analyzed, finds the inspection data critical word that feature in database describes most similar case and patient It compares and analyzes, the detailed features of the disease is obtained according to the sample training and statistics that obtain from history medical record data, are found out Potential sufferer, while depicting disease model；

S22, the disease model depicted is matched with the diagnostic model in database, obtains and believes with symptom described in the patient Cease immediate diagnostic model result；

S3, auxiliary diagnosis result is shown

S31, the diagnostic model result that the present invention calculates is showed into doctor, patient data information is with sector diagram, Line Chart Equal various ways are counted, and give certain diagnosis reference proposition；

S32, to the sufferer gone to a doctor come to visit situation be ranked up and mark, auxiliary doctor pay a return visit and

Understand the conditions of sufferer, and handles in time.

2. a kind of working method of medical diagnosis auxiliary system according to claim 1, which is characterized in that the nature language Speech processing, as carrying out semantic feature to the symptom information indicates to excavate with semantic, extracts the main suit using neural network In internal characteristics and pathology, excavate the semantic feature in the symptom information.

3. a kind of working method of medical diagnosis auxiliary system according to claim 1, which is characterized in that the history disease It counts one by one according to cardinal symptom, the recent illness, pervious illness history, nearest physical examination report for including sufferer, the pervious disease Disease history includes past medical history, family history.

4. a kind of working method of medical diagnosis auxiliary system according to claim 1, which is characterized in that the database Including case case details, the history medical history record of all patients, the diagnoses and treatment record of patient, case, doctor comes personally guide, Processing scheme and method, the pertinent literatures such as disease, report etc..

5. a kind of working method of medical diagnosis auxiliary system according to claim 1, which is characterized in that further include sufferer The data such as main suit, present illness history, both medical history and personal information oneself are filled in, symptom information is inputted in server, according to corresponding number It is investigated that disease of bringing vexation on oneself, matches disease model, gives corresponding solution.

Technical field

The present invention relates to medical diagnosis technical fields, more specifically to a kind of work of medical diagnosis auxiliary system Method.

Background technique

The classification of present illness is more and more thinner, and Patient Global's state of an illness becomes increasingly complex, and the requirement to the diagnosis and treatment of doctor is also got over Come higher, in short supply instantly in medical resource, the quantity that doctor needs to diagnose patient daily is also more and more, and doctor is caused to hold very much Easily there is promnesia, chaotic situation generation, the case where inevitably will cause mistaken diagnosis, fail to pinpoint a disease in diagnosis.Therefore one kind can be directed to patient Multiple inspection data, provide big data analysis infer possible diagnostic result for doctor's reference, thus reduce fail to pinpoint a disease in diagnosis, mistaken diagnosis The method of generation becomes the task of top priority.

Summary of the invention

The present invention to solve the above-mentioned problems, proposes a kind of working method of medical diagnosis auxiliary system, solve due to Classification of diseases is too thin, too miscellaneous and doctors'work burden is excessive, cause doctor to be easy to fail to pinpoint a disease in diagnosis, mistaken diagnosis the problem of.

In order to achieve the above objectives, the present invention provides a kind of working method of medical diagnosis auxiliary system, comprising the following steps:

S1, disease retrieval.

S11, the symptom information in main suit is inputted in server, data sampling is checked to patient, the inspection data are adopted Sample includes extracting this physical examination situation of patient and patient history's inspection situation.

S12, server carry out natural language processing to the symptom information, identify keyword and correlation in symptom information Sentence, and extract corresponding keyword.

S2, machine learning.

S21, pathology is analyzed, finds feature in database and describes most similar case and the inspection data pass of patient Keyword compares and analyzes, and obtains the detailed features of the disease according to the sample training and statistics that obtain from history medical record data, Potential sufferer is found out, while depicting disease model.

S22, the disease model depicted is matched with the diagnostic model in database, is obtained and disease described in the patient The immediate diagnostic model result of shape information.

S3, auxiliary diagnosis result is shown.

S31, the diagnostic model result that the present invention calculates is showed into doctor, patient data information is with sector diagram, line The various ways such as shape figure are counted, and give certain diagnosis reference proposition.

S32, situation of coming to visit to the sufferer gone to a doctor are ranked up and mark, and auxiliary doctor pays a return visit and understand the disease of sufferer Disease situation, and handle in time.

Under preferred embodiment, the natural language processing, as carrying out semantic feature to the symptom information is indicated and semanteme It excavates, internal characteristics and pathology in the main suit is extracted using neural network, excavate the semantic feature in the symptom information.

Under preferred embodiment, the history medical record data includes that the cardinal symptom of sufferer, recent illness, pervious illness are gone through History, nearest physical examination report, the pervious illness history includes past medical history, family history.

Under preferred embodiment, the database includes case case details, and the history medical history record of all patients, patient's examines Disconnected treatment record, case, doctor come personally guide, processing scheme and method, the pertinent literatures such as disease, report etc..

Under preferred embodiment, the present invention can fill in the number such as main suit, present illness history, both medical history and personal information with sufferer oneself According to by symptom information input server, according to corresponding data lookup illness, matching disease model gives corresponding solution party Case.

The present invention is based on big data platforms, and the accuracy of data analysis can be further increased by the continuous accumulation of data, Internet interrogation accuracy compared with prior art improves 70%.The present invention constantly expands new point according to the logical operation of data Vocabulary and logical algorithm are analysed, diagnosis ability is improved by auxiliary diagnosis, reduces mistaken diagnosis, rate of missed diagnosis.Doctor is according to practical feelings Condition and query result of the invention, suit the remedy to the case, and can reach 100% accuracy rate.

Detailed description of the invention

Fig. 1 is specific steps flow diagram of the present invention.

Specific embodiment

The present invention is further explained and is illustrated with specific embodiment with reference to the accompanying drawings of the specification.For of the invention real The step number in example is applied, is arranged only for the purposes of illustrating explanation, any restriction is not done to the sequence between step, is implemented The execution sequence of each step in example can be adaptively adjusted according to the understanding of those skilled in the art.

As shown in Figure 1, a kind of working method of medical diagnosis auxiliary system, comprising the following steps:

1, disease is retrieved.

(1) state of an illness is described in doctor, and symptom information is inputted in server, checks data sampling to patient, described Check that data sampling includes extracting this physical examination situation of patient and patient history's inspection situation.

(2) server carries out natural language processing to the symptom information, identifies keyword and correlation in symptom information Sentence, and extract corresponding keyword.The natural language processing as carries out semantic feature expression and language to the symptom information Justice is excavated, and extracts internal characteristics and pathology in the main suit using neural network, and the semanteme excavated in the symptom information is special Sign.Specific method is: neuron calculates character representation automatically, the feature vector after exporting as learning of neuron, the nerve net Network includes convolutional neural networks, recurrent neural network etc., and the present invention is not limited to a certain specific networks.

2, machine learning.

(1) pathology is analyzed using big data, finds feature in database and describes most similar case and patient It checks that data critical word compares and analyzes, which is obtained according to the sample training and statistics that obtain from history medical record data Detailed features find out potential sufferer, while depicting disease model.The history medical record data includes the basic disease of sufferer Shape, recent illness, pervious illness history, nearest physical examination report, the pervious illness history includes past medical history, family History.

(2) disease model depicted is matched with the diagnostic model in database, is obtained and disease described in the patient The immediate diagnostic model result of shape information.The database includes case case details, the history medical history record of all patients, The diagnoses and treatment record of patient, case, doctor come personally guide, processing scheme and method, the pertinent literatures such as disease, report etc..

3, auxiliary diagnosis result is shown.

(1) the diagnostic model result calculated of the present invention is showed into doctor, patient data information is with sector diagram, linear The various ways such as figure are counted, and give certain diagnosis reference proposition.

(2) the visiting situation of the sufferer gone to a doctor is ranked up and is marked, auxiliary doctor pays a return visit and understand the disease of sufferer Disease situation, and handle in time.

The invention also includes sufferers oneself to fill in the data such as main suit, present illness history, both medical history and personal information, by symptom information It inputs in server, illness is searched according to corresponding data, disease model is matched, gives corresponding solution.

The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto, Anyone skilled in the art within the technical scope of the present disclosure, according to the technique and scheme of the present invention and its Inventive concept is subject to equivalent substitution or change, should be covered by the protection scope of the present invention.