CN110610766A - Apparatus and storage medium for deriving probability of disease based on symptom feature weight - Google Patents
Apparatus and storage medium for deriving probability of disease based on symptom feature weight Download PDFInfo
- Publication number
- CN110610766A CN110610766A CN201910842570.8A CN201910842570A CN110610766A CN 110610766 A CN110610766 A CN 110610766A CN 201910842570 A CN201910842570 A CN 201910842570A CN 110610766 A CN110610766 A CN 110610766A
- Authority
- CN
- China
- Prior art keywords
- disease
- symptom
- filtering
- information
- result
- 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
- 201000010099 disease Diseases 0.000 title claims abstract description 171
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 171
- 208000024891 symptom Diseases 0.000 title claims abstract description 163
- 238000001914 filtration Methods 0.000 claims abstract description 69
- 238000005065 mining Methods 0.000 claims abstract description 7
- 238000000034 method Methods 0.000 claims description 19
- 238000003745 diagnosis Methods 0.000 description 12
- 238000013473 artificial intelligence Methods 0.000 description 9
- 238000009795 derivation Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 206010028980 Neoplasm Diseases 0.000 description 2
- ATUOYWHBWRKTHZ-UHFFFAOYSA-N Propane Chemical compound CCC ATUOYWHBWRKTHZ-UHFFFAOYSA-N 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 241000234282 Allium Species 0.000 description 1
- 235000002732 Allium cepa var. cepa Nutrition 0.000 description 1
- 206010006187 Breast cancer Diseases 0.000 description 1
- 208000026310 Breast neoplasm Diseases 0.000 description 1
- 206010008342 Cervix carcinoma Diseases 0.000 description 1
- 206010009944 Colon cancer Diseases 0.000 description 1
- 208000001333 Colorectal Neoplasms Diseases 0.000 description 1
- 208000000461 Esophageal Neoplasms Diseases 0.000 description 1
- 206010060891 General symptom Diseases 0.000 description 1
- 206010058467 Lung neoplasm malignant Diseases 0.000 description 1
- 206010030155 Oesophageal carcinoma Diseases 0.000 description 1
- 208000006105 Uterine Cervical Neoplasms Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 201000010881 cervical cancer Diseases 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 201000004101 esophageal cancer Diseases 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 201000005202 lung cancer Diseases 0.000 description 1
- 208000020816 lung neoplasm Diseases 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000001294 propane Substances 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000007306 turnover Effects 0.000 description 1
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
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- 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
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Abstract
The invention provides a device and a storage medium for deducing disease probability based on symptom characteristic weight, comprising: a symptom characteristic storage unit for storing a plurality of pieces of medical knowledge information, each piece of medical knowledge information consisting of a symptom characteristic and a disease label; a filtering strategy generating unit, which is used for mining the medical knowledge information to generate a filtering strategy; the retrieval unit is used for receiving the entity label from the consultation information or the feedback information, retrieving the symptom characteristic storage unit by using the entity label, and obtaining the symptom characteristic and the disease label corresponding to the consultation information or the feedback information as a retrieval result; and the filtering unit is used for filtering the retrieval result according to the filtering strategy to obtain a filtering result, judging the filtering result through a preset rule, and selecting to generate a problem according to the filtering result to prompt a user to input feedback information or to output the filtering result as a final result.
Description
Technical Field
The present invention relates to the fields of natural language processing, artificial intelligence, and big data analysis, and more particularly, to an apparatus and a storage medium for deriving a probability of a disease based on symptom feature weights.
Background
In the 70 s of the 20 th century, related research of a medical expert system for assisting a doctor in diagnosis by artificial intelligence begins to appear, various medical expert systems appear for over forty years, but no product is widely applied clinically, and even IBM Watson is only used for demonstration application in disease diagnosis such as tumor and cancer; the scientific research and development flying intelligent medical assistant robot becomes the first artificial intelligent robot passing through the qualification test of the national medical practitioners in the country because the robot passes through the comprehensive written examination evaluation of the qualification test of the national medical practitioners by the score of more than 96 points of the qualified line, and is only applied to some examination points at present. Therefore, the research of artificial intelligent auxiliary diagnosis and other comprehensive applications starts early, but the artificial intelligent auxiliary diagnosis and other comprehensive applications are not popularized and applied to the present. The reason for this is that the last kilometer of the technology and medical scene is not communicated, that is, the condition collection has defects, and the medical expert system is far beyond the medical knowledge level of the ordinary doctor from the viewpoint of technical research, and the doctor-patient conversation itself collects the condition of the patient according to the diagnosis idea of the doctor in the medical diagnosis process. The last kilometer cannot exceed the distance, and powerful artificial intelligence is also useless for doctors.
To get through this last kilometer, "a method of collecting a disease condition based on associated symptom derivation" has been proposed. The method is a good idea, namely, the method is based on the principle of diagnostics, and is like a doctor to quickly, effectively and accurately collect the illness state according to the patient chief complaints, and the subjective diagnosis process of the doctor is simulated by using technical means. The disease condition collection method based on the derivation of the associated symptoms is actually a keyword retrieval technology; the device for deducing the disease probability based on the symptom characteristic weight is an artificial intelligence technology. It is believed that with the interest in this patent achievement in the industry, the last mile of hurdle will be cleared for the widespread use of artificial intelligence-aided diagnosis. Besides the aforesaid comprehensive application of artificial intelligence aided diagnosis, the application research of artificial intelligence aided doctors in some key points of diagnosis and treatment is continuously making breakthrough, especially in medical imaging, such as the accuracy of medical image screening and identification of esophageal cancer, lung cancer, glycoreticular lesion, breast cancer, colorectal cancer and cervical cancer is over the medical expert level. With the breakthrough of more key points, the intelligence degree of the artificial intelligence auxiliary diagnosis comprehensive application is higher and higher.
Disclosure of Invention
In view of the deficiencies of the prior art, the present invention provides an apparatus for deriving probability of disease based on symptom feature weights, comprising:
a symptom characteristic storage unit for storing a plurality of pieces of medical knowledge information, each piece of medical knowledge information consisting of a symptom characteristic and a disease label;
a filtering strategy generating unit, which is used for mining the medical knowledge information to generate a filtering strategy;
the retrieval unit is used for receiving the entity label from the consultation information or the feedback information, retrieving the symptom characteristic storage unit by using the entity label, and obtaining the symptom characteristic and the disease label corresponding to the consultation information or the feedback information as a retrieval result;
and the filtering unit is used for filtering the retrieval result according to the filtering strategy to obtain a filtering result, judging the filtering result through a preset rule, and selecting to generate a problem according to the filtering result to prompt a user to input feedback information or to output the filtering result as a final result.
The device for deducing the disease probability based on the symptom characteristic weight is characterized in that the symptom characteristic storage unit is used for compressing and storing the medical knowledge information in a mode of combining common symptom characteristics among disease labels.
The apparatus for deriving probability of disease based on symptom feature weights, wherein the compressed storage comprises:
the medical knowledge information is divided according to disease types, common symptom characteristics of each type of disease are extracted, the common symptom characteristics and the corresponding disease types are stored in the symptom characteristic storage unit, and the medical knowledge information consisting of unique symptom characteristics, the disease types and the disease labels is also stored in the symptom characteristic storage unit.
The device for deducing the disease probability based on the symptom characteristic weight is characterized in that the common symptom characteristic is endowed with a low weight value, and the unique symptom characteristic is endowed with a high weight value.
The apparatus for deriving probability of disease based on symptom feature weight, wherein the retrieving unit comprises: and obtaining the disease weight of each disease label according to the weight value of the symptom characteristic corresponding to the consultation information or the feedback information, so as to count the probability of suffering from each disease, and extracting one or more disease labels with the highest probability and the symptom characteristic corresponding to the disease labels as the retrieval result.
The present invention also proposes a storage medium for storing a program for executing the method of:
a symptom characteristic storage step of storing a plurality of pieces of medical knowledge information, each piece of medical knowledge information consisting of a symptom characteristic and a disease label;
a filtering strategy generating step of mining the medical knowledge information to generate a filtering strategy;
a retrieval step, receiving an entity label from the consultation information or the feedback information, and retrieving the symptom characteristic storage unit by using the entity label to obtain a symptom characteristic and a disease label corresponding to the consultation information or the feedback information as a retrieval result;
and a filtering step, namely filtering the retrieval result according to the filtering strategy to obtain a filtering result, judging the filtering result through a preset rule, and selecting to generate a problem according to the filtering result to prompt a user to input feedback information or selecting to output the filtering result as a final result.
The storage medium, wherein the medical knowledge information is compressed and stored in a manner of incorporating common symptom characteristics among disease signatures.
The storage medium, wherein the compressed storage comprises:
the medical knowledge information is divided according to disease types, common symptom characteristics of each type of disease are extracted, the common symptom characteristics and the corresponding disease types are stored in the symptom characteristic storage unit, and the medical knowledge information consisting of unique symptom characteristics, the disease types and the disease labels is also stored in the symptom characteristic storage unit.
The storage medium, wherein the common symptom feature is assigned a low weight value, and the unique symptom feature is assigned a high weight value.
The storage medium, wherein the retrieving step comprises: and obtaining the disease weight of each disease label according to the weight value of the symptom characteristic corresponding to the consultation information or the feedback information, so as to count the probability of suffering from each disease, and extracting one or more disease labels with the highest probability and the symptom characteristic corresponding to the disease labels as the retrieval result.
The invention also provides an acquisition method for deducing the disease probability based on the symptom characteristic weight. The acquisition method is characterized by comprising the following steps: s01: constructing a database of symptom characteristics and association relations thereof according to the association relations among different symptoms of diseases in medical classics; s02: since different symptom characteristics contribute differently to different types of disease promotion and prediction, accordingly, different weights are assigned; the role played by the clinics for identifying a type of disease group or a specific disease is different, and therefore, the weight information is embodied; s03: according to information such as symptom characteristics of diseases, a mode that symptom characteristic weights are weighted and modeled on different disease type symptom characteristics is adopted, and one type or a certain specific disease can be preliminarily deduced based on symptom characteristic association relation data and symptom use frequency data, so that the disease condition data acquisition efficiency of the system is greatly improved.
Drawings
FIG. 1 is a graph of the relationship of the time dimension and the space dimension;
FIG. 2 is a flow chart of the present invention;
FIG. 3 is a flow chart of the present invention for managing and maintaining.
Detailed Description
An acquisition method for deducing disease probability based on symptom characteristic weight. Based on the symptom characteristics and the association relation data, according to the information of the sex, the age, the selected disease type and the like of the patient, the system automatically searches out a set of the symptom characteristics conforming to the disease type, and because the weight marking is carried out on the main set of the symptom characteristics of the disease of the type, the probability closer to the specific disease of the type can be deduced by associating the symptom characteristics according to the selection frequency and the times of the symptoms of the patient. Meanwhile, the probability result obtained by the patient through associated symptom derivation can often more accurately guess other symptoms of the patient, so that the disease condition acquisition efficiency of the system can be greatly improved.
Symptoms are characterized by an abnormal state in which the patient is presented with the disease. Each disease has its specific etiology and pathology, and in particular has a certain evolution of development, and presents different symptomatic characteristics. That is, the symptoms are very different for each disease. An acquisition method for deducing disease probability based on the characteristic weight of clinical symptoms. Firstly, only common symptom characteristics of diseases or a certain class of diseases are presented to a user as common labels of a first classification, and the user can select a plurality of labels; determining a first-level classification through the label weight; and then presenting the common label of the secondary classification to a user, wherein the user can select a plurality of labels and determine the secondary classification. And so on. And finally, presenting the specific disease label to a user, and finally deducing the disease symptom characteristics with the highest first three probabilities through the selection of a plurality of labels by the user. And repeating the steps to obtain all characteristic data of the diseased symptoms of the patient.
In order to make the aforementioned features and effects of the present invention more comprehensible, embodiments accompanied with figures are described in detail below.
The invention provides a method for collecting disease symptom characteristic data based on artificial intelligence semantic calculation (symptom characteristics are calculated as a weight). The weights are essentially keywords of the disease symptoms, i.e. keywords describing the disease symptoms, and can therefore be referred to as semantics. In addition, the weight of each keyword in different types of diseases is different, and the keyword can be set as a corresponding data symbol after being quantized, so that the keyword becomes a virtual keyword.
The invention mainly uses natural language processing, artificial intelligence and big data analysis method according to the symptom characteristic of disease, carries out natural language and semantic processing and calculation to the disease and the symptom characteristic of specific disease under a certain class, adopts general symptom characteristic collection flow, constructs a semantic algorithm and model, and carries out intelligent probability derivation to corresponding disease through self-complaint related symptom characteristic, thereby accurately guessing other symptoms of patient, and improving the disease collection efficiency of system. Wherein intelligence is embodied in that the machine can give basic probabilities as long as simple keywords are selected. That is, the machine may assist in the acquisition.
Doctors see a set of scientific and relatively fixed medical logic, which is a judgment and reasoning process, like peeling onions, peeling along the disease layer, and finally, the rest is determined to be the conclusion, and then laboratory examination is performed to confirm diagnosis.
The system of the invention comprises: a disease structured symptom signature storage unit configured to store structured medical knowledge information having a number of entity labels (e.g., as shown in table 1 below); a policy generation unit configured to mine the structured medical knowledge information to generate a filtering policy; the retrieval unit is configured to receive the medical consultation information or the feedback information and retrieve the structural symptom characteristics and the medical knowledge information of the entity labels corresponding to the medical consultation information or the feedback information; and the filtering unit is configured to filter the retrieval result according to the filtering strategy, judge the filtering result according to a preset rule, and select to generate and output a question according to the filtering result so as to prompt a user to input feedback information or generate and output result information according to the filtering result.
Table 1:
according to the acquisition method and the acquisition mechanism for deducing the disease probability based on the symptom characteristic weight, provided by the invention, the data acquisition dynamics, the acquisition path and the result generation of each time are proved to be consistent, and meanwhile, the three diseases with the maximum probability are deduced through the calculation of the symptom characteristics, so that the accuracy and the depth of the acquisition method and the result are guaranteed.
1. Disease and symptom characteristic system: in general, a disease or a class of diseases may be considered a class of diseases that has symptoms characteristic of being common in nature; the attributes under the ontology can have classification characteristics of first-level, second-level, even third-level and fourth-level; there may be common labels (symptom signatures) between the primary classification and the primary classification; there may be common labels (symptom features) between the secondary classifications; there may be common labels (symptom signatures) between the third and third classes; there may also be common labels (symptom signatures) between the individual diseases under the tertiary classification. The above expression "may be present" means that the common label (symptom characteristic) may or may not be present;
2. acquisition of symptom characteristics (i.e. label): in the early stage, construction is required to be completed through prior expert knowledge before system implementation;
3. establishment of symptom feature (i.e. label) weights: the label (symptom signature) is different in the role of identifying a disease group or a specific disease from the clinic, and is thus embodied as weight information. Different symptom characteristics contribute differently to different disease diagnoses, and accordingly, different weights are given according to the leading experience, so that one type or a specific disease probability can be preliminarily deduced, and meanwhile, data collection of disease symptom characteristics can be completed. Such as: the weighting tables for the three diseases "A", "B" and "C" are shown in Table 1 below.
Table 1:
4. designing a temporal logic and a spatial logic: in general, a particular disease under a class of disease systems may be as many as hundreds, and the symptom signature (symptom signature) of a particular disease may be as many as tens. This means that if the symptom labels (symptom features) of all specific diseases in a certain disease system are provided to the user at one time, and it is assumed that the mobile phone screen can display 20 labels, the user needs to turn over 50 times to get a full view of the labels. Obviously, this is a very poor user experience. To do so "the user does not have to slide the screen too much, i.e. can select the symptom tab" requires the introduction of a time dimension. I.e. first only the common labels of the first category are presented to the user, who can select a plurality of labels. The primary classification is determined by the label weight. And then presenting the common label of the secondary classification to a user, wherein the user can select a plurality of labels and determine the secondary classification. And so on. And finally, presenting the specific disease label to the user, and finally determining the first three diseases with the highest probability by the user through selection of a plurality of labels.
5. Classification or disease probability design: the first step is as follows: calculating the weight of each disease involved according to the symptom selected by the user; the second step is that: and calculating the disease probability according to the weight of each disease. Examples are: suppose that: the user selects the common label 1, the A label 1 and the C label 5, and the steps of calculating the disease probability are as follows: first, the individual disease weights are calculated. The weight of the disease A is 1+ 5-6; the weight of the disease B is 1; disease propane weight 1+50 51; the total weight is: 58 +1+ 50; the second step is that: and calculating the disease probability according to the weight of each disease. Probability of disease A: 6/58 ═ 10.3%; the probability of the disease B: 1/58 ═ 1.7%; the disease probability is: 51/58 is 87.9%.
6. The present invention addresses the problem of reasoning with different complexity, and considering that the keyword weights of different disease symptoms are different, algorithms and models will also vary, and therefore should be performed in models with different degrees of precision, i.e., the development and collection of a disease will be better in a certain class of diseases. At the same time, it is proposed to integrate organically the system database used for monitoring, storing and displaying large amounts of data with reasoning at the time of acquisition.
The relationship between the time dimension and the space dimension is shown in fig. 1.
Processing method when there is no co-classification:
if the classification is performed in the time dimension and some classifications do not have a common label, the next class common label should be presented to the user so as not to miss the classification. And so on until a label of a particular disease is presented. The logic is shown in fig. 2.
Managing the background function:
the management background provides a function of maintaining a certain data table, and the basic logic is shown in fig. 3.
The following are system examples corresponding to the above method examples, and this embodiment can be implemented in cooperation with the above embodiments. The related technical details mentioned in the above embodiments are still valid in this embodiment, and are not described herein again in order to reduce repetition. Accordingly, the related-art details mentioned in the present embodiment can also be applied to the above-described embodiments.
The invention also provides a device for deducing the probability of diseases based on symptom characteristic weight, which comprises the following steps:
a symptom characteristic storage unit for storing a plurality of pieces of medical knowledge information, each piece of medical knowledge information consisting of a symptom characteristic and a disease label;
a filtering strategy generating unit, which is used for mining the medical knowledge information to generate a filtering strategy;
the retrieval unit is used for receiving the entity label from the consultation information or the feedback information, retrieving the symptom characteristic storage unit by using the entity label, and obtaining the symptom characteristic and the disease label corresponding to the consultation information or the feedback information as a retrieval result;
and the filtering unit is used for filtering the retrieval result according to the filtering strategy to obtain a filtering result, judging the filtering result through a preset rule, and selecting to generate a problem according to the filtering result to prompt a user to input feedback information or to output the filtering result as a final result.
The device for deducing the disease probability based on the symptom characteristic weight is characterized in that the symptom characteristic storage unit is used for compressing and storing the medical knowledge information in a mode of combining common symptom characteristics among disease labels.
The apparatus for deriving probability of disease based on symptom feature weights, wherein the compressed storage comprises:
the medical knowledge information is divided according to disease types, common symptom characteristics of each type of disease are extracted, the common symptom characteristics and the corresponding disease types are stored in the symptom characteristic storage unit, and the medical knowledge information consisting of unique symptom characteristics, the disease types and the disease labels is also stored in the symptom characteristic storage unit.
The device for deducing the disease probability based on the symptom characteristic weight is characterized in that the common symptom characteristic is endowed with a low weight value, and the unique symptom characteristic is endowed with a high weight value.
The apparatus for deriving probability of disease based on symptom feature weight, wherein the retrieving unit comprises: and obtaining the disease weight of each disease label according to the weight value of the symptom characteristic corresponding to the consultation information or the feedback information, so as to count the probability of suffering from each disease, and extracting one or more disease labels with the highest probability and the symptom characteristic corresponding to the disease labels as the retrieval result.
The present invention also proposes a storage medium for storing a program for executing the method of:
a symptom characteristic storage step of storing a plurality of pieces of medical knowledge information, each piece of medical knowledge information consisting of a symptom characteristic and a disease label;
a filtering strategy generating step of mining the medical knowledge information to generate a filtering strategy;
a retrieval step, receiving an entity label from the consultation information or the feedback information, and retrieving the symptom characteristic storage unit by using the entity label to obtain a symptom characteristic and a disease label corresponding to the consultation information or the feedback information as a retrieval result;
and a filtering step, namely filtering the retrieval result according to the filtering strategy to obtain a filtering result, judging the filtering result through a preset rule, and selecting to generate a problem according to the filtering result to prompt a user to input feedback information or selecting to output the filtering result as a final result.
The storage medium, wherein the medical knowledge information is compressed and stored in a manner of incorporating common symptom characteristics among disease signatures.
The storage medium, wherein the compressed storage comprises: the medical knowledge information is divided according to disease types, common symptom characteristics of each type of disease are extracted, the common symptom characteristics and the corresponding disease types are stored in the symptom characteristic storage unit, and the medical knowledge information consisting of unique symptom characteristics, the disease types and the disease labels is also stored in the symptom characteristic storage unit. The storage medium, wherein the common symptom feature is assigned a low weight value, and the unique symptom feature is assigned a high weight value. The storage medium, wherein the retrieving step comprises: and obtaining the disease weight of each disease label according to the weight value of the symptom characteristic corresponding to the consultation information or the feedback information, so as to count the probability of suffering from each disease, and extracting one or more disease labels with the highest probability and the symptom characteristic corresponding to the disease labels as the retrieval result.
Claims (10)
1. An apparatus for deriving a probability of disease based on symptom feature weights, comprising:
a symptom characteristic storage unit for storing a plurality of pieces of medical knowledge information, each piece of medical knowledge information consisting of a symptom characteristic and a disease label;
a filtering strategy generating unit, which is used for mining the medical knowledge information to generate a filtering strategy;
the retrieval unit is used for receiving the entity label from the consultation information or the feedback information, retrieving the symptom characteristic storage unit by using the entity label, and obtaining the symptom characteristic and the disease label corresponding to the consultation information or the feedback information as a retrieval result;
and the filtering unit is used for filtering the retrieval result according to the filtering strategy to obtain a filtering result, judging the filtering result through a preset rule, and selecting to generate a problem according to the filtering result to prompt a user to input feedback information or to output the filtering result as a final result.
2. The apparatus according to claim 1, wherein the symptom characteristic storage unit stores the medical knowledge information in a compressed manner by combining common symptom characteristics among disease labels.
3. The apparatus of claim 2, wherein the compressed storage comprises:
the medical knowledge information is divided according to disease types, common symptom characteristics of each type of disease are extracted, the common symptom characteristics and the corresponding disease types are stored in the symptom characteristic storage unit, and the medical knowledge information consisting of unique symptom characteristics, the disease types and the disease labels is also stored in the symptom characteristic storage unit.
4. The apparatus of claim 3, wherein the common symptom feature is assigned a low weight value, and the unique symptom feature is assigned a high weight value.
5. The apparatus according to claim 1 or 4, wherein the search unit comprises: and obtaining the disease weight of each disease label according to the weight value of the symptom characteristic corresponding to the consultation information or the feedback information, so as to count the probability of suffering from each disease, and extracting one or more disease labels with the highest probability and the symptom characteristic corresponding to the disease labels as the retrieval result.
6. A storage medium characterized by storing a program for executing a method of:
a symptom characteristic storage step of storing a plurality of pieces of medical knowledge information, each piece of medical knowledge information consisting of a symptom characteristic and a disease label;
a filtering strategy generating step of mining the medical knowledge information to generate a filtering strategy;
a retrieval step, receiving an entity label from the consultation information or the feedback information, and retrieving the symptom characteristic storage unit by using the entity label to obtain a symptom characteristic and a disease label corresponding to the consultation information or the feedback information as a retrieval result;
and a filtering step, namely filtering the retrieval result according to the filtering strategy to obtain a filtering result, judging the filtering result through a preset rule, and selecting to generate a problem according to the filtering result to prompt a user to input feedback information or selecting to output the filtering result as a final result.
7. The storage medium of claim 6, wherein the medical knowledge information is compressed for storage in a manner that incorporates common symptom characteristics among disease signatures.
8. The storage medium of claim 7, wherein the compressed storage comprises:
the medical knowledge information is divided according to disease types, common symptom characteristics of each type of disease are extracted, the common symptom characteristics and the corresponding disease types are stored in the symptom characteristic storage unit, and the medical knowledge information consisting of unique symptom characteristics, the disease types and the disease labels is also stored in the symptom characteristic storage unit.
9. The storage medium of claim 8, wherein the common symptom signature is assigned a low weight value and the unique symptom signature is assigned a high weight value.
10. The storage medium of claim 6 or 9, wherein the retrieving step comprises: and obtaining the disease weight of each disease label according to the weight value of the symptom characteristic corresponding to the consultation information or the feedback information, so as to count the probability of suffering from each disease, and extracting one or more disease labels with the highest probability and the symptom characteristic corresponding to the disease labels as the retrieval result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910842570.8A CN110610766A (en) | 2019-09-06 | 2019-09-06 | Apparatus and storage medium for deriving probability of disease based on symptom feature weight |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910842570.8A CN110610766A (en) | 2019-09-06 | 2019-09-06 | Apparatus and storage medium for deriving probability of disease based on symptom feature weight |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110610766A true CN110610766A (en) | 2019-12-24 |
Family
ID=68892308
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910842570.8A Pending CN110610766A (en) | 2019-09-06 | 2019-09-06 | Apparatus and storage medium for deriving probability of disease based on symptom feature weight |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110610766A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111445993A (en) * | 2020-03-25 | 2020-07-24 | 北京嘉和海森健康科技有限公司 | Disease hierarchical relation processing method and device and electronic equipment |
| CN111462909A (en) * | 2020-03-30 | 2020-07-28 | 安徽科大讯飞医疗信息技术有限公司 | Disease evolution tracking and disease condition prompting method and device and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106845061A (en) * | 2016-11-02 | 2017-06-13 | 百度在线网络技术（北京）有限公司 | Intelligent interrogation system and method |
| CN109102886A (en) * | 2018-08-20 | 2018-12-28 | 重庆柚瓣家科技有限公司 | The disease of old people reasoning diagnostic system of more reasoning pattern fusions |
| CN109817329A (en) * | 2019-01-21 | 2019-05-28 | 暗物智能科技(广州)有限公司 | A kind of medical treatment interrogation conversational system and the intensified learning method applied to the system |
-
2019
- 2019-09-06 CN CN201910842570.8A patent/CN110610766A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106845061A (en) * | 2016-11-02 | 2017-06-13 | 百度在线网络技术（北京）有限公司 | Intelligent interrogation system and method |
| CN109102886A (en) * | 2018-08-20 | 2018-12-28 | 重庆柚瓣家科技有限公司 | The disease of old people reasoning diagnostic system of more reasoning pattern fusions |
| CN109817329A (en) * | 2019-01-21 | 2019-05-28 | 暗物智能科技(广州)有限公司 | A kind of medical treatment interrogation conversational system and the intensified learning method applied to the system |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111445993A (en) * | 2020-03-25 | 2020-07-24 | 北京嘉和海森健康科技有限公司 | Disease hierarchical relation processing method and device and electronic equipment |
| CN111445993B (en) * | 2020-03-25 | 2023-08-08 | 北京嘉和海森健康科技有限公司 | Hierarchical relationship processing method and device for diseases and electronic equipment |
| CN111462909A (en) * | 2020-03-30 | 2020-07-28 | 安徽科大讯飞医疗信息技术有限公司 | Disease evolution tracking and disease condition prompting method and device and electronic equipment |
| CN111462909B (en) * | 2020-03-30 | 2024-04-05 | 讯飞医疗科技股份有限公司 | Disease evolution tracking and disease condition prompting method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ambekar et al. | Disease risk prediction by using convolutional neural network | |
| KR20170061222A (en) | The method for prediction health data value through generation of health data pattern and the apparatus thereof | |
| CN110675944A (en) | Triage method and device, computer equipment and medium | |
| JP2018067303A (en) | Diagnosis support method, program and apparatus | |
| CN111564223B (en) | Infectious disease survival probability prediction method, and prediction model training method and device | |
| CN110391021A (en) | A kind of disease inference system based on medical knowledge map | |
| CN112151130B (en) | Decision support system based on literature retrieval and construction method | |
| CN112541066B (en) | Text-structured-based medical and technical report detection method and related equipment | |
| Falis et al. | Ontological attention ensembles for capturing semantic concepts in ICD code prediction from clinical text | |
| CN111524570B (en) | Ultrasonic follow-up patient screening method based on machine learning | |
| CN111581969B (en) | Medical term vector representation method, device, storage medium and electronic equipment | |
| Chen et al. | A deep-learning based ultrasound text classifier for predicting benign and malignant thyroid nodules | |
| Mandal | Machine learning algorithms for the creation of clinical healthcare enterprise systems | |
| CN110610766A (en) | Apparatus and storage medium for deriving probability of disease based on symptom feature weight | |
| CN112837765B (en) | Automatic coding method and system for disease diagnosis | |
| CN112967803A (en) | Early mortality prediction method and system for emergency patients based on integrated model | |
| CN117352159A (en) | Method, system and storage medium for evidence-based treatment of difficult and complicated diseases based on electronic medical records | |
| CN107679174A (en) | Construction method, device and the server of Knowledge Organization System | |
| CN115862897A (en) | Syndrome monitoring method and system based on clinical data | |
| CN109192306A (en) | A kind of judgment means of diabetes, equipment and computer readable storage medium | |
| CN110033862B (en) | Traditional Chinese medicine quantitative diagnosis system based on weighted directed graph and storage medium | |
| CN110289065A (en) | A kind of auxiliary generates the control method and device of medical electronic report | |
| CN112700861A (en) | Accompanying symptom interaction method and accompanying symptom interaction system | |
| Wu et al. | OA-MedSQL: Order-Aware Medical Sequence Learning for Clinical Outcome Prediction | |
| Kandel | Deep Learning Techniques for Medical Image Classification |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191224 |