CN104679997A - Method for building disease risk coefficient evaluation model - Google Patents
Method for building disease risk coefficient evaluation model Download PDFInfo
- Publication number
- CN104679997A CN104679997A CN201510063639.9A CN201510063639A CN104679997A CN 104679997 A CN104679997 A CN 104679997A CN 201510063639 A CN201510063639 A CN 201510063639A CN 104679997 A CN104679997 A CN 104679997A
- Authority
- CN
- China
- Prior art keywords
- disease
- risk coefficient
- signal
- assessment models
- coefficient
- 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
Landscapes
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses a method for building a disease risk coefficient evaluation model. The method comprises a first step of acquiring signals; a second step of processing the signals; a third step of obtaining disease types, wherein (1) if the disease type is known, performing a fourth step; (2) if the disease type is unknown, performing pattern recognition; the fourth step of calculating a disease risk coefficient, wherein the step comprises a. building the evaluation model; b. extracting features; c. obtaining the risk coefficient. Compared with the prior art, the method disclosed by the invention can evaluate the severity of the disease of a patient, and help the patient to know the disease development. The method is quite suitable for automatically diagnosing and analyzing after various medical and electronic products acquire the signals, and also suitable for monitoring and tracking the condition of the patient after the confirmed diagnosis of the disease is performed by a hospital.
Description
Technical field
The present invention relates to a kind of electronic information medical domain, particularly relate to a kind of method building disease risk coefficient assessment models.
Background technology
Bio signal comprises abundant information.At medical field, by carrying out analyzing some index that can obtain reflecting function of human body state to these signals, and then treat for medical diagnosis on disease and offer help.In order to learn whether user suffers from certain disease, and the bio signal of collection is carried out signal transacting.Extract the feature be applicable to, after Sample Storehouse comparison, deducibility user may suffer from certain disease.And the mode of this automatic diagnosis, because cost is low, efficiency is high, be more and more applied in various domestic medicine, ambulatory medical device, to replace doctor to a certain extent, complete some simple medicals diagnosis on disease.
Summary of the invention
Object of the present invention is just to build a kind of by inputting biology, medical signals export the order of severity that a quantification index reflects certain disease, and disease time is predicted, the method for the disease risk coefficient assessment models that can offer help for relevant disease diagnosis and treatment to a certain extent.
To achieve these goals, the technical solution used in the present invention is: a kind of method building disease risk coefficient assessment models, and method step is as follows:
Step one, collection signal
Biology sensor is utilized to gather bio signal;
Step 2, signal transacting
Filtering and noise reduction is carried out to the bio signal gathered;
Step 3, obtains disease type
(1) if known disease type, the 4th step is directly carried out;
(2) if unidentified illness type, then pattern-recognition is carried out
A. feature extraction is carried out to the bio signal gathered;
B. be input in SVM classifier, obtain classification results, then carry out the 4th step;
Step 4, calculates disease risk coefficient.
Calculate disease risk coefficient method as follows,
A. the foundation of assessment models
This model, by suffering from the signal of this serious Disease colony, after the feature selecting this kind of disease discrimination the highest is extracted, is trained in input SVDD and is obtained;
B. feature extraction
By the classification of diseases that step 3 obtains, signal characteristic used when extracting the assessment models training of this disease;
C. danger coefficient is obtained
By the signal characteristic extracted, be input in the assessment models based on SVDD, export danger coefficient.
When calculating danger coefficient, this model supports vector based on SVDD constitutes a suprasphere, if more near from this ball centre, represent that the difference of itself and severe patient is less, then the explanation state of an illness is more serious; Otherwise the state of an illness is lighter, by the feature of test sample book and the distance at this suprasphere center, provide corresponding coefficient.
Compared with prior art, the invention has the advantages that: the order of severity that patient suffers from the disease can be weighed, contribute to it and understand PD.After being highly suitable for various medical treatment, electronic product collection bio signal, carry out automatic diagnosis and analysis.Also be applicable to after hospital's diagnosed disease, to monitoring and the tracking of the state of an illness.
Embodiment
Embodiment: the invention will be further described below, a kind of method building disease risk coefficient assessment models, method step is as follows:
Step one, collection signal
Biology sensor is utilized to gather bio signal;
Step 2, signal transacting
Filtering and noise reduction is carried out to the bio signal gathered;
Step 3, obtains disease type
(1) if known disease type, the 4th step is directly carried out;
(2) if unidentified illness type, then pattern-recognition is carried out
A. feature extraction is carried out to the bio signal gathered;
B. be input in SVM classifier, obtain classification results, then carry out the 4th step;
Step 4, calculates disease risk coefficient.
A. the foundation of assessment models
This model, by suffering from the signal of this serious Disease colony, after the feature selecting this kind of disease discrimination the highest is extracted, is trained in input SVDD and is obtained;
B. feature extraction
The classification of diseases that obtained by step 3 (sorter automatic diagnosis or user known), signal characteristic used when extracting the assessment models training of this disease;
C. danger coefficient is obtained
By the signal characteristic extracted, be input in the assessment models based on SVDD, export danger coefficient.
Analytically the angle of geometry is seen, this model supports vector based on SVDD constitutes a suprasphere.If from this ball centre more close to, represent that the difference of itself and severe patient is less, then illustrate that the state of an illness is more serious; Otherwise the state of an illness is lighter.By the feature of test sample book and the distance at this suprasphere center, provide corresponding coefficient.If distance is nearer, then coefficient is higher; Distance is far away, then coefficient is lower.
Realize system of the present invention and generally include following a few part: signals collecting, signal transacting, feature extraction, classifier design.Wherein crucial part is feature extraction and classifier design part.Support Vector data description (SVDD) is a kind of sorter for oneclass classification problem, adopts this sorter can calculate the core position of target sample feature, thus judges the distance of test sample book and certain class sample.Collection signal, and carry out filtering and noise reduction process; From signal extraction feature, if do not know respective type, and put into sorter and classify; If known respective type, according to classification results or known type, extract suitable feature and be input to the assessment models (based on SVDD) of corresponding classification, exporting the coefficient of the order of severity of this signal of reflection in this kind of disease.This assessment models is that crowd's (heartbeat signal as long-term morbidity) training of this disease by suffering from the order of severity obtains.Adopt this kind of method, the order of severity that patient suffers from the disease can be weighed, contribute to it and understand PD.After being highly suitable for various medical treatment, electronic product collection bio signal, carry out automatic diagnosis and analysis.Also be applicable to after hospital's diagnosed disease, to monitoring and the tracking of the state of an illness.
Above exhaustive presentation is carried out to a kind of method building disease risk coefficient assessment models provided by the present invention, apply specific case herein to set forth principle of the present invention and embodiment, the explanation of above embodiment just understands method of the present invention and core concept thereof for helping; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, all will change in specific embodiments and applications, to change of the present invention and improve will be possible, and design and the scope of accessory claim defined can not be exceeded, in sum, this description should not be construed as limitation of the present invention.
Claims (3)
1. build a method for disease risk coefficient assessment models, it is characterized in that, method step is as follows:
Step one, collection signal
Biology sensor is utilized to gather bio signal;
Step 2, signal transacting
Filtering and noise reduction is carried out to the bio signal gathered;
Step 3, obtains disease type
(1) if known disease type, the 4th step is directly carried out;
(2) if unidentified illness type, then pattern-recognition is carried out
A. feature extraction is carried out to the bio signal gathered;
B. be input in SVM classifier, obtain classification results, then carry out the 4th step;
Step 4, calculates disease risk coefficient.
2. a kind of method building disease risk coefficient assessment models according to claim 1, is characterized in that: step 4, calculates disease risk coefficient method as follows,
A. the foundation of assessment models
This model, by suffering from the signal of this serious Disease colony, after the feature selecting this kind of disease discrimination the highest is extracted, is trained in input SVDD and is obtained;
B. feature extraction
By the classification of diseases that step 3 obtains, signal characteristic used when extracting the assessment models training of this disease;
C. danger coefficient is obtained
By the signal characteristic extracted, be input in the assessment models based on SVDD, export danger coefficient.
3. a kind of method building disease risk coefficient assessment models according to claim 2, it is characterized in that: when step c calculates danger coefficient, this model supports vector based on SVDD constitutes a suprasphere, if from this ball centre more close to, represent that the difference of itself and severe patient is less, then illustrate that the state of an illness is more serious; Otherwise the state of an illness is lighter, by the feature of test sample book and the distance at this suprasphere center, provide corresponding coefficient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510063639.9A CN104679997A (en) | 2015-02-06 | 2015-02-06 | Method for building disease risk coefficient evaluation model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510063639.9A CN104679997A (en) | 2015-02-06 | 2015-02-06 | Method for building disease risk coefficient evaluation model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104679997A true CN104679997A (en) | 2015-06-03 |
Family
ID=53315030
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510063639.9A Pending CN104679997A (en) | 2015-02-06 | 2015-02-06 | Method for building disease risk coefficient evaluation model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104679997A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184107A (en) * | 2015-10-20 | 2015-12-23 | 沈阳国际旅行卫生保健中心 | Health risk pre-warning system for human body circulatory system |
| CN105303041A (en) * | 2015-10-20 | 2016-02-03 | 沈阳国际旅行卫生保健中心 | Human digestive system health risk early-warning system |
| CN107391901A (en) * | 2017-05-05 | 2017-11-24 | 陈昕 | Establish the method and server of public ward conditions of patients assessment models |
| CN110875087A (en) * | 2018-09-03 | 2020-03-10 | 广州呼吸健康研究院 | Chronic lung disease management system |
-
2015
- 2015-02-06 CN CN201510063639.9A patent/CN104679997A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184107A (en) * | 2015-10-20 | 2015-12-23 | 沈阳国际旅行卫生保健中心 | Health risk pre-warning system for human body circulatory system |
| CN105303041A (en) * | 2015-10-20 | 2016-02-03 | 沈阳国际旅行卫生保健中心 | Human digestive system health risk early-warning system |
| CN107391901A (en) * | 2017-05-05 | 2017-11-24 | 陈昕 | Establish the method and server of public ward conditions of patients assessment models |
| CN110875087A (en) * | 2018-09-03 | 2020-03-10 | 广州呼吸健康研究院 | Chronic lung disease management system |
| CN110875087B (en) * | 2018-09-03 | 2024-06-07 | 广州呼吸健康研究院 | Chronic pulmonary disease management system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Heart sound classification based on scaled spectrogram and partial least squares regression | |
| Wang et al. | Localizing microaneurysms in fundus images through singular spectrum analysis | |
| Alaskar et al. | The implementation of pretrained AlexNet on PCG classification | |
| CN106653001B (en) | Method and system for identifying baby crying | |
| Chen et al. | Computerized wrist pulse signal diagnosis using modified auto-regressive models | |
| CN110706786B (en) | Non-contact intelligent psychological parameter analysis and evaluation system | |
| Banerjee et al. | PhotoECG: Photoplethysmographyto estimate ECG parameters | |
| CN104679997A (en) | Method for building disease risk coefficient evaluation model | |
| Milani et al. | Abnormal heart sound classification using phonocardiography signals | |
| KR20170047423A (en) | Automatic diagnosis system of digital xray based on cad | |
| Nascimento et al. | Heart arrhythmia classification based on statistical moments and structural co-occurrence | |
| CN109273094A (en) | A kind of construction method and building system of the Kawasaki disease risk evaluation model based on Boosting algorithm | |
| CN112819045A (en) | Method, apparatus, electronic device and medium for user physiological state classification | |
| CN103886184A (en) | Construction method for heart pathology recognition model | |
| Li et al. | Multi-modal cardiac function signals classification algorithm based on improved DS evidence theory | |
| Guergueb et al. | Melanoma skin cancer detection using recent deep learning models | |
| Yang et al. | Automated intra-patient and inter-patient coronary artery disease and congestive heart failure detection using EFAP-Net | |
| CN104679967A (en) | Method for judging reliability of psychological test | |
| Wiley et al. | Automatic detection of excessive glycemic variability for diabetes management | |
| CN110532926A (en) | Pig neurogenic disease intelligence Forecasting Method based on deep learning | |
| Silviya et al. | Deep learning based plant leaf disease detection and classification | |
| Hossain et al. | MetaParkinson: A Cyber-Physical Deep Meta-Learning Framework for n-Shot Diagnosis and Monitoring of Parkinson's Patients | |
| Sayed et al. | Arrhythmia classification based on novel distance series transform of phase space trajectories | |
| CN108338777A (en) | A kind of pulse signal determination method and device | |
| CN108062570A (en) | A kind of pattern recognition system for screening lung cancer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into 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: 20150603 |