CN113040711B - Cerebral apoplexy incidence risk prediction system, equipment and storage medium - Google Patents
Cerebral apoplexy incidence risk prediction system, equipment and storage medium Download PDFInfo
- Publication number
- CN113040711B CN113040711B CN202110233465.1A CN202110233465A CN113040711B CN 113040711 B CN113040711 B CN 113040711B CN 202110233465 A CN202110233465 A CN 202110233465A CN 113040711 B CN113040711 B CN 113040711B
- Authority
- CN
- China
- Prior art keywords
- data
- classifier
- weak
- sample
- weak classifier
- 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.)
- Active
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Fuzzy Systems (AREA)
- Evolutionary Computation (AREA)
- Pathology (AREA)
- Mathematical Physics (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a cerebral apoplexy incidence risk prediction system, which comprises: a data acquisition unit: the method is used for collecting health data information of target people and performing desensitization treatment; pretreatment unit: the method is used for standardizing basic data information of target groups and marking stroke risk categories; feature screening unit: the method is used for screening cerebral apoplexy characteristic data based on an IV value analysis method to obtain characteristic data with higher model predictive value, and a data set is formed; model building unit: the method is used for training data by using an Adaboost reinforcement learning method based on logistic regression, and establishing a fusion model; model prediction unit: the method is used for predicting the sample data to be tested through the fusion model to obtain the stroke risk category. According to the invention, accurate analysis of stroke risk categories is realized based on the health data information of the target population, and the disease risk prediction efficiency can be improved.
Description
Technical Field
The invention belongs to the technical field of health management, and particularly relates to a cerebral apoplexy onset risk prediction system, equipment and a storage medium.
Background
"cerebral stroke" is also called "stroke" and is an acute cerebrovascular disease, which seriously threatens the life and health of human beings. Cerebral apoplexy has the characteristics of high morbidity, high mortality and high disability rate, and especially causes serious death or leaves the consequences of different degrees of disability for the old people, which extremely afflicts patients and families.
The medical community currently considers the prevention of stroke onset as the best control measure. However, in the prior art, risk prediction is only carried out unilaterally from clinical symptoms, so that the aims of accurate identification and early prevention cannot be achieved. The cerebral apoplexy incidence risk prediction model can be used for identifying high-risk individuals in the crowd and providing preventive intervention measures for preventing cerebral apoplexy for the high-risk individuals, so that the cerebral apoplexy can be identified and intervened prematurely. Therefore, the research design of the cerebral apoplexy incidence risk prediction model which is accurate in evaluation, convenient and easy to use has important clinical value and has important significance for preventing and controlling cerebral apoplexy.
Disclosure of Invention
In view of the above, the invention provides a cerebral apoplexy onset risk prediction system, equipment and a storage medium, which are used for providing a cerebral apoplexy onset risk prediction scheme with convenience, easiness in use and high accuracy for high-risk individuals.
In a first aspect of the present invention, a system for predicting risk of developing cerebral apoplexy is disclosed, the system comprising:
a data acquisition unit: the method is used for collecting health data information of target people and performing desensitization treatment;
pretreatment unit: the method is used for standardizing basic data information of target groups and marking stroke risk categories;
feature screening unit: the method is used for screening cerebral apoplexy characteristic data based on an IV value analysis method to obtain characteristic data with higher model predictive value, and a data set is formed;
model building unit: the method is used for training data by using an Adaboost reinforcement learning method based on logistic regression, and establishing a fusion model;
model prediction unit: the method is used for predicting the sample data to be tested through the fusion model to obtain the stroke risk category.
Preferably, in the data acquisition unit, the target population health data information includes name, gender, age, weight, occupation, residence, blood pressure, heart rate, blood oxygen, atrial fibrillation, smoking history, alcoholism history, diabetes history, cardiovascular disease history, family cerebral apoplexy history, obesity, carotid stenosis associated symptoms, limb weakness associated symptoms, speech disorder associated symptoms and corresponding associated symptom duration.
Preferably, the normalization process includes: performing type conversion, data filling and data deleting on irregular and non-uniform data; the data conversion includes binary type data conversion and continuous type data conversion.
Preferably, the model building unit specifically includes:
weak classification subunit: constructing weak classifiers by using a plurality of LR classifiers, initializing a sample weight value, and training an ith weak classifier based on LR by adopting a one-VS-rest form; taking the maximum cerebral apoplexy probability value as the output of each weak classifier, wherein the number of LR classifiers in each weak classifier is equal to the cerebral apoplexy risk class number;
calculating the output weight value alpha of the ith weak classifier i ：α i ＝max(ln((1-f i )/f i )+ln2,0)，f i The sum of weight values of samples misclassified by the weak classifier is used for training the samples;
strong classification subunit: updating the strong classifier based on the weak classifiers, and calculating the error rate of classification of the strong classifier;
a circulation subunit: judging whether a cycle ending condition is reached, if yes, ending the cycle, otherwise, recalculating a sample weight value according to the classification error rate of the ith weak classifier, and adding a new sample weight value into a sample for training; and the cycle ending condition is to judge whether the classification error rate of the strong classifier is 0, if so, ending the cycle, otherwise, judging whether the number of the weak classifiers is greater than or equal to the set maximum number of the weak classifiers, and if so, ending the cycle.
Preferably, the expression of the updating strong classifier is:wherein H is fq For a strong classifier before update, +.>For the updated strong classifier, η 1 ∈[0,1]For learning rate alpha j Output weight value for jth weak classifier,/>The input of the weak classifier is that l is the return value of the argmax function, namely the stroke risk class level predicted value.
Preferably, the expression for calculating the error rate of the strong classifier class is:wherein sign is a sign function, T k Is characteristic of the kth training sample, if the kth sample is a positive sample, S k For corresponding T k Is the stroke risk level, m is the training set sample numberA number.
Preferably, the recalculating the sample weight value according to the classification error rate of the ith weak classifier, and adding the new sample weight value to the sample specifically includes:
calculating a new sample weight value of the (i+1) th weak classifier according to the classification error rate of the (i) th weak classifier: e, e i+1k ＝e ik ·exp(α i ·[H i fq (T k )≠S k ]),k∈[1,2,…,m]，α i The output weight value e of the ith weak classifier ik The kth sample weight value, e, for the ith weak classifier i+1k The kth sample weight value for the (i+1) th weak classifier; and processing the weight value newly generated by the sample and carrying out normalization processing.
In a second aspect of the present invention, an electronic device is disclosed, comprising: at least one processor, at least one memory, a communication interface, and a bus;
the processor, the memory and the communication interface complete communication with each other through the bus;
the memory stores program instructions executable by the processor, which are called by the processor to implement the system according to the first aspect of the invention.
In a third aspect of the present invention, a computer-readable storage medium is disclosed, the computer-readable storage medium storing computer instructions that cause the computer to implement the system according to the first aspect of the present invention.
Compared with the prior art, the invention has the following beneficial effects:
1) According to the invention, by means of technologies such as data analysis and processing and the like and combining an IV value analysis method, feature screening is carried out on a cerebral apoplexy data sample, features with low cerebral apoplexy risk prediction value are removed, feature dimension is reduced, then a cerebral apoplexy risk prediction model is constructed through an Adaboost algorithm fused with logistic regression, classification categories to which cerebral apoplexy risks to be identified belong are calculated, a novel solution is provided for the traditional cerebral apoplexy identification and diagnosis problem, accurate analysis on cerebral apoplexy risk categories is realized based on target crowd health data information, and diagnosis efficiency is improved.
2) According to the invention, the weak classifier is constructed by a plurality of LR classifiers, the one-VS-rest mode is adopted, the weak classifier is trained, the strong classifier is formed based on the plurality of weak classifiers, and the accuracy of predicting the cerebral apoplexy incidence risk can be improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a block diagram of a stroke risk prediction system according to the present invention;
FIG. 2 is a schematic diagram of a model building block according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will clearly and fully describe the technical aspects of the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.
Referring to fig. 1, the present invention provides a cerebral apoplexy onset risk prediction system, which includes: the device comprises a data acquisition unit 10, a preprocessing unit 20, a feature screening unit 30, a model establishing unit 40 and a model predicting unit 50;
the data acquisition unit 10 is used for acquiring health data information of target people and performing desensitization treatment;
the target population health data information includes name, gender, age, weight, occupation, residence, blood pressure, heart rate, blood oxygen, atrial fibrillation, smoking history, alcoholism history, diabetes history, cardiovascular disease history, family stroke history, whether obesity, and whether carotid stenosis syndrome, limb weakness syndrome, speech disorder syndrome and corresponding syndrome duration. And code desensitization protection is carried out on sensitive privacy information.
The preprocessing unit 20 is used for standardizing basic data information of target crowd and labeling stroke risk categories;
the normalization process includes: performing type conversion, data filling and data deleting on irregular and non-uniform data; the data conversion includes binary type data conversion and continuous type data conversion. Binary data conversion such as sex field "male" or "female", residence field "south" or "north", blood pressure ≡140/90mmHg field "yes" or "no", smoking history, alcoholism history, atrial fibrillation, carotid stenosis, accompanying limb weakness, accompanying speech disorder, diabetes history, cardiovascular disease history, family stroke history, etc. "yes" or "no", respectively, may represent "0" or "1"; continuous data conversion such as blood oxygen [ 95-99%, 90-94%, 85-89%, 70-84% to "0", "1", "2", "3" ], heart rate (times/min) [ 55-90, 91-100, 101-110, 111-130, 131-180 to "0", "1", "2", "3", "4" ], etc. And filling data, namely filling the average value of the relevant null value fields of each health item, so as to improve the accuracy of model training. The stroke risk type is marked as a single category, namely the category label only corresponds to one of a risk category set (low risk, medium risk and high risk).
The feature screening unit 30 is configured to screen feature data of stroke based on an IV value analysis method, obtain feature data with higher model predictive value, and form a data set;
the IV value analysis method measures the influence degree of a certain feature on a target, and the basic idea is to compare and calculate the association degree according to the ratio of the hit black-white samples of the feature to the ratio of the total black-white samples, and the IV value analysis method is carried out on the stroke sample data to obtain the screened sample feature data, thereby being beneficial to reducing the interference of noise information, accelerating the convergence rate in model training and reducing the time cost. For example, for the health data information of a certain target group, 14 characteristics with great influence on cerebral apoplexy are finally selected by an IV value analysis method and used as input variables of a subsequent model, wherein the input variables comprise age, weight, smoking history, alcoholism history, blood pressure, heart rate, blood oxygen, atrial fibrillation, carotid stenosis, limb weakness, symptom duration, diabetes history, cardiovascular disease history and family cerebral apoplexy history.
The model building unit 40 is configured to train data and build a fusion model by using an Adaboost reinforcement learning method based on logistic regression;
based on an LR and Adaboost fusion model, the method comprises 1 strong classifier and a plurality of weak classifiers, wherein the weak classifiers comprise a plurality of LR classifiers, classification is carried out in a one-VS-rest mode, and the maximum cerebral apoplexy probability value is used as the output of each weak classifier; the number of LR classifiers in each weak classifier is equal to the number of stroke risk categories. Referring to fig. 2, the model building unit 40 specifically includes:
weak classification subunit 41: constructing weak classifiers by using a plurality of LR classifiers, and training an ith weak classifier based on LR by adopting a one-VS-rest form; calculating the output weight value of the ith weak classifier according to the classification error rate; taking the maximum cerebral apoplexy probability value as the output of each weak classifier, wherein the number of LR classifiers in each weak classifier is equal to the cerebral apoplexy risk class number;
specifically, let the training set be expressed as r= { (T) 1 ,S 1 ),…(T k ,S k ) }, T therein k Is characteristic of the kth training sample, T k ∈{t k1 ,t k2 ,…,t kp }，t kl (l=1, 2, …, p) is the first element of the kth sample feature, S k E {1,2,3} is the corresponding T k The stroke risk level of the device is { low risk, medium risk and high risk }, m is the number of samples of the training set and accounts for 80% of the total sample data; in addition, the maximum weak classifier number g=60 and the learning rate eta are set 1 =0.8 and η 2 =0.01, regularization parameter λ=110, LR maximum number of iterations n=100;
the current weak classifier index i=1 is initialized,initializing a sample weight value as e ik ＝1/m，k∈[1,2,…,m]Initializing a strong classifier H fq =0; weak classifier C i Consists of 3 LR classifiers, i.e. c 1 ,c 2 ,c 3 Weak classifier C i The output of (i) isa is the number of the LR classifier in each weak class, where +.>w is regression coefficient, ++>Is the input of the classifier; and taking m pieces of training sample data as input, and adjusting regression coefficients by using a gradient descent method according to the initial values, wherein the regression coefficients are as follows:
j=1, 2, … p, if the kth sample is a positive sample,if the kth sample is a negative sample, +.>Training therefore to obtain the i-th weak classifier C based on LR i I=1, 2, …, g; calculating the weak classifier classification error rate (i.e. the sum of the weight values of the training samples misclassified by the weak classifier) taking m samples as input:
calculating the output weight value alpha of the ith weak classifier i ：α i ＝max(ln((1-f i )/f i )+ln2,0)。
Strong classification subunit 42: updating the strong classifier based on the weak classifiers, and calculating the error rate of classification of the strong classifier;
the expression of the updated strong classifier is:wherein H is fq For a strong classifier before update, +.>For the updated strong classifier, η 1 ∈[0,1]For learning rate alpha j Output weight value for jth weak classifier,/>The input of the weak classifier is that l is the return value of the argmax function, namely the stroke risk class level predicted value.
The expression for calculating the error rate of the strong classifier classification is:wherein sign is a sign function, T k Is characteristic of the kth training sample, if the kth sample is a positive sample, S k For corresponding T k Is the risk level of cerebral apoplexy, and m is the number of samples of the training set.
Circulation subunit 43: judging whether a cycle ending condition is reached, if yes, ending the cycle, otherwise, recalculating a sample weight value according to the classification error rate of the ith weak classifier, and adding a new sample weight value into a sample for training; and the cycle ending condition is to judge whether the classification error rate of the strong classifier is 0, if so, ending the cycle, otherwise, judging whether the number of the weak classifiers is greater than or equal to the set maximum number of the weak classifiers, and if so, ending the cycle.
The recalculating the sample weight value according to the classification error rate of the ith weak classifier, and adding the new sample weight value into the sample specifically comprises:
calculating a new sample weight value of the (i+1) th weak classifier according to the classification error rate of the (i) th weak classifier: e, e i+1k ＝e ik ·exp(α i ·[H i fq (T k )≠S k ]),k∈[1,2,…,m]，α i The output weight value e of the ith weak classifier ik The kth sample weight value, e, for the ith weak classifier i+1k The kth sample weight value for the (i+1) th weak classifier;
let i=i+1, process the weight value newly generated by the sample and perform normalization processing:
the model prediction unit 50 is configured to predict sample data to be tested by fusing a model, so as to obtain a stroke risk category.
Through comparison, after characteristic data are screened by fully utilizing IV value analysis, the accuracy of a fusion model classifier can be remarkably improved, the convergence speed of the model is accelerated in the training process, the calculation time cost is saved, and finally the generalization capability of a prediction model can be improved; meanwhile, the weak classifier is built by a plurality of LR classifiers, the one-VS-rest mode is adopted, the weak classifier based on LR is trained, the strong classifier is built on the basis of the plurality of weak classifiers, the crowd of potential risks of cerebral apoplexy is predicted, and a certain help can be provided for patients and clinicians. The cerebral apoplexy incidence risk prediction model can realize early discovery of high-risk individuals in target groups, and has important significance for providing preventive measures for the high-risk individuals.
The invention also discloses an electronic device, comprising: at least one processor, at least one memory, a communication interface, and a bus; the processor, the memory and the communication interface complete communication with each other through the bus; the memory stores program instructions executable by the processor that the processor invokes to implement the aforementioned methods of the present invention.
The invention also discloses a computer readable storage medium storing computer instructions for causing a computer to implement all or part of the steps of the methods of the embodiments of the invention. The storage medium includes: a usb disk, a removable hard disk, a ROM, a RAM, a magnetic or optical disk, or other various media capable of storing program code.
The system embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, i.e., may be distributed over a plurality of network elements. Some or all of the modules may be selected according to the actual government office in feudal China to achieve the purpose of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (7)
1. A stroke morbidity risk prediction system, the system comprising:
a data acquisition unit: the method is used for collecting health data information of target people and performing desensitization treatment;
pretreatment unit: the method is used for standardizing basic data information of target groups and marking stroke risk categories;
feature screening unit: the method is used for screening cerebral apoplexy feature data based on an IV value analysis method to obtain feature data with model prediction value higher than a preset threshold value, and a data set is formed;
model building unit: the method is used for training data by using an Adaboost reinforcement learning method based on logistic regression, and establishing a fusion model; the model building unit specifically comprises:
weak classification subunit: constructing a weak classifier by using a plurality of LR classifiers, initializing a sample weight value, and training the weak classifier based on LR by adopting a one-VS-rest form; taking the maximum cerebral apoplexy probability value as the output of each weak classifier, wherein the number of LR classifiers in each weak classifier is equal to the cerebral apoplexy risk class number;
calculating the output weight value of the weak classifier, and outputting the weight value alpha of the ith weak classifier i The method comprises the following steps: alpha i ＝max(ln((1-f i )/f i )+ln2,0)，f i The sum of weight values of samples misclassified by the weak classifier is used for training the samples;
strong classification subunit: updating the strong classifier based on the weak classifiers, and calculating the error rate of classification of the strong classifier;
a circulation subunit: judging whether a cycle ending condition is reached, if yes, ending the cycle, otherwise, recalculating a sample weight value according to the classification error rate of the ith weak classifier, and adding a new sample weight value into a sample for training; the cycle ending condition is to judge whether the classification error rate of the strong classifier is 0, if so, ending the cycle, otherwise, judging whether the number of the weak classifiers is greater than or equal to the set maximum number of the weak classifiers, and if so, ending the cycle;
model prediction unit: the method comprises the steps of predicting sample data to be tested through a fusion model to obtain stroke risk categories;
the target population health data information includes name, gender, age, weight, occupation, residence, blood pressure, heart rate, blood oxygen, atrial fibrillation, smoking history, alcoholism history, diabetes history, cardiovascular disease history, family stroke history, whether obesity, and whether carotid stenosis syndrome, limb weakness syndrome, speech disorder syndrome and corresponding syndrome duration.
2. The stroke risk prediction system according to claim 1, wherein the normalization process comprises: performing type conversion, data filling and data deleting on irregular and non-uniform data; the data conversion includes binary type data conversion and continuous type data conversion.
3. The stroke risk prediction system according to claim 1, wherein the expression of the updated strong classifier is:where a is the number of the LR classifier in each weak class, H fq For a strong classifier before update, +.>For the updated strong classifier, η 1 ∈[0,1]For learning rate alpha j Output weight value for jth weak classifier,/>The input of the weak classifier is that l is the return value of the argmax function, namely the stroke risk class level predicted value.
4. The stroke risk prediction system according to claim 3, wherein the expression for calculating the error rate of the classification of the strong classifier is:wherein sign is a sign function, T k Is characteristic of the kth training sample, if the kth sample is a positive sample, S k For corresponding T k Is the risk level of cerebral apoplexy, and m is the number of samples of the training set.
5. The stroke risk prediction system according to claim 1, wherein the recalculating the sample weight value according to the classification error rate of the i-th weak classifier, and adding the new sample weight value to the sample specifically comprises:
calculating a new sample weight value of the (i+1) th weak classifier according to the classification error rate of the (i) th weak classifier: e, e i+1k ＝e ik ·exp(α i ·[H i fq (T k )≠S k ]),k∈[1,2,…,m]，α i The output weight value e of the ith weak classifier ik The kth sample weight value, e, for the ith weak classifier i+1k The kth sample weight for the (i+1) th weak classifierA heavy value; and processing the weight value newly generated by the sample and carrying out normalization processing.
6. An electronic device, comprising: at least one processor, at least one memory, a communication interface, and a bus;
the processor, the memory and the communication interface complete communication with each other through the bus;
the memory stores program instructions executable by the processor, the processor invoking the program instructions to implement the system of any of claims 1-5.
7. A computer readable storage medium storing computer instructions that cause the computer to implement the system of any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110233465.1A CN113040711B (en) | 2021-03-03 | 2021-03-03 | Cerebral apoplexy incidence risk prediction system, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110233465.1A CN113040711B (en) | 2021-03-03 | 2021-03-03 | Cerebral apoplexy incidence risk prediction system, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113040711A CN113040711A (en) | 2021-06-29 |
| CN113040711B true CN113040711B (en) | 2023-08-18 |
Family
ID=76509620
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110233465.1A Active CN113040711B (en) | 2021-03-03 | 2021-03-03 | Cerebral apoplexy incidence risk prediction system, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113040711B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113393935A (en) * | 2021-06-30 | 2021-09-14 | 吾征智能技术(北京)有限公司 | Method and system for evaluating early cancer risk based on physical examination data |
| CN113520319A (en) * | 2021-07-12 | 2021-10-22 | 吾征智能技术(北京)有限公司 | Epileptic event risk management method and system based on logistic regression |
| CN115620902A (en) * | 2021-07-15 | 2023-01-17 | 华为云计算技术有限公司 | Method and device for predicting survival risk rate |
| CN113876437B (en) * | 2021-09-13 | 2024-02-23 | 上海微创医疗机器人（集团）股份有限公司 | Storage medium, robot system, and computer device |
| CN114093448B (en) * | 2021-11-24 | 2022-07-01 | 首都医科大学附属北京天坛医院 | Construction method of disease risk prediction model |
| CN114334157A (en) * | 2022-01-19 | 2022-04-12 | 北京富通东方科技有限公司 | Cerebral apoplexy risk prediction method and device |
| CN115019919B (en) * | 2022-06-16 | 2023-04-07 | 江苏省人民医院（南京医科大学第一附属医院） | Deep learning-based method for establishing function prediction model after early recovery of stroke |
| CN117238478A (en) * | 2023-06-27 | 2023-12-15 | 开滦总医院 | Cerebral apoplexy risk early warning model generation method based on sleep monitoring |
| CN117116475A (en) * | 2023-06-27 | 2023-11-24 | 开滦总医院 | Method, system, terminal and storage medium for predicting risk of ischemic cerebral apoplexy |
| CN117457220B (en) * | 2023-12-26 | 2024-04-12 | 深圳市健怡康医疗器械科技有限公司 | Informationized management method and system for neurological stroke patients |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015066564A1 (en) * | 2013-10-31 | 2015-05-07 | Cancer Prevention And Cure, Ltd. | Methods of identification and diagnosis of lung diseases using classification systems and kits thereof |
| CN111079978A (en) * | 2019-11-20 | 2020-04-28 | 辽宁工程技术大学 | Coal and gas outburst prediction method based on logistic regression and reinforcement learning |
| CN111968746A (en) * | 2020-07-21 | 2020-11-20 | 华中科技大学协和深圳医院 | Cerebral apoplexy risk prediction method and device based on hybrid deep transfer learning |
| CN112101413A (en) * | 2020-08-12 | 2020-12-18 | 海南大学 | Intelligent system for predicting cerebral apoplexy risk |
| CN112092675A (en) * | 2020-08-31 | 2020-12-18 | 长城汽车股份有限公司 | Battery thermal runaway early warning method, system and server |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112055878B (en) * | 2018-04-30 | 2024-04-02 | 皇家飞利浦有限公司 | Adjusting a machine learning model based on the second set of training data |
-
2021
- 2021-03-03 CN CN202110233465.1A patent/CN113040711B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015066564A1 (en) * | 2013-10-31 | 2015-05-07 | Cancer Prevention And Cure, Ltd. | Methods of identification and diagnosis of lung diseases using classification systems and kits thereof |
| CN111079978A (en) * | 2019-11-20 | 2020-04-28 | 辽宁工程技术大学 | Coal and gas outburst prediction method based on logistic regression and reinforcement learning |
| CN111968746A (en) * | 2020-07-21 | 2020-11-20 | 华中科技大学协和深圳医院 | Cerebral apoplexy risk prediction method and device based on hybrid deep transfer learning |
| CN112101413A (en) * | 2020-08-12 | 2020-12-18 | 海南大学 | Intelligent system for predicting cerebral apoplexy risk |
| CN112092675A (en) * | 2020-08-31 | 2020-12-18 | 长城汽车股份有限公司 | Battery thermal runaway early warning method, system and server |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113040711A (en) | 2021-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113040711B (en) | Cerebral apoplexy incidence risk prediction system, equipment and storage medium | |
| CN109036553B (en) | Disease prediction method based on automatic extraction of medical expert knowledge | |
| CN109659033B (en) | Chronic disease state of an illness change event prediction device based on recurrent neural network | |
| US20180158552A1 (en) | Interpretable deep learning framework for mining and predictive modeling of health care data | |
| Yu et al. | Using a multi-task recurrent neural network with attention mechanisms to predict hospital mortality of patients | |
| CN113486578B (en) | Method for predicting residual life of equipment in industrial process | |
| CN109949936A (en) | A kind of Risk Forecast Method of being hospitalized again based on deep learning mixed model | |
| CN112086195B (en) | Admission risk prediction method based on self-adaptive ensemble learning model | |
| CN110532398B (en) | Automatic family map construction method based on multi-task joint neural network model | |
| EP3891760A1 (en) | Method and data processing apparatus for generating real-time alerts about a patient | |
| CN108492877B (en) | Cardiovascular disease auxiliary prediction method based on DS evidence theory | |
| Moreira et al. | Evolutionary radial basis function network for gestational diabetes data analytics | |
| CN109036577A (en) | Diabetic complication analysis method and device | |
| CN111387938B (en) | Patient heart failure death risk prediction system based on characteristic rearrangement one-dimensional convolutional neural network | |
| Xu et al. | Intelligent emotion detection method based on deep learning in medical and health data | |
| CN116671919B (en) | Emotion detection reminding method based on wearable equipment | |
| CN110491506A (en) | Auricular fibrillation prediction model and its forecasting system | |
| CN115376638A (en) | Physiological characteristic data analysis method based on multi-source health perception data fusion | |
| Koyi et al. | A research survey on state of the art heart disease prediction systems | |
| Hao et al. | Classification of cardiovascular disease via a new softmax model | |
| CN117976200A (en) | Cerebral apoplexy risk prediction method and system | |
| Al Duhayyim et al. | An Ensemble Machine Learning Technique for Stroke Prognosis. | |
| CN111696674A (en) | Deep learning method and system for electronic medical record | |
| CN110633368A (en) | Deep learning classification method for early colorectal cancer unstructured data | |
| Siddiqa et al. | Robust Length of Stay Prediction Model for Indoor Patients. |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |