CN113571176A - Depression identification method based on blood routine test data - Google Patents
Depression identification method based on blood routine test data Download PDFInfo
- Publication number
- CN113571176A CN113571176A CN202110681222.4A CN202110681222A CN113571176A CN 113571176 A CN113571176 A CN 113571176A CN 202110681222 A CN202110681222 A CN 202110681222A CN 113571176 A CN113571176 A CN 113571176A
- Authority
- CN
- China
- Prior art keywords
- data
- blood routine
- depression
- test data
- routine test
- 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
- 239000008280 blood Substances 0.000 title claims abstract description 36
- 210000004369 blood Anatomy 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000009666 routine test Methods 0.000 title claims abstract description 26
- 238000003745 diagnosis Methods 0.000 claims abstract description 12
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 7
- 201000010099 disease Diseases 0.000 claims abstract description 5
- 238000012360 testing method Methods 0.000 claims description 22
- 238000012549 training Methods 0.000 claims description 16
- 238000013528 artificial neural network Methods 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 5
- 238000009534 blood test Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 210000002569 neuron Anatomy 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 2
- 238000007689 inspection Methods 0.000 claims 1
- 230000003014 reinforcing effect Effects 0.000 claims 1
- 238000013135 deep learning Methods 0.000 abstract description 2
- 239000010410 layer Substances 0.000 description 47
- 230000006870 function Effects 0.000 description 11
- 230000008569 process Effects 0.000 description 9
- 239000013598 vector Substances 0.000 description 8
- 238000013507 mapping Methods 0.000 description 5
- 208000020401 Depressive disease Diseases 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 230000004913 activation Effects 0.000 description 2
- 238000003759 clinical diagnosis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000008506 pathogenesis Effects 0.000 description 2
- 208000019901 Anxiety disease Diseases 0.000 description 1
- 206010012374 Depressed mood Diseases 0.000 description 1
- 208000019022 Mood disease Diseases 0.000 description 1
- 230000036506 anxiety Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 238000013399 early diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 208000020016 psychiatric disease Diseases 0.000 description 1
- 239000002356 single layer Substances 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
Images
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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/70—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mental therapies, e.g. psychological therapy or autogenous training
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Epidemiology (AREA)
- General Physics & Mathematics (AREA)
- Primary Health Care (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Child & Adolescent Psychology (AREA)
- Developmental Disabilities (AREA)
- Hospice & Palliative Care (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention provides a depression identification method based on blood routine test data. By preprocessing a large amount of blood routine test data samples, unsupervised feature learning is introduced before the traditional supervised algorithm, the data features of input information are enhanced, and the initial amount of parameters is standardized by understanding and judging the data in advance. The model for judging whether the person to be tested has the depression or not is constructed by only using blood routine test data based on the diagnostic label through the semi-supervised prediction characteristic of deep learning, and the problems of low disease identification efficiency and accuracy and higher experience dependence of the existing diagnosis method for the depression are solved.
Description
Technical Field
The invention relates to the field of testing medicine, in particular to a depression identification method based on blood routine test data.
Background
Depressive disorder, also called depressive disorder, is a mood disorder disease mainly characterized by persistent and low mood, which has high incidence rate and easily repeated disease condition and is one of mental disorder diseases seriously harming human health. Depression, a common disease, has now become the second largest "killer" in humans, and patients gradually have a tendency to become younger, especially with a continuous rise in the prevalence of students. Due to the complexity of pathogenesis of depression, the medical treatment and prevention of depression in China at present are in the situation of low recognition rate, and related pathogenesis is still in the exploration stage. At present, clinically, the identification and diagnosis of depression is mainly based on traditional psychiatric interviews, and the subjectivity of the clinical diagnosis makes early diagnosis still challenging. Through the communication exchange between the doctor and the patient and the filling of the relevant questionnaire scales, the method is lack of objectivity, depends on the working experience of the doctor and the corresponding professional background knowledge, and is time-consuming and labor-consuming.
Nowadays, with the continuous progress of bioinformatic technology, ways of monitoring depression are also continuously abundant. There have been individual studies to identify patients with depression through data such as electroencephalogram (EEG). The early discovery of depression tendency and mild depression can help timely and actively intervene, and can avoid further aggravation of depression. Therefore, the finding of a rapid and accurate depression identification and diagnosis method is of great significance.
Disclosure of Invention
The invention aims to provide a depression identification method based on blood routine test data. By the deep learning algorithm, a model for determining whether or not a subject has depression is constructed using only blood routine test data. The blood routine is a basic physical examination item, and is generally involved in physical examinations such as college entrance, employee attendance and the like. Through the correlation analysis of the conventional blood data and the depression, the problems of low disease identification efficiency and accuracy and higher experience dependence of the existing diagnosis method for the depression can be solved.
In order to achieve the purpose, the invention provides a depression identification method based on blood routine test data, and the following technical scheme is adopted.
A method for identifying depression based on blood routine test data is designed, which comprises the following steps.
S1: obtaining blood routine test sample data from a physical examination hospital and a mental disease hospital and carrying out data preprocessing, wherein the test sample data is merged according to blood routine detection items; and filling up the dimension missing value of the test sample data. And classifying data types as training data aiming at the keywords in the diagnosis result by screening the sample.
S2: initializing parameters between neural network layers by using a self-coding learning mode of a Restricted Boltzmann Machine (RBM) through a layer-by-layer learning strategy based on test sample data, namely directly assigning multiplicative bias and a weight connection matrix in the RBM to a weight matrix and bias of a corresponding layer, and strengthening data characteristics of input information.
S3: the diagnosis label of the existing sample is used as feedback learning of the output network neuron, the design stage of the tail end supervised classifier is completed, a model capable of classifying the blood routine test data characteristics is trained, and finally a diagnosis result network model is obtained to realize the classification of depression tendency.
Preferably, in S1, the specific steps of the blood routine test data preprocessing are as follows.
The test sample data is test index data of a blood routine laboratory, and the data preprocessing process of the test sample comprises data standardization, merging, area transposition and variable screening.
Furthermore, when the sample data shows that the patient receives the drug intervention, the positive samples are intensively excluded in the process of screening the sample data.
Further, the positive data set and the negative data set are paired, and the data set is randomly divided into a training set, a verification set and a test set according to the proportion of 80%, 10% and 10%.
Preferably, in S2, the sample information is used to train the RBM through a layer-by-layer learning strategy, and the network structure thereof is divided into two layers: display layervData for representing input and output, hidden layerhIs understood as the intrinsic expression of a variable.
Preferably, in S3, the supervised classifier is a feedforward neural network model with the data features output by the RBM network as input, and can update the weights through network-wide feedback, and through back propagation learning, the internal weights of the neural network can be automatically updated according to the error information obtained in each iteration to obtain the data classification model.
Preferably, the network learning comprises the following specific steps.
First, initialization between the input and the first hidden layer is performed. Let the input information bexAnd normalizing the data to obtain a visible layer:。
then, initializing parameters from the first hidden layer to the second hidden layer, and obtaining the output of the first hidden layer in the deep feedforward neural network by using the following formula:whereinThe data is normalized for the activation function on the first hidden layer to obtain:。
by doing so, the parameter initialization process is completed after each layer of parameters in the feature learning stage is trained, and the process belongs to unsupervised learning.
The second is fine tuning work, namely supervised learning. Regarding fine adjustment stageThe solution of (2) is consistent with the traditional deep feedforward neural network, and the end-to-end network is integrally finely adjusted by using a back propagation algorithm. The optimized objective function in the parameter initialization stage is solved by using a contrast divergence algorithm (an approximation algorithm), and the probability of the input sample under the distribution is maximized by constructing a log-likelihood function, so that the optimal parameter set is obtainedθ。
Each layer of RBM network can only ensure that the weight in the layer where the RBM network is located is optimal for the feature vector mapping of the layer, and is not optimal for the feature vector mapping of the whole model. Therefore, the back propagation network also propagates error information from top to bottom to the unit nodes of each layer and fine-tunes the overall model.
Further, in the process of model training, the classifier is evaluated by using the classification confusion matrix so as to complete the evaluation of the classifier.
The depression identification method based on the blood routine test data has the beneficial effect.
1. The invention provides a method for identifying depression by using blood conventional indexes aiming at the defects of the existing depression clinical diagnosis method, and improves the efficiency and the precision of integral depression diagnosis.
2. The invention aims at the high-order spatial characteristics of the conventional blood test data, uses the self-coding function of RBM, strengthens the data characteristics and improves the accuracy of machine learning classification.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below.
Fig. 1 is a schematic diagram of a model network of a depression identification method based on blood routine test data according to the present invention.
Fig. 2 is a machine learning process diagram of a depression identification method based on blood routine test data according to the present invention.
Fig. 3 is a model training flowchart of a depression identification method based on blood routine test data according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
A depression identification method based on blood routine test data comprises the following steps.
Obtaining blood routine test sample data from a physical examination hospital and a mental disease hospital and carrying out data preprocessing, wherein the test sample data is merged according to blood routine detection items; and filling up the dimension missing value of the test sample data. And classifying data types as training data aiming at the keywords in the diagnosis result by screening the sample.
In one embodiment of the invention, the test sample data is extracted from a Laboratory Information System (LIS) to obtain various test index data within three years. And classifying the test sample data according to the depression and anxiety related words in the diagnosis result, and after obtaining the test sample data, performing data preprocessing on the test sample data so that the test sample data meets the requirement of classifier training. The data preprocessing comprises the steps of carrying out mean filling of missing values and normal standardization on the test sample data.
Based on the test sample data, through a layer-by-layer learning strategy, the parameter initialization between the neural network layers is replaced by a self-coding learning mode of the RBM, namely (hidden layer) multiplicative bias and a weight connection matrix in the RBM are directly assigned to a weight matrix and bias of a corresponding layer, and the data characteristics of input information are enhanced.
In RBM, for a data setThe constructed energy expression is:. WhereinWIs an input layervTo the hidden layerhThe weights of (2) connect the matrices. It is composed ofThe weight connection matrix from the hidden layer to the output is converted,afor a multiplicative bias on the visible layer,bis a multiplicative bias on the hidden layer.
The feedback learning of neurons is trained by using the diagnostic label of the existing sample, the design stage of a supervised classifier is completed, a model capable of classifying the conventional blood test data characteristics is trained, and finally a diagnostic result model is obtained to realize the classification of depression tendency.
Referring to fig. 1-2, the process of training the model parameters is mainly divided into two steps.
(1) Unsupervised part of the training, each layer in the whole network is separate and unsupervised, and when feature vectors are mapped to different feature spaces, feature information is preserved as much as possible.
(2) With supervised part training, a feedforward (BP) neural network is constructed at the last layer to receive the feature vectors output by the RBMs. The final output is essentially a classifier function, and the process of training the classifier belongs to supervised learning.
The optimized objective function for the initialization of the level parameters isWhereintRefers to hidden layer ordinal numbers. The optimization work of the objective function is divided into two steps:
first, initialization between the input and the first hidden layer is performed. Let the input data bexAnd normalizing the data to obtain a visible layer:。
obtaining parameters using an optimized objective functionW *,a *,b *It is taken as a parameter matrix of the first level node, i.e.W 1=W *,b 1=b *(ii) a Next, initializing parameters from the first hidden layer to the second hidden layer: the output of the first hidden layer in the deep feedforward neural network is first obtained using the following formula:whereinThe data is normalized for the activation function on the first hidden layer to obtain:。
similarly, parameters are obtained by using the optimization objective function, and the obtained parameters are assigned toW 2,b 2(ii) a By doing so, the parameter initialization process can be completed after each layer of parameters in the feature learning stage are trained.
The second is fine tuning. And regarding the solution of the fine tuning stage, the method is consistent with the traditional deep feedforward neural network, and the end-to-end network overall fine tuning is carried out by utilizing a back propagation algorithm. The optimized objective function in the parameter initialization stage is solved by using a contrast divergence algorithm (an approximation algorithm), and the probability of the input sample under the distribution is maximized by constructing a log-likelihood function, so that the optimal parameter set is obtainedθ。
In summary, the weight training goal of the first stage is to make the mapping of the weight of each layer to the feature vector of the layer reach the optimal value. The second stage goal of training the weights is to minimize the error between the final prediction result and the actual value of the entire model. The first stage weight training highlights single-layer optimization, and the second stage weight training highlights overall optimization.
And in the second stage of weight training, the conventional blood attribute data is transmitted to a visible layer of the network, an output vector is obtained through the model trained in the first stage and is used as an input layer of the top layer BP, the power value is predicted, the power value is compared with an actual label, the difference value of the power value and the actual label is transmitted back to a bottom layer network structure of the model, and the weight of the whole model is finely adjusted.
Each layer of RBM network can only ensure that the weight in the layer where the RBM network is located is optimal for the feature vector mapping of the layer, and is not optimal for the feature vector mapping of the whole model. Therefore, the back propagation network also propagates error information from top to bottom to the unit nodes of each layer, and fine-tunes the model network.
By adopting the depression identification method based on the blood routine test data, the identification efficiency and the accuracy of the depression can be greatly improved. In the embodiment of the invention, more than 3000 test samples are adopted, and the accuracy rate of 0.93, the true yin rate of 0.95, the false positive rate of 0.04, the true positive rate of 0.89 and the false negative rate of 0.10 can be achieved through 22 indexes of the blood routine.
The above description is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can substitute or change the technical solution of the present invention within the technical scope of the present invention.
Claims (6)
1. A method for identifying depression based on blood routine test data, comprising the steps of:
step 1: obtaining blood routine test sample data from a physical examination hospital and a mental disease hospital and carrying out data preprocessing, wherein the test sample data are merged according to blood routine detection items, and data division is carried out according to diagnosis results;
step 2: reinforcing data characteristics of input information through a layer-by-layer learning strategy based on inspection sample data;
and step 3: the feedback learning of neurons is trained by using the diagnostic label of the existing sample, the design stage of a supervised classifier is completed, a model capable of classifying the conventional blood test data characteristics is trained, and finally a diagnostic result model is obtained to realize the classification of depression tendency.
2. A method of identifying depression based on blood routine test data as claimed in claim 1, wherein: the information used to predict input for depression disease is 22 indices of blood routine.
3. A method of identifying depression based on blood routine test data as claimed in claim 1, wherein: and filling up the dimension missing value of the test sample data.
4. And screening the samples, and taking keywords in the diagnosis result as a classification basis.
5. A method of identifying depression based on blood routine test data as claimed in claim 1, wherein: the parameter initialization between the neural network levels is replaced by a self-coding learning mode of the RBM, namely (hidden layer) multiplicative bias and a weight connection matrix in the RBM are directly assigned to the weight matrix and the bias of the corresponding level to be used as the parameter initialization of the neural network.
6. A method of identifying depression based on blood routine test data as claimed in claim 1, wherein: the positive data set and the negative data set are paired, and the data sets are randomly divided into a training set, a verification set and a test set according to the proportion of 80%, 10% and 10%.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110681222.4A CN113571176A (en) | 2021-06-19 | 2021-06-19 | Depression identification method based on blood routine test data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110681222.4A CN113571176A (en) | 2021-06-19 | 2021-06-19 | Depression identification method based on blood routine test data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113571176A true CN113571176A (en) | 2021-10-29 |
Family
ID=78162279
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110681222.4A Pending CN113571176A (en) | 2021-06-19 | 2021-06-19 | Depression identification method based on blood routine test data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113571176A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140257708A1 (en) * | 2008-03-04 | 2014-09-11 | Ridge Diagnostics, Inc. | Diagnosing and monitoring depression disorders |
CN111899869A (en) * | 2020-08-03 | 2020-11-06 | 东南大学 | Depression patient identification system and identification method thereof |
CN111899894A (en) * | 2020-08-03 | 2020-11-06 | 东南大学 | System and method for evaluating prognosis drug effect of depression patient |
CN112002412A (en) * | 2020-08-23 | 2020-11-27 | 吾征智能技术(北京)有限公司 | System, equipment and storage medium for deducing disease based on blood routine data |
CN112635069A (en) * | 2020-12-14 | 2021-04-09 | 内蒙古卫数数据科技有限公司 | Intelligent pulmonary tuberculosis identification method based on conventional test data |
-
2021
- 2021-06-19 CN CN202110681222.4A patent/CN113571176A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140257708A1 (en) * | 2008-03-04 | 2014-09-11 | Ridge Diagnostics, Inc. | Diagnosing and monitoring depression disorders |
CN111899869A (en) * | 2020-08-03 | 2020-11-06 | 东南大学 | Depression patient identification system and identification method thereof |
CN111899894A (en) * | 2020-08-03 | 2020-11-06 | 东南大学 | System and method for evaluating prognosis drug effect of depression patient |
CN112002412A (en) * | 2020-08-23 | 2020-11-27 | 吾征智能技术(北京)有限公司 | System, equipment and storage medium for deducing disease based on blood routine data |
CN112635069A (en) * | 2020-12-14 | 2021-04-09 | 内蒙古卫数数据科技有限公司 | Intelligent pulmonary tuberculosis identification method based on conventional test data |
Non-Patent Citations (1)
Title |
---|
郭银淼 等: "中性粒细胞与淋巴细胞比值和抑郁症严重程度的相关性", 《世界最新医学信息文摘》, vol. 18, no. 55, pages 9 - 11 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111292853B (en) | Multi-parameter-based cardiovascular disease risk prediction network model and construction method thereof | |
CN111524602B (en) | Old person's memory and cognitive function aassessment screening early warning system | |
WO2021120936A1 (en) | Chronic disease prediction system based on multi-task learning model | |
KR100794516B1 (en) | System and method for diagnosis and clinical test selection using case based machine learning inference | |
JP7381815B1 (en) | Passage anomaly detection system based on adaptive resampling deep encoder network | |
WO2022126800A1 (en) | Symptom, blood data and medical image fused method for comprehensively recognizing various suspected diseases | |
Jalali et al. | Advanced analytics for outcome prediction in intensive care units | |
Mahyoub et al. | Comparison analysis of machine learning algorithms to rank alzheimer’s disease risk factors by importance | |
CN112967803A (en) | Early mortality prediction method and system for emergency patients based on integrated model | |
CN113593708A (en) | Sepsis prognosis prediction method based on integrated learning algorithm | |
Adi et al. | Stroke risk prediction model using machine learning | |
CN115482924A (en) | Method and device for establishing autism spectrum disorder children dysnoesia diagnosis model | |
CN116705300A (en) | Medical decision assistance method, system and storage medium based on sign data analysis | |
CN116259415A (en) | Patient medicine taking compliance prediction method based on machine learning | |
Pillai | Cardiac disease prediction with tabular neural network | |
Sharma et al. | Model for predicting cardiac health using deep learning classifier | |
CN116564521A (en) | Chronic disease risk assessment model establishment method, medium and system | |
Siddiqa et al. | Robust Length of Stay Prediction Model for Indoor Patients. | |
CN113571176A (en) | Depression identification method based on blood routine test data | |
Oliullah et al. | Analyzing the effectiveness of several machine learning methods for heart attack prediction | |
Umut et al. | Prediction of sepsis disease by Artificial Neural Networks | |
Deepa et al. | Experimental evaluation of artificial intelligence assisted heart disease prediction using deep learning principle | |
Paliwal et al. | An efficient method for predicting heart disease problem using fitness value | |
TW202143248A (en) | System and method of biomedical data prediction risk including a data collection unit, a data processing unit and a judgment unit | |
CN117637154B (en) | Nerve internal department severe index prediction method and system based on optimization algorithm |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20211029 |
|
WD01 | Invention patent application deemed withdrawn after publication |