CN111738302A - System for classifying and diagnosing Alzheimer disease based on multi-modal data - Google Patents
System for classifying and diagnosing Alzheimer disease based on multi-modal data Download PDFInfo
- Publication number
- CN111738302A CN111738302A CN202010465444.8A CN202010465444A CN111738302A CN 111738302 A CN111738302 A CN 111738302A CN 202010465444 A CN202010465444 A CN 202010465444A CN 111738302 A CN111738302 A CN 111738302A
- Authority
- CN
- China
- Prior art keywords
- module
- image data
- data
- probability
- prediction
- 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.)
- Granted
Links
Images
Classifications
-
- 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
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- 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
-
- 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
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Public Health (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Evolutionary Biology (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Probability & Statistics with Applications (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a system for classifying and diagnosing Alzheimer disease based on multi-modal data, which comprises: the graphical user interface module is used for presenting an interface for program function calling to a user in a graphical mode; the 3D MRI image data preprocessing module is used for preprocessing a 3D MRI image of a patient and selecting a 2D slice from the 3D data; the image data classification prediction module is used for carrying out feature extraction on the 2D slices, carrying out classification prediction and giving the probability of three states (Alzheimer disease, mild cognitive impairment and normality) of the patient; the non-image data classification prediction module is used for performing classification prediction on the non-image data of the patient and giving the probability of three states of the patient; and the probability-based integration module is used for giving a final prediction result of the three state probabilities. The method carries out accurate classification prediction on the Alzheimer disease patient in a multi-modal (image and non-image) data situation, and assists medical workers in diagnosis and treatment.
Description
Technical Field
The invention relates to the technical field of medical diagnosis equipment for Alzheimer disease, in particular to a system for carrying out classification diagnosis on Alzheimer disease based on multi-modal data.
Background
Alzheimer's Disease (AD) is a neurodegenerative Disease commonly known as senile dementia. Mild Cognitive Impairment (MCI) is a state of Cognitive Impairment that is intermediate between normal aging and Mild AD, but does not meet the diagnostic criteria for AD.
At present, clinical diagnosis of early AD and AD-related MCI lacks reliable indexes, most patients belong to middle and late AD stages when the diagnosis is confirmed, and the treatment effect is poor. Therefore, the early diagnosis and early intervention of AD and AD-related MCI patients are particularly important.
Magnetic Resonance Imaging (MRI) is a neuroimaging method widely used for diagnosis of AD. The invention analyzes the MRI data and non-image physical examination data of MCI/AD patients to construct a classification diagnosis model based on a deep learning method, and assists doctors in analyzing and diagnosing the illness state. The research result of the subject can be applied to clinical diagnosis and intervention of AD and AD-related MCI patients, and can reduce huge burden for families and society of the patients.
The technical basis is that firstly, classification research on AD has precedent, and certain achievements are made on the extraction of AD characteristics to reduce data dimensionality and improve model accuracy, so that a basis is provided for researching diagnosis classification models of AD and MCI.
Deep learning is a branch of machine learning. It accomplishes a specific task on new data by training the model on a given data set. Compared with the traditional medical image identification method, the potential nonlinear relation in the medical image can be mined by deep learning, and the feature extraction rate is higher. From many experimental data, the results of deep learning are indeed more accurate than those of other methods. In recent years, deep learning is increasingly applied to medical image recognition, and provides important basis for further clinical application research.
A piece of cardiac mri AI analysis software using deep learning based cardiac mri as a core technology has been approved and approved by FDA510(k) and CE in europe. This suggests that the method of deep learning is feasible in clinical prediction and classification.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a system for classifying and diagnosing Alzheimer disease based on multi-modal data, which classifies and predicts Alzheimer disease patients under the condition of multi-modal (image and non-image) data and assists medical workers in diagnosis and treatment.
In order to achieve the purpose, the technical scheme provided by the invention is as follows: a system for the differential diagnosis of alzheimer's disease based on multimodal data comprising:
the graphical user interface module is used for presenting an interface for program function calling to a user in a graphical mode for the user to use;
the 3D MRI image data preprocessing module is used for preprocessing a 3D MRI image of a patient and selecting a 2D slice from the 3D data for processing by the subsequent image data classification and prediction module;
the image data classification prediction module is used for extracting the features of the 2D slices obtained by the 3D MRI image data preprocessing module, performing classification prediction by the extracted features, and giving the probability of three states of the patient, wherein the three states are Alzheimer disease AD, mild cognitive impairment MCI and normal NC respectively;
the non-image data classification prediction module is used for performing classification prediction on the non-image data of the patient and giving the probability of the three states of the patient; wherein the non-image data comprises a questionnaire, physical examination indicators;
and the probability-based integration module is used for integrating the probabilities of the three states given by the two classification prediction modules and giving a final prediction result of the probabilities of the three states.
Further, the graphical user interface module comprises a pre-training model selection module, a data set selection module, a model training parameter setting module, and a trained model selection module, wherein:
the pre-training model selection module selects pre-training model weights from equipment, and then performs model weight adjustment training on owned data;
the data set selection module selects a specific data set from equipment for model weight adjustment training;
the model training parameter setting module is used for setting parameters during model weight adjustment training, and the parameters comprise batch size and training round number;
and the trained model selection module selects a model after weight adjustment training from equipment, and performs classification prediction on data to be predicted.
Further, the 3D MRI image data preprocessing module includes a 3D MRI image data reading module, a 2D slice entropy calculation module, and an entropy value-based slice selection module, wherein:
the 3D MRI image data reading module causes the program to read a 3D MRI file in the format of nii suffix from the device;
the 2D slice entropy calculation module carries out entropy calculation on each slice of the read 3D MRI image, and the calculation formula is
Where E is the entropy of the slice, piIs the probability of a certain pixel in the slice, and replaces the probability with the frequency when calculating;
the slice selection module based on the entropy values sorts according to the entropy values of all slices, the entropy values can reflect the information quantity of a system in an information theory, the larger the entropy value is, the larger the information quantity is, therefore, the slice selection module finally selects a plurality of slices with the maximum entropy values, the default number is 32, and a user can set the number of the slices by himself.
Further, the image data classification prediction module comprises a neural network feature extraction module and a feature classification module, wherein:
the neural network feature extraction module performs feature extraction on the 2D image data so as to be classified by the feature classification module, and the neural network feature extraction module performs feature extraction by using a plurality of different convolutional neural network models, including DenseNet, inclusion V4 and VGG 16;
and the feature classification module receives features extracted by the neural network and performs probability calculation of the three states by using the fully-connected neural network and the Softmax function.
Further, the non-image data classification prediction module comprises a data cleaning module and a classification module, wherein:
the data cleaning module is used for cleaning non-image data and removing or completing data with field missing according to statistical data;
the classification module is used for classifying non-image data and performing probability calculation of three states by using an XGboost tool.
Further, the probability-based integration module receives results given by the image data classification prediction module and the non-image data classification prediction module, normalizes the results based on the prediction probabilities of the results, and carries out weighted summation according to the accuracy rates of the results when the results are trained to give a final prediction result; the normalization and weighted summation method comprises the following steps:
normalization: for three types of probability predicted value vectors p given by each classification prediction module1、p2、p3Selecting the maximum value p thereofmaxDividing each probability by the maximum value, wherein the operation is carried out on the probability prediction given by each classification prediction module;
weighted summation: after the three types of probability prediction value vectors are normalized, multiplying the normalized three types of probability prediction value vectors by the accuracy of the corresponding classification prediction module in the training process, and then carrying out vector addition;
adding all the vectors to obtain a final vector p1、p2、p3Each element of the vector is divided by (p)1+p2+p3) The final probabilistic prediction is obtained.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. the multi-modal data are used for classification prediction, data of a single mode can be one-sided, and data of different modes can achieve the effect of knowledge complementation.
2. The predicted values given by the classification prediction modules are integrated by using a probability-based method, and the accuracy in the training process of the corresponding model is taken as the weight, which is completely different from the simple voting of 'minority obeying majority' formula in most of the existing integration methods.
In summary, the system of the present invention utilizes image data and non-image data extraction information for classification, and the use of different modality data together can achieve the advantage of knowledge complementation. For 3D MRI images, the method does not directly use a neural network for processing, but firstly selects a partial slice with the largest information amount by calculating an entropy value and then uses the neural network for processing. In addition, for the prediction results given by a plurality of classifiers, the invention does not use a simple voting mode of 'minority obeying majority', but uses a probability-based mode to vote, and higher weight is given to a model with stronger confidence, thereby achieving the purpose of improving the self-confidence of prediction.
Drawings
FIG. 1 is a flow chart of a data path based system of the present invention.
Detailed Description
The present invention will be further described with reference to the following specific examples.
The system for performing a classification diagnosis of alzheimer's disease based on multimodal data provided in this embodiment is a classification diagnosis system of alzheimer's disease developed using Python language and running on a device equipped with a Windows system or a Linux system, and includes:
and the graphical user interface module is used for presenting an interface for program function calling to a user in a graphical mode, so that the user without programming basis can use the graphical user interface module more simply and conveniently.
And the 3D MRI image data preprocessing module is used for preprocessing the 3D MRI image of the patient and selecting a 2D slice from the 3D data so as to be processed by the subsequent image data classification and prediction module.
And the image data classification and prediction module is used for extracting the features of the 2D slices obtained by the 3D MRI image data preprocessing module, performing classification and prediction through the extracted features, and giving the probability of three states of the patient, wherein the three states are Alzheimer Disease (AD), Mild Cognitive Impairment (MCI) and Normal (NC) respectively.
And the non-image data classification prediction module is used for performing classification prediction on the non-image data (questionnaire and physical examination indexes) of the patient and giving the probability of three states on the patient.
And the probability-based integration module is used for integrating the probabilities of the three states given by the two classification prediction modules and giving a final prediction result of the probabilities of the three states.
The graphical user interface module comprises a pre-training model selection module, a data set selection module, a model training parameter setting module and a trained model selection module, wherein:
the pre-training model selection module selects pre-training model weights from equipment, and then model weight adjustment training can be performed on owned data;
the data set selection module selects a specific data set from equipment for model weight adjustment training;
the model training parameter setting module is used for setting parameters (such as batch size and training round number) during model weight adjustment training;
and the trained model selection module selects a model after weight adjustment training from equipment, and performs classification prediction on data to be predicted.
The 3D MRI image data preprocessing module comprises a 3D MRI image data reading module, a 2D slice entropy calculation module and an entropy value-based slice selection module, wherein:
the 3D MRI image data reading module causes the program to read a 3D MRI file in the format of nii suffix from the device;
the 2D slice entropy calculation module carries out entropy calculation on each slice of the read 3D MRI image, and the calculation formula is
Where E is the entropy of the slice, piIs the probability of a certain pixel in the slice (the probability is replaced by the frequency in the calculation);
the slice selection module based on the entropy value sorts according to the entropy value of each slice, the entropy value can reflect the information quantity of a system in an information theory, and the larger the entropy value is, the larger the information quantity is, so that the module finally selects a plurality of slices with the maximum entropy value (the default is 32 slices, and the number of the slices can be set by a user).
The image data classification prediction module comprises a neural network feature extraction module and a feature classification module, wherein:
the neural network feature extraction module performs feature extraction on 2D image data so as to be classified by the feature classification module, the module can perform feature extraction by using a plurality of different convolutional neural network models, and alternative convolutional neural network models are all models with excellent performance which are widely used in the field of computer vision, such as DenseNet, inclusion V4, VGG16 and the like;
and the feature classification module receives the features extracted by the neural network and performs probability calculation of the three states by using the fully-connected neural network and the Softmax function.
The non-image data classification prediction module comprises a data cleaning module and a classification module, wherein:
the data cleaning module is used for cleaning non-image data and removing or completing data with field missing according to statistical data;
the classification module is used for classifying non-image data and performing probability calculation of three states by using an XGboost tool.
The probability-based integration module receives results given by the image data classification prediction module and the non-image data classification prediction module, normalizes the results based on the prediction probabilities of the results and carries out weighted summation according to the accuracy rates of the results when the results are trained to give a final prediction result; the normalization and weighted summation method comprises the following steps:
normalization: three types of probability predicted values p given by each classification prediction module1、p2、p3(vector) selecting the maximum value p thereofmaxDividing each probability by the maximum value, wherein the operation is carried out on the probability prediction given by each classification prediction module;
weighted summation: after the three types of probability prediction value vectors are normalized, multiplying the normalized three types of probability prediction value vectors by the accuracy of the corresponding classification prediction module in the training process, and then carrying out vector addition;
adding all the vectors to obtain a final vector p1、p2、p3Each element of the vector is divided by (p)1+p2+p3) The final probabilistic prediction is obtained.
The following is a specific operation process of the above system of this embodiment, as shown in fig. 1, the flow is as follows:
when the user uses the program, the user can see the graphical user interface by opening the program of the graphical user interface module. The user can select a pre-trained model in a pre-training model selection module of the graphical user interface, set custom parameters in a model training parameter setting module, and then select a data set required to be used for training in a data set selection module, so that the training of the model can be started.
In the model training process, the 3D MRI image data preprocessing module takes out 3D MRI image data in the nii format from a data set catalog selected by a user and then transmits the 3D MRI image data to the 2D slice entropy calculation module; the 2D slice entropy calculation module can calculate the entropy according to a formula
And calculating the entropy value of each 2D slice in the 3D data, sequencing according to the entropy value after the calculation is finished, and selecting a plurality of slices with the maximum entropy by a slice selection module based on the entropy value (32 slices are acquiescent, and the number of slices can be customized by a user).
When the slice selection module selects a slice, the slice is stored for the subsequent image data classification prediction module to train. The image data classification prediction module reads and trains 2D slices from the slices stored in the slice selection module, namely, the parameters in the pre-training model selected by the user in the graphical user interface are updated, so that the model can better distinguish MRI 2D slice data. During the training process, the read 2D slice is transmitted to the neural network feature extraction module. After the feature extraction process of convolutional neural networks such as DenseNet, inclusion V4 and VGG16, the extracted features are transmitted to a feature classification module. And the feature classification module performs classification probability calculation of three states on the extracted features by using a fully-connected neural network and a Softmax function. And comparing the state with the original label according to the obtained probability of the three states, and if the probability of the three states is incorrect, calculating gradient updating model parameters.
In the graphical user interface, in addition to the user selecting a pre-training model and an image dataset catalog for training, the user may also select an XGboost tool and a non-image dataset for model training. After the user selects the model and the non-image data set, the data cleaning module will remove the data with the missing field first, and then the classification module will use the processed data for training.
After the model for image data classification and the model for non-image data classification are trained, the user can input unknown data for prediction. When a user has MRI image data of a patient and non-image data such as physical examination indexes, the image data can obtain a plurality of triples after passing through a 3D MRI image data preprocessing module and an image data classification prediction module, namely probability prediction values of three states given by different image data classification prediction modules; the non-image data can also obtain a triple, namely the probability predicted values of the three states, after passing through the non-image classification prediction module. And transmitting the obtained multiple triples to the probability-based integration module for processing. Finally, the user can diagnose the final probability prediction of the program as a reference index. Thus, the system accomplishes the task of assisting diagnosis.
The above-mentioned embodiments are merely preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, so that the changes in the shape and principle of the present invention should be covered within the protection scope of the present invention.
Claims (6)
1. A system for performing a differential diagnosis of alzheimer's disease based on multimodal data, comprising:
the graphical user interface module is used for presenting an interface for program function calling to a user in a graphical mode for the user to use;
the 3D MRI image data preprocessing module is used for preprocessing a 3D MRI image of a patient and selecting a 2D slice from the 3D data for processing by the subsequent image data classification and prediction module;
the image data classification prediction module is used for extracting the features of the 2D slices obtained by the 3D MRI image data preprocessing module, performing classification prediction by the extracted features, and giving the probability of three states of the patient, wherein the three states are Alzheimer disease AD, mild cognitive impairment MCI and normal NC respectively;
the non-image data classification prediction module is used for performing classification prediction on the non-image data of the patient and giving the probability of the three states of the patient; wherein the non-image data comprises a questionnaire, physical examination indicators;
and the probability-based integration module is used for integrating the probabilities of the three states given by the two classification prediction modules and giving a final prediction result of the probabilities of the three states.
2. The system of claim 1, wherein the system is configured to classify alzheimer's disease based on multimodal data, and wherein: the graphical user interface module comprises a pre-training model selection module, a data set selection module, a model training parameter setting module and a trained model selection module, wherein:
the pre-training model selection module selects pre-training model weights from equipment, and then performs model weight adjustment training on owned data;
the data set selection module selects a specific data set from equipment for model weight adjustment training;
the model training parameter setting module is used for setting parameters during model weight adjustment training, and the parameters comprise batch size and training round number;
and the trained model selection module selects a model after weight adjustment training from equipment, and performs classification prediction on data to be predicted.
3. The system of claim 1, wherein the system is configured to classify alzheimer's disease based on multimodal data, and wherein: the 3D MRI image data preprocessing module comprises a 3D MRI image data reading module, a 2D slice entropy calculation module and an entropy value-based slice selection module, wherein:
the 3D MRI image data reading module causes the program to read a 3D MRI file in the format of nii suffix from the device;
the 2D slice entropy calculation module carries out entropy calculation on each slice of the read 3D MRI image, and the calculation formula is
Where E is the entropy of the slice, piIs the probability of a certain pixel in the slice, and replaces the probability with the frequency when calculating;
the slice selection module based on the entropy values sorts according to the entropy values of all slices, the entropy values can reflect the information quantity of a system in an information theory, the larger the entropy value is, the larger the information quantity is, therefore, the slice selection module finally selects a plurality of slices with the maximum entropy values, the default number is 32, and a user can set the number of the slices by himself.
4. The system of claim 1, wherein the system is configured to classify alzheimer's disease based on multimodal data, and wherein: the image data classification prediction module comprises a neural network feature extraction module and a feature classification module, wherein:
the neural network feature extraction module performs feature extraction on the 2D image data so as to be classified by the feature classification module, and the neural network feature extraction module performs feature extraction by using a plurality of different convolutional neural network models, including DenseNet, inclusion V4 and VGG 16;
and the feature classification module receives features extracted by the neural network and performs probability calculation of the three states by using the fully-connected neural network and the Softmax function.
5. The system of claim 1, wherein the system is configured to classify alzheimer's disease based on multimodal data, and wherein: the non-image data classification prediction module comprises a data cleaning module and a classification module, wherein:
the data cleaning module is used for cleaning non-image data and removing or completing data with field missing according to statistical data;
the classification module is used for classifying non-image data and performing probability calculation of three states by using an XGboost tool.
6. The system of claim 1, wherein the system is configured to classify alzheimer's disease based on multimodal data, and wherein: the probability-based integration module receives results given by the image data classification prediction module and the non-image data classification prediction module, normalizes the results based on the prediction probabilities of the results and carries out weighted summation according to the accuracy rates of the results when the results are trained to give a final prediction result; the normalization and weighted summation method comprises the following steps:
normalization: for three types of probability predicted value vectors p given by each classification prediction module1、p2、p3Selecting the maximum value p thereofmaxDividing each probability by the maximum value, wherein the operation is carried out on the probability prediction given by each classification prediction module;
weighted summation: after the three types of probability prediction value vectors are normalized, multiplying the normalized three types of probability prediction value vectors by the accuracy of the corresponding classification prediction module in the training process, and then carrying out vector addition;
adding all the vectors to obtain a final vector p1、p2、p3Each element of the vector is divided by (p)1+p2+p3) The final probabilistic prediction is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010465444.8A CN111738302B (en) | 2020-05-28 | 2020-05-28 | System for classifying and diagnosing Alzheimer's disease based on multi-modal data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010465444.8A CN111738302B (en) | 2020-05-28 | 2020-05-28 | System for classifying and diagnosing Alzheimer's disease based on multi-modal data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111738302A true CN111738302A (en) | 2020-10-02 |
| CN111738302B CN111738302B (en) | 2023-06-20 |
Family
ID=72646475
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010465444.8A Active CN111738302B (en) | 2020-05-28 | 2020-05-28 | System for classifying and diagnosing Alzheimer's disease based on multi-modal data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111738302B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113205111A (en) * | 2021-04-07 | 2021-08-03 | 零氪智慧医疗科技(天津)有限公司 | Identification method and device suitable for liver tumor and electronic equipment |
| CN113538333A (en) * | 2021-06-08 | 2021-10-22 | 中南民族大学 | Alzheimer disease diagnosis method based on brain block feature weighted expression |
| CN113763343A (en) * | 2021-08-31 | 2021-12-07 | 同济大学 | Alzheimer's disease detection method based on deep learning and computer readable medium |
| CN114170162A (en) * | 2021-11-25 | 2022-03-11 | 深圳先进技术研究院 | Image prediction method, image prediction device and computer storage medium |
| CN116417135A (en) * | 2023-02-17 | 2023-07-11 | 中国人民解放军总医院第二医学中心 | Processing method and device for predicting early Alzheimer's disease type based on brain image |
| TWI811097B (en) * | 2021-09-09 | 2023-08-01 | 南韓商智聰醫治股份有限公司 | Method and apparatus for determining a degree of dementia of a user |
| CN116843377A (en) * | 2023-07-25 | 2023-10-03 | 河北鑫考科技股份有限公司 | Consumption behavior prediction method, device, equipment and medium based on big data |
| CN117496133A (en) * | 2024-01-03 | 2024-02-02 | 山东工商学院 | Closed bus R-CNN temperature fault monitoring method based on multi-mode data |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070253625A1 (en) * | 2006-04-28 | 2007-11-01 | Bbnt Solutions Llc | Method for building robust algorithms that classify objects using high-resolution radar signals |
| US20150141491A1 (en) * | 2012-07-11 | 2015-05-21 | The University Of Birmingham | Therapeutic Targets for Alzheimer's Disease |
| CN107067395A (en) * | 2017-04-26 | 2017-08-18 | 中国人民解放军总医院 | A kind of nuclear magnetic resonance image processing unit and method based on convolutional neural networks |
| CN107944473A (en) * | 2017-11-06 | 2018-04-20 | 南京邮电大学 | A kind of physiological signal emotion identification method based on the subjective and objective fusion of multi-categorizer |
| CN108257673A (en) * | 2018-01-12 | 2018-07-06 | 南通大学 | Risk value Forecasting Methodology and electronic equipment |
| CN109165692A (en) * | 2018-09-06 | 2019-01-08 | 中国矿业大学 | A kind of user's personality prediction meanss and method based on Weakly supervised study |
| CN109508653A (en) * | 2018-10-26 | 2019-03-22 | 南京邮电大学 | A kind of subjective and objective individual combat Emotion identification method merged based on EEG signals with psychology |
| CN109589092A (en) * | 2018-10-08 | 2019-04-09 | 广州市本真网络科技有限公司 | Method and system are determined based on the Alzheimer's disease of integrated study |
| WO2019169816A1 (en) * | 2018-03-09 | 2019-09-12 | 中山大学 | Deep neural network for fine recognition of vehicle attributes, and training method thereof |
| CN110236543A (en) * | 2019-05-23 | 2019-09-17 | 东华大学 | The more classification diagnosis systems of Alzheimer disease based on deep learning |
-
2020
- 2020-05-28 CN CN202010465444.8A patent/CN111738302B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070253625A1 (en) * | 2006-04-28 | 2007-11-01 | Bbnt Solutions Llc | Method for building robust algorithms that classify objects using high-resolution radar signals |
| US20150141491A1 (en) * | 2012-07-11 | 2015-05-21 | The University Of Birmingham | Therapeutic Targets for Alzheimer's Disease |
| CN107067395A (en) * | 2017-04-26 | 2017-08-18 | 中国人民解放军总医院 | A kind of nuclear magnetic resonance image processing unit and method based on convolutional neural networks |
| CN107944473A (en) * | 2017-11-06 | 2018-04-20 | 南京邮电大学 | A kind of physiological signal emotion identification method based on the subjective and objective fusion of multi-categorizer |
| CN108257673A (en) * | 2018-01-12 | 2018-07-06 | 南通大学 | Risk value Forecasting Methodology and electronic equipment |
| WO2019169816A1 (en) * | 2018-03-09 | 2019-09-12 | 中山大学 | Deep neural network for fine recognition of vehicle attributes, and training method thereof |
| CN109165692A (en) * | 2018-09-06 | 2019-01-08 | 中国矿业大学 | A kind of user's personality prediction meanss and method based on Weakly supervised study |
| CN109589092A (en) * | 2018-10-08 | 2019-04-09 | 广州市本真网络科技有限公司 | Method and system are determined based on the Alzheimer's disease of integrated study |
| CN109508653A (en) * | 2018-10-26 | 2019-03-22 | 南京邮电大学 | A kind of subjective and objective individual combat Emotion identification method merged based on EEG signals with psychology |
| CN110236543A (en) * | 2019-05-23 | 2019-09-17 | 东华大学 | The more classification diagnosis systems of Alzheimer disease based on deep learning |
Non-Patent Citations (2)
| Title |
|---|
| RACHNA JAIN 等: "Convolutional neural network based Alzheimer’s disease classification from magnetic resonance brain images" * |
| 吴庆园: "动脉硬化评价方法及其风险因素干预研究", 《CNKI硕士学位论文 医药卫生科技》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113205111A (en) * | 2021-04-07 | 2021-08-03 | 零氪智慧医疗科技(天津)有限公司 | Identification method and device suitable for liver tumor and electronic equipment |
| CN113538333A (en) * | 2021-06-08 | 2021-10-22 | 中南民族大学 | Alzheimer disease diagnosis method based on brain block feature weighted expression |
| CN113538333B (en) * | 2021-06-08 | 2022-06-17 | 中南民族大学 | Alzheimer disease diagnosis method based on brain block feature weighted expression |
| CN113763343A (en) * | 2021-08-31 | 2021-12-07 | 同济大学 | Alzheimer's disease detection method based on deep learning and computer readable medium |
| CN113763343B (en) * | 2021-08-31 | 2024-03-29 | 同济大学 | Deep learning-based Alzheimer's disease detection method and computer-readable medium |
| TWI811097B (en) * | 2021-09-09 | 2023-08-01 | 南韓商智聰醫治股份有限公司 | Method and apparatus for determining a degree of dementia of a user |
| CN114170162A (en) * | 2021-11-25 | 2022-03-11 | 深圳先进技术研究院 | Image prediction method, image prediction device and computer storage medium |
| CN116417135A (en) * | 2023-02-17 | 2023-07-11 | 中国人民解放军总医院第二医学中心 | Processing method and device for predicting early Alzheimer's disease type based on brain image |
| CN116417135B (en) * | 2023-02-17 | 2024-03-08 | 中国人民解放军总医院第二医学中心 | Processing method and device for predicting early Alzheimer's disease type based on brain image |
| CN116843377A (en) * | 2023-07-25 | 2023-10-03 | 河北鑫考科技股份有限公司 | Consumption behavior prediction method, device, equipment and medium based on big data |
| CN117496133A (en) * | 2024-01-03 | 2024-02-02 | 山东工商学院 | Closed bus R-CNN temperature fault monitoring method based on multi-mode data |
| CN117496133B (en) * | 2024-01-03 | 2024-03-22 | 山东工商学院 | Closed bus R-CNN temperature fault monitoring method based on multi-mode data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111738302B (en) | 2023-06-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111738302A (en) | System for classifying and diagnosing Alzheimer disease based on multi-modal data | |
| CN109886273B (en) | CMR image segmentation and classification system | |
| CN112418329B (en) | Cervical OCT image classification method and system based on multi-scale textural feature fusion | |
| CN107506797A (en) | One kind is based on deep neural network and multi-modal image alzheimer disease sorting technique | |
| CN109544518B (en) | Method and system applied to bone maturity assessment | |
| CN111009321A (en) | Application method of machine learning classification model in juvenile autism auxiliary diagnosis | |
| CN113314205A (en) | Efficient medical image labeling and learning system | |
| CN117218453B (en) | Incomplete multi-mode medical image learning method | |
| CN114549469A (en) | Deep neural network medical image diagnosis method based on confidence degree calibration | |
| Ebrahim et al. | Alzheimer disease early detection using convolutional neural networks | |
| CN113786185B (en) | Static brain network feature extraction method and system based on convolutional neural network | |
| CN114881105A (en) | Sleep staging method and system based on transformer model and contrast learning | |
| CN112837276A (en) | Brain glioma segmentation method based on cascaded deep neural network model | |
| CN116843667A (en) | Brain image classification and brain cognition scoring prediction method based on multitask learning | |
| CN107590806B (en) | Detection method and system based on brain medical imaging | |
| CN117352164A (en) | Multi-mode tumor detection and diagnosis platform based on artificial intelligence and processing method thereof | |
| CN117338234A (en) | Diopter and vision joint detection method | |
| CN117012370A (en) | Multi-mode disease auxiliary reasoning system, method, terminal and storage medium | |
| CN116564539A (en) | Medical similar case recommending method and system based on information extraction and entity normalization | |
| Tobias et al. | Android Application for Chest X-ray Health Classification From a CNN Deep Learning TensorFlow Model | |
| Nisha et al. | SGD-DABiLSTM based MRI Segmentation for Alzheimer’s disease Detection | |
| CN117765322B (en) | Classifying system capable of explaining deep learning based on multi-modal data | |
| CN114445643B (en) | Fmri brain function connection data feature extraction method based on significant sparse strong correlation | |
| CN116070120B (en) | Automatic identification method and system for multi-tag time sequence electrophysiological signals | |
| CN113995379B (en) | Sleep apnea-hypopnea syndrome evaluation method and device based on target detection framework |
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 |