CN112132203B - Fractional flow reserve measurement method and system based on intravascular ultrasound image - Google Patents
Fractional flow reserve measurement method and system based on intravascular ultrasound image Download PDFInfo
- Publication number
- CN112132203B CN112132203B CN202010986669.8A CN202010986669A CN112132203B CN 112132203 B CN112132203 B CN 112132203B CN 202010986669 A CN202010986669 A CN 202010986669A CN 112132203 B CN112132203 B CN 112132203B
- Authority
- CN
- China
- Prior art keywords
- intravascular ultrasound
- image
- flow reserve
- fractional flow
- ultrasound images
- 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
-
- 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
- G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
- G06F18/2193—Validation; Performance evaluation; Active pattern learning techniques based on specific statistical tests
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- 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
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Probability & Statistics with Applications (AREA)
- Multimedia (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The application discloses a fractional flow reserve measurement method and a fractional flow reserve measurement system based on intravascular ultrasound images, wherein the method comprises the following steps: acquiring a batch of intravascular ultrasound images and dividing the batch of intravascular ultrasound images into a training set and a verification set to obtain a training set image and a verification set image; optimizing a preset model based on the training set image and the verification set image to obtain a machine learning model; acquiring an intravascular ultrasound image to be detected and extracting image characteristics to obtain characteristic data; and inputting the characteristic data into a machine learning model to obtain the fractional flow reserve of the intravascular ultrasound image to be detected. The system comprises: the device comprises a dividing module, an optimizing module, a characteristic module and an output module. By using the application, the FFR value can be calculated according to the IVUS image, and the application has the characteristics of high accuracy and low calculation requirement. The fractional flow reserve measurement method and system based on intravascular ultrasound images can be widely applied to the field of medical image processing.
Description
Technical Field
The application belongs to the field of medical image processing, and particularly relates to a fractional flow reserve measurement method and system based on intravascular ultrasound images.
Background
The fractional flow reserve (Fractional flow reserve, FFR) is mainly the ratio of the downstream blood pressure of coronary artery stenosis to the upstream blood pressure of the stenosis, the smaller the ratio is, the heavier the influence degree of the stenosis on blood flow is reflected, the fractional flow reserve is an evaluation index which is effectively applied to clinical medical diagnosis at present, is an important means for quantitative and fixed-point evaluation diagnosis of coronary artery physiological conditions and functions at present, the prior art for obtaining FFR values mainly comprises two types of FFR calculation based on coronary angiography and FFR calculation based on OCT, and the two types of methods have high calculation requirements and cannot be applied to clinic well.
Disclosure of Invention
In order to solve the technical problems, the application aims to provide a fractional flow reserve measurement method and a fractional flow reserve measurement system based on intravascular ultrasound images, which realize that FFR values can be calculated according to IVUS images and have the characteristics of higher accuracy and low calculation requirements.
The first technical scheme adopted by the application is as follows: a fractional flow reserve measurement method based on intravascular ultrasound images, comprising the steps of:
acquiring a batch of intravascular ultrasound images and dividing the batch of intravascular ultrasound images into a training set and a verification set to obtain a training set image and a verification set image;
optimizing a preset model based on the training set image and the verification set image to obtain a machine learning model;
acquiring an intravascular ultrasound image to be detected and extracting image characteristics to obtain characteristic data;
and inputting the characteristic data into a machine learning model to obtain the fractional flow reserve of the intravascular ultrasound image to be detected.
Further, the step of optimizing the preset model based on the training set image and the verification set image to obtain a machine learning model specifically includes:
carrying out fractional flow reserve calculation and feature extraction processing on the training set image, and training a preset model by combining a deep learning algorithm to obtain a trained model;
and performing fractional flow reserve calculation and feature extraction processing on the training set image, and inputting the fractional flow reserve calculation and feature extraction processing into a trained model to obtain a machine learning model.
Further, the machine learning model is a network architecture of an automatic encoder type, and specifically includes two layers of LSTM units.
Further, the step of obtaining an intravascular ultrasound image to be detected and extracting image features to obtain feature data specifically includes:
acquiring an intravascular ultrasound image to be detected and intercepting a corresponding frame image of which the fractional flow reserve needs to be calculated;
and obtaining feature vectors according to the corresponding frame images, and constructing a feature vector set to obtain feature data.
Further, LSTM units of different layers use different frame images as input data.
Further, the characteristic data includes a location of blood flow, a vascular endothelial boundary width, a lumen radius, a medium and outer elastic membrane radius, an atherosclerosis level, an adventitia radius, and a collateral vessel distribution location.
The second technical scheme adopted by the application is as follows: a fractional flow reserve measurement system based on intravascular ultrasound images, comprising the following modules:
the dividing module is used for acquiring a batch of intravascular ultrasound images and dividing the batch of intravascular ultrasound images into a training set and a verification set to obtain a training set image and a verification set image;
the optimization module is used for optimizing the preset model based on the training set image and the verification set image to obtain a machine learning model;
the feature module is used for acquiring an intravascular ultrasound image to be detected and extracting image features to obtain feature data;
and the output module is used for inputting the characteristic data into the machine learning model to obtain the fractional flow reserve of the intravascular ultrasound image to be detected.
The method and the system have the beneficial effects that: the machine learning model is optimized through a large number of training sets and validation sets, and the fractional flow reserve is calculated by extracting eigenvalues from the IVUS image based on the optimized machine learning model and combining the eigenvalues with the neural network.
Drawings
FIG. 1 is a flow chart of the steps of a fractional flow reserve measurement method based on intravascular ultrasound images of the present application;
FIG. 2 is a block diagram of a fractional flow reserve measurement system based on intravascular ultrasound images in accordance with the present application;
FIG. 3 is a schematic diagram of the architecture of a first layer LSTM cell in accordance with an embodiment of the application;
fig. 4 is a schematic diagram of the architecture of a second layer LSTM cell according to an embodiment of the present application.
Detailed Description
The application will now be described in further detail with reference to the drawings and to specific examples. The step numbers in the following embodiments are set for convenience of illustration only, and the order between the steps is not limited in any way, and the execution order of the steps in the embodiments may be adaptively adjusted according to the understanding of those skilled in the art.
As shown in fig. 1, the present application provides a fractional flow reserve measurement method based on intravascular ultrasound images, the method comprising the steps of:
s1, acquiring a batch of intravascular ultrasound images and dividing the batch of intravascular ultrasound images into a training set and a verification set to obtain a training set image and a verification set image;
s2, optimizing a preset model based on the training set image and the verification set image to obtain a machine learning model;
s3, acquiring an intravascular ultrasound image to be detected and extracting image features to obtain feature data;
s4, inputting the characteristic data into a machine learning model to obtain the fractional flow reserve of the intravascular ultrasound image to be detected.
In particular, intravascular ultrasound (intravenous ultrasound, IVUS) refers to medical imaging techniques using special catheters with ultrasound probes attached at their ends, in combination with non-invasive ultrasound techniques and invasive catheter techniques.
Further as a preferred embodiment of the method, the step of optimizing the preset model based on the training set image and the verification set image to obtain a machine learning model specifically includes:
carrying out fractional flow reserve calculation and feature extraction processing on the training set image, and training a preset model by combining a deep learning algorithm to obtain a trained model;
and performing fractional flow reserve calculation and feature extraction processing on the training set image, and inputting the fractional flow reserve calculation and feature extraction processing into a trained model to obtain a machine learning model.
Further as a preferred embodiment of the method, the machine learning model is a network architecture of an automatic encoder type, and specifically includes two layers of LSTM units.
In particular, the LSTM unit specifically employs a ConvLSTM unit, and the standard LTSM exhibits excellent performance in many text and speech related tasks. To process spatiotemporal data, convLSTM was introduced. Here the fully connected gate interactions of the hidden states and inputs are replaced by convolution layers.
Further as a preferred embodiment of the method, the step of obtaining an intravascular ultrasound image to be measured and extracting image features to obtain feature data specifically includes:
acquiring an intravascular ultrasound image to be detected and intercepting a corresponding frame image of which the fractional flow reserve needs to be calculated;
and obtaining feature vectors according to the corresponding frame images, and constructing a feature vector set to obtain feature data.
Further as a preferred embodiment of the method, LSTM units with different layers use different frame images as input data.
Specifically, the control equation of the first layer LSTM cell is as follows,
at time m, the inputs to LSTM are three: input value x of network at present moment m Output value h of LSTM at last moment m-1 Cell state c at the previous time m-1 . While the main idea of LSTM is to learn long and short time information by controlling three switches, including 1) responsible for controlling the continued preservation of long-term state c; 2) Is responsible for controlling the input of the instant status to the long-term status c; 3) Negative poleIt is the responsibility to control whether the long-term state c is taken as the output of the current LSTM. Specifically, the three switches are controlled by three gates, the forget gate: it determines the cell state c at the previous time m-1 How much remains at the current time c m The method comprises the steps of carrying out a first treatment on the surface of the An input door: it determines the input x of the network at the current moment m How much is saved to cell state c m Output door: control unit state c m How much is output to the current output value h of LSTM m . The specific formula is described as follows:
where is the convolution operator,is a Hadamard product, subscript m represents a time step, superscript l represents a layer index, x e R d Is the input vector of LSTM unit extracted from I3D feature, f epsilon R h Is forgetting to gate activation, i.e. R h Is the activation of the input gate, o E R h Is the activation of the output gate c e R h Is a cell state vector, h.epsilon.R h Is a hidden state vector. W epsilon R h×d ，U∈R h×h ，b∈R h Is a weight matrix sum learned during trainingDeviation vector. The dimensions of the input vector and hidden state vector of LSTM are d and h, respectively, and the architecture of the specific first layer LSTM cell is referred to in fig. 3.
The control equation for the second layer LSTM cell is as follows:
wherein the subscript n represents the frame number, pc εR h Is a merged cell state vector, ph ε R h Is a merged hidden state vector. yc, yh, sc, sh are potential variables calculated within the merge function, and the architecture of the particular second tier LSTM unit is referred to in fig. 4.
In particular, it can be seen that the inputs of the first layer LSTM cells will be modified I3D functions, while the inputs of the higher layer LSTM cells will come from the module that computes the hidden state vector by merging the k LSTM cells of the previous layer with the LSTM of the LSTM and the last frame of the LSTM. Different layers capture information from different frame sizes. The higher level cell state is a function of the previous level cell state. Thus, the representation calculated at the last layer gives a compact representation of the complete image set. And further obtaining FFR values corresponding to the corresponding frame numbers of the images.
Further as a preferred embodiment of the method, the characteristic data includes a location of blood flow, a vascular endothelial boundary width, a lumen radius, a media and outer elastic membrane radius, a degree of atherosclerosis, an adventitia radius, and a collateral vessel distribution location.
Specifically, after IVUS detection is performed on the blood vessel, an image set is obtained, the number of frames corresponding to the local blood vessel length of the FFR value part to be calculated can be intercepted, and each feature vector is obtained from the image. Wherein the feature vectors are assembled, and the data comprise the position of blood flow, the width of vascular endothelial boundary, the radius of lumen, the radius of media and outer elastic membrane, the atherosclerosis degree, the radius of adventitia, the distribution position of collateral blood vessels and the like. Corresponding generated data x, f, c, g, W, U, b, etc.
As shown in fig. 2, a fractional flow reserve measurement system based on intravascular ultrasound images, comprising:
the dividing module is used for acquiring a batch of intravascular ultrasound images and dividing the batch of intravascular ultrasound images into a training set and a verification set to obtain a training set image and a verification set image;
the optimization module is used for optimizing the preset model based on the training set image and the verification set image to obtain a machine learning model;
the feature module is used for acquiring an intravascular ultrasound image to be detected and extracting image features to obtain feature data;
and the output module is used for inputting the characteristic data into the machine learning model to obtain the fractional flow reserve of the intravascular ultrasound image to be detected.
The content in the method embodiment is applicable to the system embodiment, the functions specifically realized by the system embodiment are the same as those of the method embodiment, and the achieved beneficial effects are the same as those of the method embodiment.
While the preferred embodiment of the present application has been described in detail, the application is not limited to the embodiment, and various equivalent modifications and substitutions can be made by those skilled in the art without departing from the spirit of the application, and these equivalent modifications and substitutions are intended to be included in the scope of the present application as defined in the appended claims.
Claims (5)
1. A fractional flow reserve measurement method based on intravascular ultrasound images, comprising the steps of:
acquiring a batch of intravascular ultrasound images and dividing the batch of intravascular ultrasound images into a training set and a verification set to obtain a training set image and a verification set image;
optimizing a preset model based on the training set image and the verification set image to obtain a machine learning model;
acquiring an intravascular ultrasound image to be detected and extracting image characteristics to obtain characteristic data;
inputting the characteristic data into a machine learning model to obtain the fractional flow reserve of the intravascular ultrasound image to be tested;
the machine learning model is a network architecture of an automatic encoder type and specifically comprises two layers of LSTM units;
different layers of LSTM units take different frame images as input data;
the control equation for the first layer LSTM cell is as follows:
wherein, is a convolution operator,is a Hadamard product, subscript m represents a time step, superscript l represents a layer index, x e R d Is an input vector of LSTM cells extracted from I3D features, f represents forgetting gate activation, o represents activation of output gates, c represents cell state vector, h represents hidden state vector, W, U and b are weight matrix and bias vector learned during training;
the control equation for the second layer LSTM cell is as follows:
where the subscript n denotes the number of frames, pc is the merged cell state vector, ph is the merged hidden state vector, yc, yh, sc, sh is the potential variable calculated within the merge function.
2. The method for measuring fractional flow reserve based on intravascular ultrasound images according to claim 1, wherein the step of optimizing the preset model based on the training set image and the verification set image to obtain a machine learning model specifically comprises the following steps:
carrying out fractional flow reserve calculation and feature extraction processing on the training set image, and training a preset model by combining a deep learning algorithm to obtain a trained model;
and performing fractional flow reserve calculation and feature extraction processing on the training set image, and inputting the fractional flow reserve calculation and feature extraction processing into a trained model to obtain a machine learning model.
3. The method for measuring fractional flow reserve based on intravascular ultrasound images according to claim 2, wherein the steps of acquiring intravascular ultrasound images to be measured and extracting image features to obtain feature data comprise:
acquiring an intravascular ultrasound image to be detected and intercepting a corresponding frame image of which the fractional flow reserve needs to be calculated;
and obtaining feature vectors according to the corresponding frame images, and constructing a feature vector set to obtain feature data.
4. A fractional flow reserve measurement method based on intravascular ultrasound images according to claim 3 wherein the characteristic data comprises blood flow location, vascular endothelial boundary width, lumen radius, medium and outer elastic membrane radius, atherosclerosis level, adventitia radius and collateral vessel distribution location.
5. A fractional flow reserve measurement system based on intravascular ultrasound images, comprising the following modules:
the dividing module is used for acquiring a batch of intravascular ultrasound images and dividing the batch of intravascular ultrasound images into a training set and a verification set to obtain a training set image and a verification set image;
the optimization module is used for optimizing the preset model based on the training set image and the verification set image to obtain a machine learning model;
the feature module is used for acquiring an intravascular ultrasound image to be detected and extracting image features to obtain feature data;
the output module is used for inputting the characteristic data into the machine learning model to obtain the fractional flow reserve of the intravascular ultrasound image to be detected;
the machine learning model is a network architecture of an automatic encoder type and specifically comprises two layers of LSTM units;
different layers of LSTM units take different frame images as input data;
the control equation for the first layer LSTM cell is as follows:
wherein, is a convolution operator,is a Hadamard product, subscript m represents a time step, superscript l represents a layer index, x e R d Is an input vector of an LSTM unit extracted from an I3D feature, f represents forgetting gate activation, o represents activation of an output gate, c represents a unit state vector, h represents a hidden state vectorW, U and b are weight matrices and bias vectors learned during training;
the control equation for the second layer LSTM cell is as follows:
where the subscript n denotes the number of frames, pc is the merged cell state vector, ph is the merged hidden state vector, yc, yh, sc, sh is the potential variable calculated within the merge function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010986669.8A CN112132203B (en) | 2020-09-18 | 2020-09-18 | Fractional flow reserve measurement method and system based on intravascular ultrasound image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010986669.8A CN112132203B (en) | 2020-09-18 | 2020-09-18 | Fractional flow reserve measurement method and system based on intravascular ultrasound image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112132203A CN112132203A (en) | 2020-12-25 |
| CN112132203B true CN112132203B (en) | 2023-09-29 |
Family
ID=73841442
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010986669.8A Active CN112132203B (en) | 2020-09-18 | 2020-09-18 | Fractional flow reserve measurement method and system based on intravascular ultrasound image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112132203B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014205231A1 (en) * | 2013-06-19 | 2014-12-24 | The Regents Of The University Of Michigan | Deep learning framework for generic object detection |
| CN108451540A (en) * | 2017-02-17 | 2018-08-28 | 深圳先进技术研究院 | A kind of blood flow reserve fraction measurement method and apparatus |
| CN110853029A (en) * | 2017-11-15 | 2020-02-28 | 深圳科亚医疗科技有限公司 | Method, system, and medium for automatically predicting blood flow characteristics based on medical images |
| CN110946619A (en) * | 2019-11-27 | 2020-04-03 | 杨靖 | Intravascular ultrasonic automatic imaging omics analysis system and analysis method |
| CN111627017A (en) * | 2020-05-29 | 2020-09-04 | 昆山戎影医疗科技有限公司 | Blood vessel lumen automatic segmentation method based on deep learning |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11514325B2 (en) * | 2018-03-21 | 2022-11-29 | The Regents Of The University Of California | Method and system for phase recovery and holographic image reconstruction using a neural network |
-
2020
- 2020-09-18 CN CN202010986669.8A patent/CN112132203B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014205231A1 (en) * | 2013-06-19 | 2014-12-24 | The Regents Of The University Of Michigan | Deep learning framework for generic object detection |
| CN108451540A (en) * | 2017-02-17 | 2018-08-28 | 深圳先进技术研究院 | A kind of blood flow reserve fraction measurement method and apparatus |
| CN110853029A (en) * | 2017-11-15 | 2020-02-28 | 深圳科亚医疗科技有限公司 | Method, system, and medium for automatically predicting blood flow characteristics based on medical images |
| CN110946619A (en) * | 2019-11-27 | 2020-04-03 | 杨靖 | Intravascular ultrasonic automatic imaging omics analysis system and analysis method |
| CN111627017A (en) * | 2020-05-29 | 2020-09-04 | 昆山戎影医疗科技有限公司 | Blood vessel lumen automatic segmentation method based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| Learning physical properties in complex visual scenes: An intelligent machine for perceiving blood flow dynamics from static CT angiography imaging;Zhifan Gao 等;《Neural Networks》;第82-93页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112132203A (en) | 2020-12-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Joshi et al. | Detection and prediction of diabetes mellitus using back-propagation neural network | |
| Mandic et al. | On the characterization of the deterministic/stochastic and linear/nonlinear nature of time series | |
| KR102304339B1 (en) | Method and computing device for providing analysis information on intravascular ultrasound images of blood vessels using artificial neural networks | |
| CN108509823A (en) | The detection method and device of QRS complex | |
| Kepp et al. | Segmentation of retinal low-cost optical coherence tomography images using deep learning | |
| CN112132203B (en) | Fractional flow reserve measurement method and system based on intravascular ultrasound image | |
| Liu et al. | Multi-scale ResNet and BiGRU automatic sleep staging based on attention mechanism | |
| CN111370120A (en) | Method for detecting diastolic dysfunction based on heart sound signals | |
| US20240006069A1 (en) | Medical Event Prediction Using a Personalized Dual-Channel Combiner Network | |
| Shanthi et al. | Prediction of blood glucose concentration ahead of time with feature based neural network | |
| Talal et al. | Machine learning‐based classification of multiple heart disorders from PCG signals | |
| CN113171059B (en) | Postoperative END risk early warning of multi-modal monitoring information and related equipment | |
| Ramachandran et al. | Classification of Electrocardiography Hybrid Convolutional Neural Network‐Long Short Term Memory with Fully Connected Layer | |
| Yang et al. | A deep learning method for intraoperative age-agnostic and disease-specific cardiac output monitoring from arterial blood pressure | |
| Sinha et al. | Automated detection of coronary artery disease comparing arterial fat accumulation using CNN | |
| Wang et al. | Neural network using for prediction spinal diseases | |
| CN115177260A (en) | Intelligent electrocardio-heart sound diagnosis method and device based on artificial neural network | |
| CN111297388B (en) | Fractional flow reserve measuring device | |
| Fakhry et al. | Comparison of window shapes and lengths in short-time feature extraction for classification of heart sound signals | |
| Yang et al. | AWD-stacking: An enhanced ensemble learning model for predicting glucose levels | |
| Esteves et al. | Learning a reconnecting regularization term for blood vessel variational segmentation | |
| Du et al. | Estimating residual stenosis for an arteriovenous shunt using a flexible fuzzy classifier | |
| Pahwa et al. | Utility of Machine learning in the Management of Normal Pressure Hydrocephalus (NPH): A Systematic Review | |
| US20240138773A1 (en) | Sequentially-reduced artificial intelligence based systems and methods for cardiovascular transfer functions | |
| CN113397511B (en) | Blood pressure measuring method and device |
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 |