CN116705307A - AI model-based heart function assessment method, system and storage medium for children - Google Patents
AI model-based heart function assessment method, system and storage medium for children Download PDFInfo
- Publication number
- CN116705307A CN116705307A CN202310981653.1A CN202310981653A CN116705307A CN 116705307 A CN116705307 A CN 116705307A CN 202310981653 A CN202310981653 A CN 202310981653A CN 116705307 A CN116705307 A CN 116705307A
- Authority
- CN
- China
- Prior art keywords
- data
- video data
- function
- heart
- right ventricle
- 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
- 230000004217 heart function Effects 0.000 title claims abstract description 106
- 238000000034 method Methods 0.000 title claims abstract description 45
- 210000005241 right ventricle Anatomy 0.000 claims description 129
- 230000006870 function Effects 0.000 claims description 69
- 230000008859 change Effects 0.000 claims description 60
- 230000002861 ventricular Effects 0.000 claims description 60
- 238000011156 evaluation Methods 0.000 claims description 58
- 238000001514 detection method Methods 0.000 claims description 48
- 238000007781 pre-processing Methods 0.000 claims description 29
- 238000003745 diagnosis Methods 0.000 claims description 27
- 230000002159 abnormal effect Effects 0.000 claims description 26
- 230000005856 abnormality Effects 0.000 claims description 24
- 230000007488 abnormal function Effects 0.000 claims description 22
- 238000002604 ultrasonography Methods 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 19
- 230000000747 cardiac effect Effects 0.000 claims description 17
- 238000004458 analytical method Methods 0.000 claims description 16
- 238000000926 separation method Methods 0.000 claims description 16
- 238000003709 image segmentation Methods 0.000 claims description 13
- 238000000605 extraction Methods 0.000 claims description 11
- 230000008602 contraction Effects 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 206010064911 Pulmonary arterial hypertension Diseases 0.000 claims description 6
- 208000008166 Right Ventricular Dysfunction Diseases 0.000 claims description 5
- 230000006814 right ventricular dysfunction Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims description 3
- 208000019622 heart disease Diseases 0.000 description 17
- 238000012549 training Methods 0.000 description 10
- 238000012795 verification Methods 0.000 description 8
- 230000011218 segmentation Effects 0.000 description 7
- 230000003205 diastolic effect Effects 0.000 description 6
- 238000001914 filtration Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 239000000284 extract Substances 0.000 description 4
- 238000010200 validation analysis Methods 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013145 classification model Methods 0.000 description 2
- 230000008828 contractile function Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000004064 dysfunction Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000004171 remote diagnosis Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 208000020446 Cardiac disease Diseases 0.000 description 1
- 208000002330 Congenital Heart Defects Diseases 0.000 description 1
- 206010016803 Fluid overload Diseases 0.000 description 1
- 206010049694 Left Ventricular Dysfunction Diseases 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 208000028831 congenital heart disease Diseases 0.000 description 1
- 210000005240 left ventricle Anatomy 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000001356 surgical procedure Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- 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)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The application discloses a children heart function assessment method, a system and a storage medium based on an AI model.
Description
Technical Field
The present application relates to the field of data processing and data transmission, and more particularly, to a method, system and storage medium for evaluating heart function of children based on an AI model.
Background
Assessment of pediatric cardiac function is an important link in diagnosis and treatment of pediatric cardiac disease. Currently, many infants and children suffer from congenital heart disease, whose heart structure or receiving surgical treatment results in long-term stress and/or volume overload of the right ventricle. As more children survive to adulthood, right ventricular dysfunction becomes a common factor leading to increased morbidity and mortality. Assessment of heart function in children involves quantitative and qualitative analysis of various parts of the heart to discover and intervene early in possible abnormalities. However, existing cardiac function assessment techniques are often directed to adults and focus on assessment of left ventricular function, right ventricular thin wall and morphology is complex, and right ventricular dysfunction often results in left ventricular dysfunction. Functional assessment of the right ventricle is more difficult relative to the left ventricle.
Accordingly, the present invention provides an innovative artificial intelligence model-based approach to the problems presented in the assessment of heart function in children. Compared with the prior art, the invention combines the deep learning algorithm and clinical data, can evaluate the heart function of children more accurately, reduces the dependence on manual analysis through automatic analysis, and realizes more effective and rapid heart function evaluation.
Disclosure of Invention
In view of the foregoing, it is an object of the present invention to provide an AI model-based heart function evaluation method, system and storage medium for children, which are capable of more effectively and more rapidly evaluating heart functions of children.
The first aspect of the invention provides a children heart function assessment method based on an AI model, which comprises the following steps:
acquiring heart ultrasonic video data;
preprocessing the heart ultrasonic video data to obtain preprocessed video data;
inputting the preprocessed video data into a preset children heart function evaluation network model for analysis, performing image segmentation on the preprocessed video data, and calculating according to the segmented image data to obtain the area change proportion of the right ventricle;
predicting the right ventricular function state to obtain a probability value of right ventricular function abnormality, and analyzing by combining the right ventricular area change proportion to obtain right ventricular function detection data;
Analyzing according to the right ventricle function detection data to generate a children heart function assessment report;
and sending the heart function evaluation report of the child to a preset terminal for display.
In this scheme, still include:
acquiring historical child heart ultrasound video data and related clinical data;
preprocessing the historical heart ultrasonic video data of the child to obtain sample data;
and analyzing according to the sample data and the related clinical data, and establishing a preset heart function evaluation network model of the children.
In this scheme, still include:
analyzing according to the heart ultrasonic video data, and judging whether the heart ultrasonic video data is in a preset format or not;
if not, carrying out format conversion on the heart ultrasonic video data, and converting the data format of the heart ultrasonic video data into a preset format; otherwise, no processing is performed.
In this scheme, the preprocessing the cardiac ultrasound video data to obtain preprocessed video data includes:
denoising the heart ultrasonic video data;
carrying out standardized processing on the denoised heart ultrasonic video data to obtain preprocessed video data;
Judging whether the video frame number of the preprocessed video data meets a preset requirement or not;
if not, inserting a blank frame into the preprocessed video data;
if yes, do not do any processing.
In this scheme, carry out the image segmentation to the preliminary treatment video data to calculate according to the image data after segmentation and obtain right ventricle area change proportion, include:
performing feature extraction on the preprocessed video data through a nested U-shaped structure network to obtain first feature data;
dividing the right ventricle area of each frame of image in the preprocessed video data according to the first characteristic data to obtain right ventricle image data;
analyzing according to the right ventricle image data to obtain a right ventricle area change curve;
extracting image data of a contraction frame and a relaxation frame according to the right ventricle area change curve;
and calculating according to the image data of the contraction frame and the relaxation frame to obtain the area change proportion of the right ventricle.
In this scheme, predict right ventricle functional state, obtain the probability value of right ventricle dysfunction, combine right ventricle area change proportion to carry out the analysis, obtain right ventricle function detection data, include:
Performing feature extraction on the preprocessed video data through a channel separation network to obtain second feature data;
comparing the second characteristic data with characteristic data of abnormal sample data in a database to obtain a probability value of right ventricle abnormal function;
judging whether the probability value of the right ventricle abnormal function is larger than a preset threshold value or not;
if yes, the right ventricle is abnormal in function; if not, the right ventricle functions normally;
analyzing according to the right ventricular image data with abnormal functions and the area change proportion of the right ventricle to obtain abnormal type data;
the abnormality type data includes pulmonary arterial hypertension and Fallotetraia.
In this aspect, the analyzing according to the right ventricular function detection data generates a pediatric cardiac function evaluation report, including:
analyzing according to the right ventricle function detection data, and generating a quantified detection result by combining clinical data;
sample data of the same detection result are screened from a database, and diagnosis suggestions are generated according to the treatment scheme of the sample data;
and integrating the quantified detection result and the diagnosis suggestion to generate a children heart function assessment report.
A second aspect of the present invention provides a system for assessing heart function in children based on an AI model, comprising:
The acquisition module is used for acquiring heart ultrasonic video data;
the preprocessing module is used for preprocessing the heart ultrasonic video data to obtain preprocessed video data;
the data analysis module is used for inputting the preprocessed video data into a preset children heart function evaluation network model for analysis, carrying out image segmentation on the preprocessed video data, and calculating according to the segmented image data to obtain the area change proportion of the right ventricle; predicting the right ventricular function state to obtain a probability value of right ventricular function abnormality, and analyzing by combining the right ventricular area change proportion to obtain right ventricular function detection data; analyzing according to the right ventricle function detection data to generate a children heart function assessment report;
and the data output module is used for sending the heart function evaluation report of the child to a preset terminal for display.
In this scheme, still include:
acquiring historical child heart ultrasound video data and related clinical data;
preprocessing the historical heart ultrasonic video data of the child to obtain sample data;
and analyzing according to the sample data and the related clinical data, and establishing a preset heart function evaluation network model of the children.
In this scheme, still include:
analyzing according to the heart ultrasonic video data, and judging whether the heart ultrasonic video data is in a preset format or not;
if not, carrying out format conversion on the heart ultrasonic video data, and converting the data format of the heart ultrasonic video data into a preset format; otherwise, no processing is performed.
In this scheme, the preprocessing the cardiac ultrasound video data to obtain preprocessed video data includes:
denoising the heart ultrasonic video data;
carrying out standardized processing on the denoised heart ultrasonic video data to obtain preprocessed video data;
judging whether the video frame number of the preprocessed video data meets a preset requirement or not;
if not, inserting a blank frame into the preprocessed video data;
if yes, do not do any processing.
In this scheme, carry out the image segmentation to the preliminary treatment video data to calculate according to the image data after segmentation and obtain right ventricle area change proportion, include:
performing feature extraction on the preprocessed video data through a nested U-shaped structure network to obtain first feature data;
dividing the right ventricle area of each frame of image in the preprocessed video data according to the first characteristic data to obtain right ventricle image data;
Analyzing according to the right ventricle image data to obtain a right ventricle area change curve;
extracting image data of a contraction frame and a relaxation frame according to the right ventricle area change curve;
and calculating according to the image data of the contraction frame and the relaxation frame to obtain the area change proportion of the right ventricle.
In this scheme, predict right ventricle functional state, obtain the probability value of right ventricle dysfunction, combine right ventricle area change proportion to carry out the analysis, obtain right ventricle function detection data, include:
performing feature extraction on the preprocessed video data through a channel separation network to obtain second feature data;
comparing the second characteristic data with characteristic data of abnormal sample data in a database to obtain a probability value of right ventricle abnormal function;
judging whether the probability value of the right ventricle abnormal function is larger than a preset threshold value or not;
if yes, the right ventricle is abnormal in function; if not, the right ventricle functions normally;
analyzing according to the right ventricular image data with abnormal functions and the area change proportion of the right ventricle to obtain abnormal type data;
the abnormality type data includes pulmonary arterial hypertension and Fallotetraia.
In this aspect, the analyzing according to the right ventricular function detection data generates a pediatric cardiac function evaluation report, including:
analyzing according to the right ventricle function detection data, and generating a quantified detection result by combining clinical data;
sample data of the same detection result are screened from a database, and diagnosis suggestions are generated according to the treatment scheme of the sample data;
and integrating the quantified detection result and the diagnosis suggestion to generate a children heart function assessment report.
A third aspect of the present invention provides a computer-readable storage medium having embodied therein an AI model-based pediatric cardiac function evaluation method program which, when executed by a processor, implements the steps of an AI model-based pediatric cardiac function evaluation method as described in any of the above.
The invention discloses a children heart function assessment method, a system and a storage medium based on an AI model.
Drawings
FIG. 1 shows a flow chart of a method for assessing heart function of a child based on an AI model of the application;
FIG. 2 is a flow chart showing a method for calculating the right ventricular area variation ratio according to the present application;
FIG. 3 is a flow chart of a method of acquiring right ventricular function test data according to the present application;
FIG. 4 shows a block diagram of a pediatric heart function assessment system based on an AI model of the application;
FIG. 5 is a schematic diagram showing the workflow of a pediatric heart function assessment system based on an AI model of the application;
FIG. 6 is a schematic diagram showing the segmentation effect of a preset pediatric cardiac function assessment network model of the application;
FIG. 7 is a schematic representation of DICE similarity coefficients of sample data according to the present application;
FIG. 8 shows a scatter plot of a right ventricular area variation ratio regression of the present application;
FIG. 9 shows a schematic representation of a subject's working profile of the present application;
FIG. 10 is a schematic diagram showing the effect of evaluating the performance of a model classification according to the present application.
Detailed Description
In order that the above-recited objects, features and advantages of the present application will be more clearly understood, a more particular description of the application will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, without conflict, the embodiments of the present application and features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.
FIG. 1 shows a flow chart of a method for assessing heart function of a child based on an AI model of the invention.
As shown in fig. 1, the invention discloses a children heart function assessment method based on an AI model, which comprises the following steps:
s102, acquiring heart ultrasonic video data;
s104, preprocessing the heart ultrasonic video data to obtain preprocessed video data;
s106, inputting the preprocessed video data into a preset children heart function evaluation network model for analysis, performing image segmentation on the preprocessed video data, and calculating according to the segmented image data to obtain a right ventricle area change proportion;
s108, predicting the right ventricular function state to obtain a probability value of right ventricular dysfunction, and analyzing by combining the right ventricular area change proportion to obtain right ventricular function detection data;
s110, analyzing according to the right ventricle function detection data to generate a children heart function assessment report;
And S112, sending the heart function evaluation report of the child to a preset terminal for display.
According to the embodiment of the invention, the heart ultrasonic video data acquired from the heart ultrasonic equipment is analyzed through the preset heart function evaluation network model of the child and the heart function evaluation report of the child is generated, and firstly, the acquired heart ultrasonic video data is subjected to format conversion and preprocessing, so that the method is suitable for detection data of heart ultrasonic equipment of different models and brands, and video data with different resolutions and frame rates can be processed. The heart ultrasonic video data are segmented and classified through a preset child heart function evaluation network model formed by two sub-networks of the nested U-shaped structure network and the channel separation convolution network, whether right ventricle function abnormality exists or not is judged, a specific abnormality type is determined, a heart function evaluation report is generated by combining clinical data, and the heart function evaluation report is sent to a preset terminal (such as a computer display terminal used by a doctor) to be displayed, so that a quantitative result and diagnosis suggestion are provided for the doctor.
The following advantages exist in detecting the heart function of children through the invention: (1) high accuracy and reliability: the AI model-based algorithm can learn from a large number of children's heart ultrasound videos and clinical data, and provides accurate right ventricular function assessment results, reducing human errors. (2) real-time assessment: the algorithm can rapidly process and analyze the heart ultrasonic image, realize the real-time evaluation of the heart function of children, provide immediate diagnosis results, and facilitate timely intervention and treatment. (3) non-invasive assessment: the algorithm evaluates based on the heart ultrasonic video and clinical data of the child, so that discomfort and pain of the traditional invasive examination are avoided, and comfort level of the child is improved. (4) aid in diagnosis and treatment decisions: the evaluation report generated by the algorithm provides quantitative results and diagnosis suggestions, provides important auxiliary information for doctors, helps the doctors to accurately judge the heart function conditions of children, and makes proper treatment schemes. (5) high efficiency and convenience: by means of automatic algorithm evaluation, the manual operation and time cost of doctors are reduced, and meanwhile, a more convenient and efficient evaluation method is provided, so that cardiac function evaluation can be applied to a wider medical environment. (6) innovative and prospective: the invention introduces the combination of the deep learning technology and the clinical data through the children heart function evaluation system based on the AI model, and provides a brand-new method for the evaluation of the children heart diseases. The method has wide application prospect and is helpful for promoting the development of the children heart disease diagnosis and treatment field.
The invention has the characteristics of simple operation, high efficiency and accuracy. The heart ultrasonic video data can be effectively processed, and the heart ultrasonic video is comprehensively analyzed and diagnosed by combining clinical data, so that powerful auxiliary decision-making tools are provided for medical professionals such as doctors and the like, and the diagnosis accuracy and classification accuracy of the heart diseases of children are improved. Meanwhile, the children heart function evaluation system based on the AI model has good expandability and adaptability, can be used for different medical institutions and clinical practice environments, and provides powerful support for early screening and treatment of heart diseases of children.
According to an embodiment of the present invention, further comprising:
acquiring historical child heart ultrasound video data and related clinical data;
preprocessing the historical heart ultrasonic video data of the child to obtain sample data;
and analyzing according to the sample data and the related clinical data, and establishing a preset heart function evaluation network model of the children.
It should be noted that, the historical heart ultrasonic video data of children can be obtained through channels of medical institutions, research institutions and the like. The collected video data encompasses samples of different ages, sexes, and case types to obtain a comprehensive and diverse training data set. Preprocessing the historical children heart ultrasonic video data to improve the accuracy of subsequent feature extraction.
The sample data is divided into a training set and a verification set, the characteristics of the training set and the verification set are extracted, the ventricular size, the heartbeat period, the diastolic frame, the systolic frame and the like are calculated, and the extracted characteristic data is marked and classified for training and verification of a model.
And establishing a neural network model, adopting a channel separation convolution network (Channel separated convolutions network), training the model by using labeling data, and carrying out disease classification and abnormality diagnosis on the heart function of the child by using learning characteristics to obtain a preset heart function evaluation network model of the child.
The pre-set pediatric cardiac functional assessment network model includes two sub-networks, a channel-separated convolutional network (Channel separated convolutions network) and a nested U-structure network (nestedU-structure network), respectively. The channel separation convolution network is a 3D convolution network dedicated to video classification for diagnosing and typing pediatric heart disease. The nested U-structure network is a network dedicated to image saliency segmentation, which is used to identify the right ventricular region of each frame in cardiac ultrasound video data, and for segmentation tasks, the number of downsampling must be reduced in view of the size of the input video, modifying the structure of the network, and embodying the number of encoders and decoders. In addition, to further optimize model performance, the learning rate, batch size, and number of iterations are tuned.
Then verifying and evaluating model performance of the preset network model for heart function evaluation of children in various ways, as shown in fig. 7, the frames of end diastole and end systole of the right ventricle and the overall average DICE similarity coefficient (Dice Similarity Coefficient); as shown in fig. 8, a scatter plot of the regression of the right ventricular area variation ratio (Fraction area change) shows that the mean absolute error (MeanAbsolute Error) and root mean square error (Root Mean Square Error) on the validation set were 4.07% and 5.09%, respectively; as shown in fig. 9, a ROC (ReceiverOperating Characteristic) curve for the case where normal and abnormal are predicted in the verification set; as shown in fig. 10, classification performance is assessed for the classification model on the validation set using accuracy, recall, F1 score.
According to an embodiment of the present invention, further comprising:
analyzing according to the heart ultrasonic video data, and judging whether the heart ultrasonic video data is in a preset format or not;
if not, carrying out format conversion on the heart ultrasonic video data, and converting the data format of the heart ultrasonic video data into a preset format; otherwise, no processing is performed.
It should be noted that, the data format of the input data of the preset pediatric cardiac function evaluation network model is a video format, such as an mp4 format, and is set to a preset format. Medical data acquired by the cardiac ultrasound device is usually DICOM format data, and cannot be directly input into a model for analysis, so that format conversion is required to be performed on the medical data, and the medical data is converted into video data in a system preset format.
According to an embodiment of the present invention, the preprocessing the cardiac ultrasound video data to obtain preprocessed video data includes:
denoising the heart ultrasonic video data;
carrying out standardized processing on the denoised heart ultrasonic video data to obtain preprocessed video data;
judging whether the video frame number of the preprocessed video data meets a preset requirement or not;
if not, inserting a blank frame into the preprocessed video data;
if yes, do not do any processing.
It should be noted that, the data preprocessing includes steps of removing noise in video, normalizing video images, and the like, firstly, denoising the heart ultrasonic video data by a noise reduction method preset by a system (such as a time domain filtering method, a frequency domain filtering method, a block filtering method, and the like), and then, performing size adjustment on the video data according to a preset size of the system to obtain normalized video data meeting the system requirements, namely, preprocessing the video data.
Because the video data of the input model needs a fixed frame number, as shown in fig. 5, the video data of the input model is 64 frames of image data, that is, the preset requirement is 64 frames, so that the video data with insufficient frame number is subjected to frame supplementing processing, and a new blank frame is inserted into the video data with insufficient frame number, so that the frame number reaches the preset requirement, and the extracted segment is ensured to contain a complete cardiac cycle, thereby improving the accuracy and stability of the subsequent analysis.
Fig. 2 shows a flowchart of a method for calculating the right ventricular area variation ratio according to the present invention.
As shown in fig. 2, according to an embodiment of the present invention, the image segmentation is performed on the preprocessed video data, and a right ventricle area change ratio is calculated according to the segmented image data, including:
s202, extracting features of the preprocessed video data through a nested U-shaped structure network to obtain first feature data;
s204, dividing the right ventricle area of each frame of image in the preprocessed video data according to the first characteristic data to obtain right ventricle image data;
s206, analyzing according to the right ventricle image data to obtain a right ventricle area change curve;
s208, extracting image data of a contraction frame and a relaxation frame according to the right ventricle area change curve;
and S210, calculating according to the image data of the contraction frame and the relaxation frame to obtain the area change proportion of the right ventricle.
It should be noted that, after the preprocessed video data is input into a preset network model for evaluating heart function of a child, as shown in fig. 5, first, an image segmentation technique is used to automatically extract the first image feature, that is, the right ventricle image feature, in the cardiac ultrasound video through a nested U-structure network (nested U-structure network), as shown in the left side of fig. 6, a single frame image selected from the cardiac ultrasound image is displayed, and visualization of segmenting the right ventricle by using a neural network is displayed, so that the right ventricle is segmented, and a right ventricle area change curve (area change curve) is output. The area change ratio of the right ventricle (Fraction Area Change) is then calculated by extracting the image data of the systolic and diastolic frames for assessing the right ventricular contractile function. As shown on the right side of fig. 6, a right ventricular area variation curve (Area change curve) of the split-network nested U-structure network output is shown. The peaks and troughs of the right ventricular area variation curve correspond to the systolic frame (s 1 … sn) and the diastolic frame (d 1 … dn), respectively, of each heartbeat. According to the calculation formula of fractional area change, the model calculates a fractional area change value once for each heartbeat period.
Fig. 3 is a flowchart showing a right ventricular function detection data acquisition method of the present invention.
As shown in fig. 3, according to an embodiment of the present invention, predicting the functional state of the right ventricle to obtain a probability value of abnormal function of the right ventricle, and analyzing in combination with the area change ratio of the right ventricle to obtain the functional detection data of the right ventricle includes:
s302, extracting features of the preprocessed video data through a channel separation network to obtain second feature data;
s304, comparing the second characteristic data with characteristic data of abnormal sample data in a database to obtain a probability value of right ventricle abnormal function;
s306, judging whether the probability value of the right ventricle abnormal function is larger than a preset threshold value;
s308, if yes, right ventricle function is abnormal; if not, the right ventricle functions normally;
s310, analyzing according to the right ventricle image data with abnormal functions and the area change proportion of the right ventricle to obtain abnormal type data;
the abnormality type data includes pulmonary arterial hypertension and Fallotetraia.
After the area change ratio of the right ventricle is obtained, the heart function of the child is evaluated through the channel separation convolution network (Channel separated convolutions network), as shown in fig. 5, and the heart disease of the child is automatically identified and classified through analyzing the heart structure and function information in the heart ultrasonic video data, including judging whether the right ventricle function abnormality and the specific abnormality type exist. After the preprocessing video data are input into the channel separation convolution network, the network extracts second characteristic data, such as shape characteristics, edge characteristics and the like, of the heart ultrasonic video data, and evaluates the right ventricle function in the heart ultrasonic video data by combining the characteristic data of the abnormal sample data in the database to output a probability value of the right ventricle function abnormality. And then determining whether the right ventricle is abnormal according to the probability value, and determining the abnormal type. The preset threshold value is obtained through analysis of the system according to the sample data, and is the lowest probability value of the right ventricle abnormal function.
According to an embodiment of the present invention, the analyzing according to the right ventricular function detection data generates a pediatric heart function evaluation report, including:
analyzing according to the right ventricle function detection data, and generating a quantified detection result by combining clinical data;
sample data of the same detection result are screened from a database, and diagnosis suggestions are generated according to the treatment scheme of the sample data;
and integrating the quantified detection result and the diagnosis suggestion to generate a children heart function assessment report.
It should be noted that, the system accurately diagnoses and classifies the heart disease of children according to the right ventricle function detection data output by the model and the related medical knowledge, provides corresponding treatment suggestions, and helps doctors to make accurate diagnosis and treatment decisions by combining clinical data according to different heart disease types. In addition, the system can be integrated with a hospital information system, so that data sharing and remote diagnosis are realized, and doctors and experts can conveniently cooperate and communicate comments.
FIG. 4 shows a block diagram of a pediatric heart function assessment system based on an AI model of the invention.
As shown in fig. 4, a second aspect of the present invention provides an AI model-based heart function assessment system for children, comprising:
The acquisition module is used for acquiring heart ultrasonic video data;
the preprocessing module is used for preprocessing the heart ultrasonic video data to obtain preprocessed video data;
the data analysis module is used for inputting the preprocessed video data into a preset children heart function evaluation network model for analysis, carrying out image segmentation on the preprocessed video data, and calculating according to the segmented image data to obtain the area change proportion of the right ventricle; predicting the right ventricular function state to obtain a probability value of right ventricular function abnormality, and analyzing by combining the right ventricular area change proportion to obtain right ventricular function detection data; analyzing according to the right ventricle function detection data to generate a children heart function assessment report;
and the data output module is used for sending the heart function evaluation report of the child to a preset terminal for display.
According to the embodiment of the invention, the heart ultrasonic video data acquired from the heart ultrasonic equipment is analyzed through the preset heart function evaluation network model of the child and the heart function evaluation report of the child is generated, and firstly, the acquired heart ultrasonic video data is subjected to format conversion and preprocessing, so that the method is suitable for detection data of heart ultrasonic equipment of different models and brands, and video data with different resolutions and frame rates can be processed. The heart ultrasonic video data are segmented and classified through a preset child heart function evaluation network model formed by two sub-networks of the nested U-shaped structure network and the channel separation convolution network, whether right ventricle function abnormality exists or not is judged, a specific abnormality type is determined, a heart function evaluation report is generated by combining clinical data, and the heart function evaluation report is sent to a preset terminal (such as a computer display terminal used by a doctor) to be displayed, so that a quantitative result and diagnosis suggestion are provided for the doctor.
The following advantages exist in detecting the heart function of children through the invention: (1) high accuracy and reliability: the AI model-based algorithm can learn from a large number of children's heart ultrasound videos and clinical data, and provides accurate right ventricular function assessment results, reducing human errors. (2) real-time assessment: the algorithm can rapidly process and analyze the heart ultrasonic image, realize the real-time evaluation of the heart function of children, provide immediate diagnosis results, and facilitate timely intervention and treatment. (3) non-invasive assessment: the algorithm evaluates based on the heart ultrasonic video and clinical data of the child, so that discomfort and pain of the traditional invasive examination are avoided, and comfort level of the child is improved. (4) aid in diagnosis and treatment decisions: the evaluation report generated by the algorithm provides quantitative results and diagnosis suggestions, provides important auxiliary information for doctors, helps the doctors to accurately judge the heart function conditions of children, and makes proper treatment schemes. (5) high efficiency and convenience: by means of automatic algorithm evaluation, the manual operation and time cost of doctors are reduced, and meanwhile, a more convenient and efficient evaluation method is provided, so that cardiac function evaluation can be applied to a wider medical environment. (6) innovative and prospective: the invention introduces the combination of the deep learning technology and the clinical data through the children heart function evaluation system based on the AI model, and provides a brand-new method for the evaluation of the children heart diseases. The method has wide application prospect and is helpful for promoting the development of the children heart disease diagnosis and treatment field.
The invention has the characteristics of simple operation, high efficiency and accuracy. The heart ultrasonic video data can be effectively processed, and the heart ultrasonic video is comprehensively analyzed and diagnosed by combining clinical data, so that powerful auxiliary decision-making tools are provided for medical professionals such as doctors and the like, and the diagnosis accuracy and classification accuracy of the heart diseases of children are improved. Meanwhile, the children heart function evaluation system based on the AI model has good expandability and adaptability, can be used for different medical institutions and clinical practice environments, and provides powerful support for early screening and treatment of heart diseases of children.
According to an embodiment of the present invention, further comprising:
acquiring historical child heart ultrasound video data and related clinical data;
preprocessing the historical heart ultrasonic video data of the child to obtain sample data;
and analyzing according to the sample data and the related clinical data, and establishing a preset heart function evaluation network model of the children.
It should be noted that, the historical heart ultrasonic video data of children can be obtained through channels of medical institutions, research institutions and the like. The collected video data encompasses samples of different ages, sexes, and case types to obtain a comprehensive and diverse training data set. Preprocessing the historical children heart ultrasonic video data to improve the accuracy of subsequent feature extraction.
The sample data is divided into a training set and a verification set, the characteristics of the training set and the verification set are extracted, the ventricular size, the heartbeat period, the diastolic frame, the systolic frame and the like are calculated, and the extracted characteristic data is marked and classified for training and verification of a model.
And establishing a neural network model, adopting a channel separation convolution network (Channel separated convolutions network), training the model by using labeling data, and carrying out disease classification and abnormality diagnosis on the heart function of the child by using learning characteristics to obtain a preset heart function evaluation network model of the child.
The pre-set pediatric cardiac functional assessment network model includes two sub-networks, a channel-separated convolutional network (Channel separated convolutions network) and a nested U-structure network (nestedU-structure network), respectively. The channel separation convolution network is a 3D convolution network dedicated to video classification for diagnosing and typing pediatric heart disease. The nested U-structure network is a network dedicated to image saliency segmentation, which is used to identify the right ventricular region of each frame in cardiac ultrasound video data, and for segmentation tasks, the number of downsampling must be reduced in view of the size of the input video, modifying the structure of the network, and embodying the number of encoders and decoders. In addition, to further optimize model performance, the learning rate, batch size, and number of iterations are tuned.
Then verifying and evaluating model performance of the preset network model for heart function evaluation of children in various ways, as shown in fig. 7, the frames of end diastole and end systole of the right ventricle and the overall average DICE similarity coefficient (Dice Similarity Coefficient); as shown in fig. 8, a scatter plot of the regression of the right ventricular area variation ratio (Fraction area change) shows that the mean absolute error (MeanAbsolute Error) and root mean square error (Root Mean Square Error) on the validation set were 4.07% and 5.09%, respectively; as shown in fig. 9, a ROC (ReceiverOperating Characteristic) curve for the case where normal and abnormal are predicted in the verification set; as shown in fig. 10, classification performance is assessed for the classification model on the validation set using accuracy, recall, F1 score.
According to an embodiment of the present invention, further comprising:
analyzing according to the heart ultrasonic video data, and judging whether the heart ultrasonic video data is in a preset format or not;
if not, carrying out format conversion on the heart ultrasonic video data, and converting the data format of the heart ultrasonic video data into a preset format; otherwise, no processing is performed.
It should be noted that, the data format of the input data of the preset pediatric cardiac function evaluation network model is a video format, such as an mp4 format, and is set to a preset format. Medical data acquired by the cardiac ultrasound device is usually DICOM format data, and cannot be directly input into a model for analysis, so that format conversion is required to be performed on the medical data, and the medical data is converted into video data in a system preset format.
According to an embodiment of the present invention, the preprocessing the cardiac ultrasound video data to obtain preprocessed video data includes:
denoising the heart ultrasonic video data;
carrying out standardized processing on the denoised heart ultrasonic video data to obtain preprocessed video data;
judging whether the video frame number of the preprocessed video data meets a preset requirement or not;
if not, inserting a blank frame into the preprocessed video data;
if yes, do not do any processing.
It should be noted that, the data preprocessing includes steps of removing noise in video, normalizing video images, and the like, firstly, denoising the heart ultrasonic video data by a noise reduction method preset by a system (such as a time domain filtering method, a frequency domain filtering method, a block filtering method, and the like), and then, performing size adjustment on the video data according to a preset size of the system to obtain normalized video data meeting the system requirements, namely, preprocessing the video data.
Because the video data of the input model needs a fixed frame number, as shown in fig. 5, the video data of the input model is 64 frames of image data, that is, the preset requirement is 64 frames, so that the video data with insufficient frame number is subjected to frame supplementing processing, and a new blank frame is inserted into the video data with insufficient frame number, so that the frame number reaches the preset requirement, and the extracted segment is ensured to contain a complete cardiac cycle, thereby improving the accuracy and stability of the subsequent analysis.
According to an embodiment of the present invention, the image segmentation is performed on the preprocessed video data, and the area change ratio of the right ventricle is calculated according to the segmented image data, including:
performing feature extraction on the preprocessed video data through a nested U-shaped structure network to obtain first feature data;
dividing the right ventricle area of each frame of image in the preprocessed video data according to the first characteristic data to obtain right ventricle image data;
analyzing according to the right ventricle image data to obtain a right ventricle area change curve;
extracting image data of a contraction frame and a relaxation frame according to the right ventricle area change curve;
and calculating according to the image data of the contraction frame and the relaxation frame to obtain the area change proportion of the right ventricle.
It should be noted that, after the preprocessed video data is input into a preset network model for evaluating heart function of a child, as shown in fig. 5, first, an image segmentation technique is used to automatically extract the first image feature, that is, the right ventricle image feature, in the cardiac ultrasound video through a nested U-structure network (nested U-structure network), as shown in the left side of fig. 6, a single frame image selected from the cardiac ultrasound image is displayed, and visualization of segmenting the right ventricle by using a neural network is displayed, so that the right ventricle is segmented, and a right ventricle area change curve (area change curve) is output. The area change ratio of the right ventricle (Fraction Area Change) is then calculated by extracting the image data of the systolic and diastolic frames for assessing the right ventricular contractile function. As shown on the right side of fig. 6, a right ventricular area variation curve (Area change curve) of the split-network nested U-structure network output is shown. The peaks and troughs of the right ventricular area variation curve correspond to the systolic frame (s 1 … sn) and the diastolic frame (d 1 … dn), respectively, of each heartbeat. According to the calculation formula of fractional area change, the model calculates a fractional area change value once for each heartbeat period.
According to an embodiment of the present invention, the predicting the functional state of the right ventricle to obtain the probability value of the abnormal function of the right ventricle, and analyzing in combination with the area change ratio of the right ventricle to obtain the functional detection data of the right ventricle includes:
performing feature extraction on the preprocessed video data through a channel separation network to obtain second feature data;
comparing the second characteristic data with characteristic data of abnormal sample data in a database to obtain a probability value of right ventricle abnormal function;
judging whether the probability value of the right ventricle abnormal function is larger than a preset threshold value or not;
if yes, the right ventricle is abnormal in function; if not, the right ventricle functions normally;
analyzing according to the right ventricular image data with abnormal functions and the area change proportion of the right ventricle to obtain abnormal type data;
the abnormality type data includes pulmonary arterial hypertension and Fallotetraia.
After the area change ratio of the right ventricle is obtained, the heart function of the child is evaluated through the channel separation convolution network (Channel separated convolutions network), as shown in fig. 5, and the heart disease of the child is automatically identified and classified through analyzing the heart structure and function information in the heart ultrasonic video data, including judging whether the right ventricle function abnormality and the specific abnormality type exist. After the preprocessing video data are input into the channel separation convolution network, the network extracts second characteristic data, such as shape characteristics, edge characteristics and the like, of the heart ultrasonic video data, and evaluates the right ventricle function in the heart ultrasonic video data by combining the characteristic data of the abnormal sample data in the database to output a probability value of the right ventricle function abnormality. And then determining whether the right ventricle is abnormal according to the probability value, and determining the abnormal type. The preset threshold value is obtained through analysis of the system according to the sample data, and is the lowest probability value of the right ventricle abnormal function.
According to an embodiment of the present invention, the analyzing according to the right ventricular function detection data generates a pediatric heart function evaluation report, including:
analyzing according to the right ventricle function detection data, and generating a quantified detection result by combining clinical data;
sample data of the same detection result are screened from a database, and diagnosis suggestions are generated according to the treatment scheme of the sample data;
and integrating the quantified detection result and the diagnosis suggestion to generate a children heart function assessment report.
It should be noted that, the system accurately diagnoses and classifies the heart disease of children according to the right ventricle function detection data output by the model and the related medical knowledge, provides corresponding treatment suggestions, and helps doctors to make accurate diagnosis and treatment decisions by combining clinical data according to different heart disease types. In addition, the system can be integrated with a hospital information system, so that data sharing and remote diagnosis are realized, and doctors and experts can conveniently cooperate and communicate comments.
A third aspect of the present invention provides a computer-readable storage medium having embodied therein an AI model-based pediatric cardiac function evaluation method program which, when executed by a processor, implements the steps of an AI model-based pediatric cardiac function evaluation method as described in any of the above.
The application discloses a children heart function assessment method, a system and a storage medium based on an AI model.
In the several embodiments provided by the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above described device embodiments are only illustrative, e.g. the division of the units is only one logical function division, and there may be other divisions in practice, such as: multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. In addition, the various components shown or discussed may be coupled or directly coupled or communicatively coupled to each other via some interface, whether indirectly coupled or communicatively coupled to devices or units, whether electrically, mechanically, or otherwise.
The units described above as separate components may or may not be physically separate, and components shown as units may or may not be physical units; can be located in one place or distributed to a plurality of network units; some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present invention may be integrated in one processing unit, or each unit may be separately used as one unit, or two or more units may be integrated in one unit; the integrated units may be implemented in hardware or in hardware plus software functional units.
Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above method embodiments may be implemented by hardware related to program instructions, and the foregoing program may be stored in a computer readable storage medium, where the program, when executed, performs steps including the above method embodiments; and the aforementioned storage medium includes: a mobile storage device, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk or an optical disk, or the like, which can store program codes.
Alternatively, the above-described integrated units of the present invention may be stored in a computer-readable storage medium if implemented in the form of software functional modules and sold or used as separate products. Based on such understanding, the technical solutions of the embodiments of the present invention may be embodied in essence or a part contributing to the prior art in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, ROM, RAM, magnetic or optical disk, or other medium capable of storing program code.
Claims (10)
1. A method for evaluating heart function of a child based on an AI model, comprising:
acquiring heart ultrasonic video data;
preprocessing the heart ultrasonic video data to obtain preprocessed video data;
inputting the preprocessed video data into a preset children heart function evaluation network model for analysis, performing image segmentation on the preprocessed video data, and calculating according to the segmented image data to obtain the area change proportion of the right ventricle;
Predicting the right ventricular function state to obtain a probability value of right ventricular function abnormality, and analyzing by combining the right ventricular area change proportion to obtain right ventricular function detection data;
analyzing according to the right ventricle function detection data to generate a children heart function assessment report;
and sending the heart function evaluation report of the child to a preset terminal for display.
2. The AI model-based method of assessing cardiac function of a child according to claim 1, further comprising:
acquiring historical child heart ultrasound video data and related clinical data;
preprocessing the historical heart ultrasonic video data of the child to obtain sample data;
and analyzing according to the sample data and the related clinical data, and establishing a preset heart function evaluation network model of the children.
3. The AI model-based method of assessing cardiac function of a child according to claim 1, further comprising:
analyzing according to the heart ultrasonic video data, and judging whether the heart ultrasonic video data is in a preset format or not;
if not, carrying out format conversion on the heart ultrasonic video data, and converting the data format of the heart ultrasonic video data into a preset format; otherwise, no processing is performed.
4. The AI model-based cardiac function assessment method of claim 1, wherein the preprocessing of the cardiac ultrasound video data to obtain preprocessed video data comprises:
denoising the heart ultrasonic video data;
carrying out standardized processing on the denoised heart ultrasonic video data to obtain preprocessed video data;
judging whether the video frame number of the preprocessed video data meets a preset requirement or not;
if not, inserting a blank frame into the preprocessed video data;
if yes, do not do any processing.
5. The AI model-based cardiac function assessment method of claim 1, wherein image segmentation of the preprocessed video data and calculation of the right ventricle area change ratio from the segmented image data comprises:
performing feature extraction on the preprocessed video data through a nested U-shaped structure network to obtain first feature data;
dividing the right ventricle area of each frame of image in the preprocessed video data according to the first characteristic data to obtain right ventricle image data;
analyzing according to the right ventricle image data to obtain a right ventricle area change curve;
Extracting image data of a contraction frame and a relaxation frame according to the right ventricle area change curve;
and calculating according to the image data of the contraction frame and the relaxation frame to obtain the area change proportion of the right ventricle.
6. The AI model-based cardiac function assessment method of claim 1, wherein predicting the right ventricular function state to obtain a probability value of a right ventricular dysfunction, and analyzing in combination with the right ventricular area variation ratio to obtain right ventricular function detection data comprises:
performing feature extraction on the preprocessed video data through a channel separation network to obtain second feature data;
comparing the second characteristic data with characteristic data of abnormal sample data in a database to obtain a probability value of right ventricle abnormal function;
judging whether the probability value of the right ventricle abnormal function is larger than a preset threshold value or not;
if yes, the right ventricle is abnormal in function; if not, the right ventricle functions normally;
analyzing according to the right ventricular image data with abnormal functions and the area change proportion of the right ventricle to obtain abnormal type data;
the abnormality type data includes pulmonary arterial hypertension and Fallotetraia.
7. The AI model-based pediatric cardiac function assessment method of claim 1, wherein the analyzing based on the right ventricular function detection data to generate a pediatric cardiac function assessment report comprises:
analyzing according to the right ventricle function detection data, and generating a quantified detection result by combining clinical data;
sample data of the same detection result are screened from a database, and diagnosis suggestions are generated according to the treatment scheme of the sample data;
and integrating the quantified detection result and the diagnosis suggestion to generate a children heart function assessment report.
8. An AI model-based pediatric cardiac function assessment system, comprising:
the acquisition module is used for acquiring heart ultrasonic video data;
the preprocessing module is used for preprocessing the heart ultrasonic video data to obtain preprocessed video data;
the data analysis module is used for inputting the preprocessed video data into a preset children heart function evaluation network model for analysis, carrying out image segmentation on the preprocessed video data, and calculating according to the segmented image data to obtain the area change proportion of the right ventricle; predicting the right ventricular function state to obtain a probability value of right ventricular function abnormality, and analyzing by combining the right ventricular area change proportion to obtain right ventricular function detection data; analyzing according to the right ventricle function detection data to generate a children heart function assessment report;
And the data output module is used for sending the heart function evaluation report of the child to a preset terminal for display.
9. The AI model-based pediatric cardiac function evaluation system of claim 8, wherein predicting the right ventricular function state to obtain a probability value for a right ventricular dysfunction, analyzing in combination with the right ventricular area variation ratio to obtain right ventricular function detection data comprises:
performing feature extraction on the preprocessed video data through a channel separation network to obtain second feature data;
comparing the second characteristic data with characteristic data of abnormal sample data in a database to obtain a probability value of right ventricle abnormal function;
judging whether the probability value of the right ventricle abnormal function is larger than a preset threshold value or not;
if yes, the right ventricle is abnormal in function; if not, the right ventricle functions normally;
analyzing according to the right ventricular image data with abnormal functions and the area change proportion of the right ventricle to obtain abnormal type data;
the abnormality type data includes pulmonary arterial hypertension and Fallotetraia.
10. A computer-readable storage medium, characterized in that it comprises an AI model-based children's heart function assessment method program, which, when executed by a processor, implements the steps of an AI model-based children's heart function assessment method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310981653.1A CN116705307A (en) | 2023-08-07 | 2023-08-07 | AI model-based heart function assessment method, system and storage medium for children |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310981653.1A CN116705307A (en) | 2023-08-07 | 2023-08-07 | AI model-based heart function assessment method, system and storage medium for children |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116705307A true CN116705307A (en) | 2023-09-05 |
Family
ID=87831542
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310981653.1A Pending CN116705307A (en) | 2023-08-07 | 2023-08-07 | AI model-based heart function assessment method, system and storage medium for children |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116705307A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108882917A (en) * | 2016-05-30 | 2018-11-23 | 深圳迈瑞生物医疗电子股份有限公司 | A kind of heart volume discriminance analysis system and method |
| TW202022713A (en) * | 2018-12-05 | 2020-06-16 | 宏碁股份有限公司 | Method and system for evaluating cardiac status, electronic device and ultrasonic scanning device |
| CN114821070A (en) * | 2022-06-02 | 2022-07-29 | 哈尔滨理工大学 | Heart MRI segmentation method based on improved U-Net type network |
| CN114898882A (en) * | 2022-06-21 | 2022-08-12 | 四川大学华西医院 | Method and system for ultrasound-based assessment of right heart function |
| CN115457653A (en) * | 2022-08-29 | 2022-12-09 | 北京邮电大学 | Low-sampling WiFi perception enhancement model training method, enhancement method, action recognition method and device |
| CN115830711A (en) * | 2022-11-28 | 2023-03-21 | 中南大学 | Sign language vocabulary recognition method, system, device and medium based on deep learning |
| CN116269496A (en) * | 2023-03-01 | 2023-06-23 | 南京大学 | Heart three-dimensional ultrasonic imaging and heart function evaluation system based on implicit neural representation |
| CN116341710A (en) * | 2023-02-03 | 2023-06-27 | 中国地质大学（武汉） | Traffic accident prediction method and device, electronic equipment and storage medium |
| CN116452579A (en) * | 2023-06-01 | 2023-07-18 | 中国医学科学院阜外医院 | Chest radiography image-based pulmonary artery high pressure intelligent assessment method and system |
-
2023
- 2023-08-07 CN CN202310981653.1A patent/CN116705307A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108882917A (en) * | 2016-05-30 | 2018-11-23 | 深圳迈瑞生物医疗电子股份有限公司 | A kind of heart volume discriminance analysis system and method |
| TW202022713A (en) * | 2018-12-05 | 2020-06-16 | 宏碁股份有限公司 | Method and system for evaluating cardiac status, electronic device and ultrasonic scanning device |
| CN114821070A (en) * | 2022-06-02 | 2022-07-29 | 哈尔滨理工大学 | Heart MRI segmentation method based on improved U-Net type network |
| CN114898882A (en) * | 2022-06-21 | 2022-08-12 | 四川大学华西医院 | Method and system for ultrasound-based assessment of right heart function |
| CN115457653A (en) * | 2022-08-29 | 2022-12-09 | 北京邮电大学 | Low-sampling WiFi perception enhancement model training method, enhancement method, action recognition method and device |
| CN115830711A (en) * | 2022-11-28 | 2023-03-21 | 中南大学 | Sign language vocabulary recognition method, system, device and medium based on deep learning |
| CN116341710A (en) * | 2023-02-03 | 2023-06-27 | 中国地质大学（武汉） | Traffic accident prediction method and device, electronic equipment and storage medium |
| CN116269496A (en) * | 2023-03-01 | 2023-06-23 | 南京大学 | Heart three-dimensional ultrasonic imaging and heart function evaluation system based on implicit neural representation |
| CN116452579A (en) * | 2023-06-01 | 2023-07-18 | 中国医学科学院阜外医院 | Chest radiography image-based pulmonary artery high pressure intelligent assessment method and system |
Non-Patent Citations (2)
| Title |
|---|
| 李响楠 等: "超声心动图在法洛四联症根治术后右心室功能评估中的作用", 心血管外科杂志, vol. 5, no. 4, pages 14 - 16 * |
| 王丽娜 等: "肺动脉收缩压、血浆N末端脑钠肽前体水平与肺动脉高压患者右心室面积变化分数的相关性研究", 实用心脑肺血管病杂志, vol. 26, no. 2, pages 131 - 133 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10869610B2 (en) | System and method for identifying cardiac arrhythmias with deep neural networks | |
| US20230131876A1 (en) | Systems and methods of identity analysis of electrocardiograms | |
| KR102125127B1 (en) | Method of brain disorder diagnosis via deep learning | |
| WO2019100562A1 (en) | Automatic recognition and classification method for electrocardiogram heartbeat based on artificial intelligence | |
| US20230005251A1 (en) | Diagnostic assistance apparatus and model generation apparatus | |
| US20220093215A1 (en) | Discovering genomes to use in machine learning techniques | |
| US10327709B2 (en) | System and methods to predict serum lactate level | |
| CN113855038B (en) | Electrocardiosignal critical value prediction method and device based on multi-model integration | |
| CN110693488A (en) | Electrocardiosignal processing system, electrocardiosignal processing method, electronic equipment and readable storage medium | |
| KR102444274B1 (en) | Apparatus and method for precise analysis of arthritis severity | |
| CN111370124A (en) | Health analysis system and method based on facial recognition and big data | |
| US20230100365A1 (en) | Automatic classification of healthy and disease conditions from images or digital standard 12-lead ECGs | |
| CN117831771B (en) | Disease risk prediction model construction method and system based on deep learning | |
| CN114926396A (en) | Mental disorder magnetic resonance image preliminary screening model construction method | |
| CN118312831A (en) | Multi-mode heart data processing optimization method and device and computer equipment | |
| US20210298625A1 (en) | System and method for detecting and predicting an occurrence of cardiac events from electrocardiograms | |
| CN116524248B (en) | Medical data processing device, method and classification model training device | |
| CN115547502B (en) | Hemodialysis patient risk prediction device based on time sequence data | |
| CN116705307A (en) | AI model-based heart function assessment method, system and storage medium for children | |
| CN116135146A (en) | Dynamic electrocardiograph analysis method and device and medical equipment | |
| CN112614091A (en) | Ultrasonic multi-section data detection method for congenital heart disease | |
| CN111297348A (en) | Method for generating electrocardiogram analysis result, electrocardiograph and storage medium | |
| RU2822304C1 (en) | Method of generating data file for ecg analysis | |
| EP3926540A1 (en) | Method for filtering normal medical image, method for interpreting medical image, and computing device implementing the methods | |
| Geetha et al. | Analysing Osteoporosis Detection: A Comparative Study of CNN and FNN |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230905 |