CN115579109A - Electrocardiogram image analysis method and device in medical environment and terminal equipment - Google Patents

Electrocardiogram image analysis method and device in medical environment and terminal equipment Download PDF

Info

Publication number
CN115579109A
CN115579109A CN202211482855.3A CN202211482855A CN115579109A CN 115579109 A CN115579109 A CN 115579109A CN 202211482855 A CN202211482855 A CN 202211482855A CN 115579109 A CN115579109 A CN 115579109A
Authority
CN
China
Prior art keywords
image
electrocardiogram
analyzed
analysis
heart disease
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
Application number
CN202211482855.3A
Other languages
Chinese (zh)
Inventor
章德云
洪申达
耿世佳
魏国栋
王凯
陈星月
傅兆吉
周荣博
俞杰
徐伟伦
鄂雁祺
齐新宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hefei Xinzhisheng Health Technology Co ltd
Original Assignee
Hefei Xinzhisheng Health Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hefei Xinzhisheng Health Technology Co ltd filed Critical Hefei Xinzhisheng Health Technology Co ltd
Priority to CN202211482855.3A priority Critical patent/CN115579109A/en
Publication of CN115579109A publication Critical patent/CN115579109A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Epidemiology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Primary Health Care (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Biomedical Technology (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

The disclosure provides an electrocardiogram image analysis method and device in a medical environment and a terminal device. The method comprises the steps of appointing an image to be analyzed in a terminal device to send an electrocardiogram image analysis request to a server, inputting an electrocardiogram waveform area image obtained by preprocessing the image to be analyzed into a pre-trained electrocardiogram analysis model by the server under the condition that the image to be analyzed is determined to be the electrocardiogram image to obtain a corresponding heart disease prevalence prediction result, carrying out back propagation on the electrocardiogram analysis model based on the heart disease prevalence prediction result to generate an internal feature visualized image, and finally sending electrocardiogram image analysis result information generated based on the obtained heart disease prevalence prediction result and the internal feature visualized image to the terminal device for presentation. The method realizes diagnosis of doctor users of the auxiliary terminal equipment, and embodies the auxiliary diagnosis effect of artificial intelligence.

Description

Electrocardiogram image analysis method and device in medical environment and terminal equipment
Technical Field
The embodiment of the disclosure relates to the technical field of medical image analysis, in particular to an electrocardiogram image analysis method, device, storage medium, terminal equipment, server and system in a medical environment.
Background
Electrocardiogram is a simple and commonly used cardiovascular disease detection technology in clinical practice at present. The development of artificial intelligence technology promotes the continuous updating of the electrocardiosignal analysis method. In the field of analysis of cardiac electrical signals, artificial intelligence has grown to maturity in the diagnosis of arrhythmias.
However, in some countries or regions, paper electrocardiograms are still the most common way for patients and doctors to record electrocardiograms. In some areas where the medical level is relatively laggard, the hospital's doctors are simply not knowledgeable about their own medical needs to provide medical services to the users.
Disclosure of Invention
The embodiment of the disclosure provides an electrocardiogram image analysis method, an electrocardiogram image analysis device, a storage medium, a terminal device, a server and an electrocardiogram image analysis system in a medical environment.
In a first aspect, an embodiment of the present disclosure provides an electrocardiogram image analysis method in a medical environment, which is applied to a first terminal device, and the method includes: acquiring an image to be analyzed; generating an electrocardiogram image analysis request based on the image to be analyzed in response to detecting an electrocardiogram image analysis operation for the image to be analyzed; sending the electrocardiogram image analysis request to a server, wherein the server inputs an electrocardiogram waveform area image obtained by preprocessing the image to be analyzed into a pre-trained electrocardiogram analysis model in response to determining that the image to be analyzed is an electrocardiogram image, obtains a heart disease probability prediction result for representing the probability of suffering from each of N preset heart diseases, wherein N is a positive integer, reversely propagates the obtained heart disease probability prediction result in the electrocardiogram analysis model to obtain a weight heat map, superimposes the weight heat map and the electrocardiogram waveform area image to obtain an internal feature visualized image, and generates and returns electrocardiogram image analysis result information based on the obtained heart disease probability prediction result and the internal feature visualized image; presenting the electrocardiogram image analysis result information in response to receiving electrocardiogram image analysis result information transmitted by the server in response to the electrocardiogram image analysis request.
In some optional embodiments, the acquiring an image to be analyzed includes: in response to detection of a selection operation for a first image captured by a camera in the first terminal device, determining the first image as the image to be analyzed; or in response to detecting the selected operation of the second image in the local album of the first terminal equipment, determining the second image as the image to be analyzed.
In some optional embodiments, the method further comprises: and presenting the prompt information in response to receiving the prompt information which is sent by the server and used for representing that the image to be analyzed is not an electrocardiogram image.
In some optional embodiments, the method further comprises: in response to receiving a further diagnosis request from a second terminal device forwarded by the server, wherein the further diagnosis request comprises an image to be diagnosed and a heart disease prevalence prediction result and an internal feature visualization image corresponding to the image to be diagnosed, presenting the image to be diagnosed and the heart disease prevalence prediction result and the internal feature visualization image corresponding to the image to be diagnosed; presenting an information input interface and receiving diagnosis suggestion information input by a user on the information input interface; and in response to the detection of the diagnosis suggestion confirmation operation aiming at the diagnosis suggestion information, forwarding the diagnosis suggestion information to the second terminal equipment through the server so that the second terminal equipment presents the diagnosis suggestion information.
In a second aspect, an embodiment of the present disclosure provides an electrocardiogram image analysis method in a medical environment, which is applied to a server, and the method includes: in response to receiving an electrocardiogram image analysis request which is sent by a first terminal device and generated based on an image to be analyzed, determining whether the image to be analyzed is an electrocardiogram image; in response to the determination that the image is the electrocardiogram image, preprocessing the image to be analyzed to obtain an electrocardiogram waveform area image; inputting the electrocardiogram waveform area image into a pre-trained electrocardiogram analysis model to obtain a heart disease incidence probability prediction result corresponding to the electrocardiogram waveform area image, wherein the heart disease incidence probability prediction result is used for representing the probability of suffering from each preset heart disease of N preset heart diseases, the electrocardiogram analysis model is used for representing the corresponding relation between the electrocardiogram waveform area image and the heart disease incidence probability prediction result, the electrocardiogram analysis model is a deep neural network, and N is a positive integer; carrying out back propagation on the obtained heart disease probability prediction result in the electrocardiogram analysis model to obtain a weight heat map; superposing the weight heat map and the electrocardiogram waveform area image to obtain an internal feature visual image; generating electrocardiogram image analysis result information based on the obtained heart disease probability prediction result and the internal feature visualized image; and sending the electrocardiogram image analysis result information to the first terminal equipment.
In some optional embodiments, the preprocessing the image to be analyzed to obtain an electrocardiogram waveform region image includes: performing waveform segmentation on the image to be analyzed to obtain a waveform segmentation result, wherein the waveform segmentation result is used for distinguishing a waveform from a background part in the image to be analyzed; generating a binary image corresponding to the image to be analyzed based on the waveform segmentation result; detecting an electrocardiogram waveform region in the binary image; intercepting an image in the binary image according to the electrocardiogram waveform area; an electrocardiogram waveform region image is generated based on the intercepted image.
In some optional embodiments, the method further comprises: in response to determining that the image is not an electrocardiogram image, generating prompt information indicating that the image to be analyzed is not an electrocardiogram image; and sending the prompt message to the first terminal equipment.
In a third aspect, an embodiment of the present disclosure provides an electrocardiogram image analysis apparatus in a medical environment, applied to a first terminal device, the apparatus including: an image acquisition unit configured to acquire an image to be analyzed; a request generation unit configured to generate an electrocardiogram image analysis request based on the image to be analyzed in response to detection of an electrocardiogram image analysis operation for the image to be analyzed; a request sending unit configured to send the request for analyzing the electrocardiogram image to a server, wherein the server inputs an electrocardiogram waveform area image obtained by preprocessing the image to be analyzed into a pre-trained electrocardiogram analysis model in response to determining that the image to be analyzed is an electrocardiogram image, obtains a heart disease probability prediction result for representing the probability of suffering from each of N preset heart diseases, wherein N is a positive integer, reversely propagates the obtained heart disease probability prediction result in the electrocardiogram analysis model to obtain a weight heat map, superimposes the weight heat map and the electrocardiogram waveform area image to obtain an internal feature visualized image, and generates and returns electrocardiogram image analysis result information based on the obtained heart disease probability prediction result and the internal feature visualized image; a result information presentation unit configured to present the electrocardiogram image analysis result information in response to receiving the electrocardiogram image analysis result information transmitted by the server in response to the electrocardiogram image analysis request.
In some optional embodiments, the image acquisition unit is further configured to: in response to detection of a selection operation for a first image captured by a camera in the first terminal device, determining the first image as the image to be analyzed; or in response to detecting the selected operation of the second image in the local album of the first terminal equipment, determining the second image as the image to be analyzed.
In some optional embodiments, the apparatus further comprises: and the prompt information presenting unit is configured to present the prompt information in response to receiving the prompt information which is sent by the server and used for representing that the image to be analyzed is not an electrocardiogram image.
In some optional embodiments, the apparatus further comprises: an image presenting unit configured to present an image to be diagnosed and a heart attack probability prediction result and an internal feature visualization image corresponding to the image to be diagnosed in response to receiving a further diagnosis request from a second terminal device forwarded via the server, the further diagnosis request including the image to be diagnosed and the heart attack probability prediction result and the internal feature visualization image corresponding to the image to be diagnosed; the interface presentation unit is configured to present an information input interface and receive diagnosis suggestion information input by a user on the information input interface; and the diagnosis suggestion sending unit is configured to respond to the detection of a diagnosis suggestion confirmation operation aiming at the diagnosis suggestion information, and forward the diagnosis suggestion information to the second terminal equipment through the server so as to enable the second terminal equipment to present the diagnosis suggestion information.
In a fourth aspect, an embodiment of the present disclosure provides an electrocardiogram image analysis apparatus applied to a server in a medical environment, the apparatus including: an image recognition unit configured to determine whether an image to be analyzed is an electrocardiogram image in response to receiving an electrocardiogram image analysis request generated based on the image to be analyzed and transmitted by a first terminal device; a waveform extraction unit configured to preprocess the image to be analyzed to obtain an electrocardiogram waveform region image in response to a determination that the image is an electrocardiogram image; a probability prediction unit configured to input the electrocardiogram waveform area image into a pre-trained electrocardiogram analysis model to obtain a heart disease probability prediction result corresponding to the electrocardiogram waveform area image, wherein the heart disease probability prediction result is used for representing the probability of suffering from each of N preset heart diseases, the electrocardiogram analysis model is used for representing the corresponding relation between the electrocardiogram waveform area image and the heart disease probability prediction result, the electrocardiogram analysis model is a deep neural network, and N is a positive integer; a back propagation unit configured to perform back propagation on the obtained heart disease probability prediction result in the electrocardiogram analysis model to obtain a weight heat map; an image generation unit configured to superimpose the weight heat map and the electrocardiogram waveform region image to obtain an internal feature visualized image; an analysis result generation unit configured to generate electrocardiogram image analysis result information based on the obtained heart disease prevalence probability prediction result and the internal feature visualized image; and an analysis result transmitting unit configured to transmit the electrocardiogram image analysis result information to the first terminal device.
In some optional embodiments, the waveform extraction unit is further configured to: performing waveform segmentation on the image to be analyzed to obtain a waveform segmentation result, wherein the waveform segmentation result is used for distinguishing a waveform from a background part in the image to be analyzed; generating a binary image corresponding to the image to be analyzed based on the waveform segmentation result; detecting an electrocardiogram waveform region in the binary image; intercepting an image from the binary image according to the electrocardiogram waveform area; an electrocardiogram waveform region image is generated based on the intercepted image.
In some optional embodiments, the apparatus further comprises: a prompt information generation unit configured to generate prompt information indicating that the image to be analyzed is not an electrocardiogram image in response to a determination that it is not an electrocardiogram image; and a prompt information sending unit configured to send the prompt information to the first terminal device.
In a fifth aspect, an embodiment of the present disclosure provides a terminal device, including: one or more processors; a storage device, on which one or more programs are stored, which, when executed by the one or more processors, cause the one or more processors to implement the method as described in any implementation manner of the first aspect.
In a sixth aspect, an embodiment of the present disclosure provides a server, including: one or more processors; a storage device, on which one or more programs are stored, which, when executed by the one or more processors, cause the one or more processors to implement the method as described in any implementation manner of the second aspect.
In a seventh aspect, embodiments of the present disclosure provide a computer-readable storage medium on which a computer program is stored, wherein the computer program, when executed by one or more processors, implements the method as described in any of the implementations of the first aspect and/or the method as described in any of the implementations of the second aspect.
In an eighth aspect, an embodiment of the present disclosure provides an electrocardiogram image analysis system in a medical environment, which includes the terminal device described in any implementation manner of the fifth aspect and the server described in any implementation manner of the sixth aspect.
In order to solve the problem that doctors in regions with relatively laggard medical levels in the prior art only need to provide medical services for users by virtue of insufficient medical knowledge, the electrocardiogram image analysis method, the device, the storage medium, the terminal device, the server and the system in the medical environment provided by the embodiment of the disclosure specify an image to be analyzed in the first terminal device to send an electrocardiogram image analysis request to the server, then the server inputs an electrocardiogram waveform area image obtained by preprocessing the image to be analyzed into a pre-trained electrocardiogram analysis model under the condition that the image to be analyzed is determined to be an electrocardiogram image, so as to obtain a corresponding heart disease probability prediction result, wherein the electrocardiogram analysis model is a deep neural network, then the obtained heart disease probability prediction result is reversely propagated in the electrocardiogram analysis model to obtain a weight heat map, then the weight heat map and the electrocardiogram waveform area image are superposed to obtain an internal feature visualized image, finally electrocardiogram image analysis result information is generated on the basis of the obtained heart disease probability prediction result and the internal feature visualized image, and the electrocardiogram image analysis result information is sent to the first terminal device to be presented. Thereby realized forming after first terminal equipment shoots the paper heart electrograph and waiting to analyze the image, through high in the clouds server real-time paper heart electrograph image analysis and return heart electrograph image analysis result, and including two information in the analysis result that returns, an information is the probability value that has every heart disease in the preset N, N is the positive integer, an information is the visual image of inside characteristic, above-mentioned two kinds of information can assist first terminal equipment's doctor user to diagnose, the supplementary diagnostic effect of artificial intelligence has been embodied.
Drawings
Other features, objects, and advantages of the disclosure will become apparent from a reading of the following detailed description of non-limiting embodiments which proceeds with reference to the accompanying drawings. The drawings are only for purposes of illustrating particular embodiments and are not to be construed as limiting the invention. In the drawings:
FIG. 1 is a system architecture diagram of one embodiment of an electrocardiographic image analysis system in a medical environment according to the present disclosure;
FIG. 2A is a timing diagram of one embodiment of an electrocardiogram image analysis system in a medical environment according to the present disclosure;
FIG. 2B is an exploded flow diagram for one embodiment of step 205, according to the present disclosure;
FIG. 3 is a flow chart of one embodiment of a method for electrocardiographic image analysis in a medical environment applied to a first terminal device in accordance with the present disclosure;
FIG. 4 is a flow chart of one embodiment of a method for electrocardiography image analysis in a medical environment applied to a server according to the present disclosure;
FIG. 5 is a schematic structural diagram of an embodiment of an electrocardiogram image analysis apparatus applied to a medical environment of a first terminal device according to the present disclosure;
FIG. 6 is a schematic diagram of an embodiment of an electrocardiogram image analysis apparatus applied to a server in a medical environment according to the present disclosure;
fig. 7 is a schematic structural diagram of a computer system of a terminal device or server suitable for implementing an embodiment of the present disclosure.
Detailed Description
The present disclosure is described in further detail below with reference to the figures and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.
It should be noted that, in the present disclosure, the embodiments and features of the embodiments may be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
Fig. 1 illustrates an exemplary system architecture 100 for one embodiment of an electrocardiographic image analysis system in a medical environment to which the present disclosure may be applied.
As shown in fig. 1, an electrocardiogram image analysis system 100 in a medical environment may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
A user may use terminal devices 101, 102, 103 to interact with a server 105 over a network 104 to receive or send messages or the like. Various communication client applications, such as an electrocardiogram image analysis application, an audio and video conference application, an image acquisition application, an image processing application, a web browser application, a shopping application, a search application, an instant messaging tool, a mailbox client, social platform software, and the like, may be installed on the terminal devices 101, 102, and 103.
The terminal devices 101, 102, 103 may be hardware or software. When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices having a display screen and supporting sound collection and/or video collection, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, motion Picture Experts compression standard Audio Layer 3), MP4 players (Moving Picture Experts Group Audio Layer IV, motion Picture Experts compression standard Audio Layer 4), laptop portable computers, desktop computers, and the like. When the terminal apparatuses 101, 102, 103 are software, they can be installed in the electronic apparatuses listed above. It may be implemented as a plurality of software or software modules (for example, for providing services such as electrocardiogram image analysis in a medical environment), or as a single software or software module. And is not particularly limited herein.
The server 105 may be a server providing various services, such as a background server providing support for an electrocardiogram image analysis-like application in the medical environment displayed on the terminal devices 101, 102, 103. The background server may analyze and otherwise process the received data such as the electrocardiogram image analysis request, and feed back a processing result (e.g., electrocardiogram image analysis result information) to the terminal device.
The server 105 may be hardware or software. When the server 105 is hardware, it may be implemented as a distributed server cluster composed of a plurality of servers, or may be implemented as a single server. When the server is software, it may be implemented as a plurality of software or software modules (for example, for providing services such as electrocardiogram image analysis), or may be implemented as a single software or software module. And is not particularly limited herein.
It should be noted that the electrocardiogram image analysis method in the medical environment applied to the first terminal device provided by the present disclosure is generally executed by the terminal devices 101, 102, 103, and accordingly, the electrocardiogram image analysis apparatus in the medical environment applied to the first terminal device is generally disposed in the terminal devices 101, 102, 103.
It should be noted that, in the present disclosure, the first terminal device may be a terminal device used by a doctor in a home environment, or may be a terminal device used by one or more doctors with corresponding medical qualifications in a medical environment.
It should be noted that the electrocardiographic image analysis method in the medical environment applied to the server provided by the present disclosure is generally executed by the server 105, and accordingly, the electrocardiographic image analysis apparatus in the medical environment applied to the server is generally provided in the server 105.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for an implementation.
With continued reference to fig. 2A, a timing sequence 200 of one embodiment of an electrocardiographic image analysis system in a medical environment is illustrated in accordance with the present disclosure. The electrocardiogram image analysis system in the medical environment in the embodiment of the disclosure may include a first terminal device and a server. The sequence 200 includes the following steps:
step 201, a first terminal device obtains an image to be analyzed.
In this embodiment, the first terminal device (e.g., terminal devices 101, 102, 103 shown in fig. 1) may acquire the image to be analyzed in various implementations.
Here, the first terminal device may acquire the image to be analyzed locally or remotely from other electronic devices network-connected to the first terminal device.
In some alternative embodiments, step 201 may be performed as follows:
in response to detecting a selected operation for a first image captured by a camera in a first terminal device, the first image is determined as an image to be analyzed.
In some alternative embodiments, step 201 may also be performed as follows:
and in response to detecting the selected operation aiming at the second image in the local album of the first terminal equipment, determining the second image as the image to be analyzed.
In step 202, the first terminal device generates an electrocardiogram image analysis request based on an image to be analyzed in response to detecting an electrocardiogram image analysis operation for the image to be analyzed.
In the present embodiment, the electrocardiographic image analyzing operation for the image to be analyzed may be various preset operations for triggering generation of an electrocardiographic image analyzing request based on the image to be analyzed. For example, an electrocardiogram analysis operation display object (e.g., an icon or button confirming submission of an electrocardiogram image analysis request) is presented first, and it is determined that an electrocardiogram image analysis operation for an image to be analyzed is detected if a first preset selection operation (e.g., a click, a double click, a slide, a hover, a drag, etc.) for the above-described electrocardiogram analysis operation display object is detected. For another example, the voice data collected by the microphone of the first terminal device may be acquired in real time, and the voice data may be recognized, and if the voice recognition result includes a preset electrocardiogram analysis instruction statement (e.g., an analysis electrocardiogram), it may be determined that an electrocardiogram image analysis operation for the image to be analyzed is detected.
Here, various implementations may be employed to generate an electrocardiogram image analysis request based on an image to be analyzed. The electrocardiogram image analysis request may include an image to be analyzed. Optionally, the electrocardiogram image analysis request may further include, in addition to the image to be analyzed, a current user identifier of a current user who is currently logged in and authenticated by the first terminal device.
In step 203, the first terminal device sends an electrocardiogram image analysis request to the server.
In the present embodiment, the first terminal device may transmit the electrocardiogram image analysis request generated in step 202 to the server. Here, the server may be a server that provides an electrocardiogram image analysis service.
In step 204, the server determines whether the image to be analyzed is an electrocardiogram image in response to receiving the electrocardiogram image analysis request sent by the first terminal device.
In this embodiment, the server may, upon receiving the electrocardiogram image analysis request sent by the first terminal device, first parse the electrocardiogram image analysis request to obtain an image to be analyzed, and then may determine whether the image to be analyzed is an electrocardiogram image by using various implementations.
In practice, since the electrocardiogram image has a background portion and a waveform portion. And, there is a clear difference in pattern rule between the background portion and the waveform portion, so that it is possible to determine whether the image to be analyzed is an electrocardiogram image using various preset electrocardiogram image determination rules.
Optionally, the server may input the image to be analyzed into a pre-trained electrocardiogram image recognition model, so as to obtain a recognition result for representing whether the image to be analyzed is an electrocardiogram image. Here, the electrocardiogram image recognition model may be trained in advance by the following first training step:
first, a first sample data set is acquired and a model structure and model parameters of an initial electrocardiogram image recognition model are determined.
Here, the first sample data in the first sample data set includes a sample image and corresponding first annotation information for characterizing whether the sample image is an electrocardiogram image.
As an example, the sample image may be obtained by at least one of:
first, electrocardiographic signal data from an open source is plotted into an electrocardiographic image.
Second, paper electrocardiograms are acquired from the clinical environment of a hospital and scanned or photographed into electrocardiogram images.
Third, a natural image is acquired from an open source natural image dataset.
The first labeling information in the first sample data can be obtained through manual labeling. For example, the sample images obtained in the first and second aspects are labeled as electrocardiogram images, and the sample image obtained in the third aspect is labeled as a non-electrocardiogram image.
Here, the initial electrocardiographic image recognition model may be various machine learning models, and the present disclosure does not specifically limit this. The initial electrocardiogram image recognition model is used for representing the corresponding relation between the image and an electrocardiogram image result used for representing whether the sample image is an electrocardiogram image. As an example, the initial electrocardiogram image recognition model may be a deep neural network.
The first sample dataset is then divided into a first training dataset, a first verification dataset, and a first test dataset.
And finally, training the initial electrocardiogram image recognition model by using a machine learning method based on the first training data set, the first verification data set and the first test data set to obtain a trained electrocardiogram image recognition model.
In step 205, the server preprocesses the image to be analyzed to obtain an electrocardiogram waveform region image in response to determining that the image is an electrocardiogram image.
In practice, the image to be analyzed uploaded by the user through the first terminal device may include other contents besides the electrocardiogram image. For example, when a paper electrocardiogram is placed on a table, the image to be analyzed may include a table portion. In addition, even if all of the images to be analyzed are electrocardiogram images, in practice, the electrocardiogram images may include other contents, such as user information and electrocardiogram acquisition time, in addition to the electrocardiogram waveform region. In order to improve the subsequent prediction effect on the heart disease affection probability, if the server determines that the image to be analyzed is an electrocardiogram image in step 204, the image to be analyzed may be preprocessed to obtain an electrocardiogram waveform area image. Here, the electrocardiogram waveform region image is used to characterize an electrocardiogram waveform region in the image to be analyzed.
In some alternative embodiments, step 205 may include steps 2051 through 2055 as shown in fig. 2B:
and step 2051, performing waveform segmentation on the image to be analyzed to obtain a waveform segmentation result.
Here, the waveform segmentation result is used to distinguish a waveform portion and a background portion in the image to be analyzed.
In practice, besides the waveform image of the electrocardiogram, the paper electrocardiogram may have background grids and background paper colors, which are not favorable for subsequent processing. For convenience of subsequent processing, various image foreground segmentation algorithms can be adopted to perform waveform segmentation on the image to be analyzed, so as to obtain a waveform segmentation result. The waveform segmentation result is used for distinguishing a waveform part and a background part in the image to be analyzed.
For example, assuming that the image size of the image to be analyzed is m × n, here, the waveform segmentation result may be represented by an m × n matrix, an element in the matrix may be 0 or 1, and when the matrix element is 0, it indicates that a pixel point at a corresponding coordinate position in the image to be analyzed is a background portion. On the contrary, when the matrix element is 1, it indicates that the pixel point at the corresponding coordinate position in the image to be analyzed is the waveform portion.
Optionally, in step 2051, the image to be analyzed is subjected to waveform segmentation to obtain a waveform segmentation result, and the image to be analyzed may be input into the electrocardiogram waveform segmentation model to obtain a waveform segmentation result.
Here, the electrocardiogram waveform segmentation model may be trained in advance by the following second training step:
first, a second sample data set is acquired and a model structure and model parameters of an initial electrocardiogram waveform segmentation model are determined.
Here, the second sample data may include a sample electrocardiogram image and a corresponding annotated electrocardiogram waveform segmentation result. Wherein the annotated electrocardiogram waveform segmentation result is used for distinguishing a waveform part and a background part in the sample electrocardiogram image. In practice, the sample electrocardiogram image may be manually labeled to mark the electrocardiogram waveform portion and the background portion thereof.
Here, the initial electrocardiogram waveform segmentation model may be various machine learning models, which are not particularly limited by the present disclosure. The initial electrocardiogram waveform segmentation model is used for representing the corresponding relation between an electrocardiogram image and a waveform segmentation result. As an example, the initial electrocardiogram waveform segmentation model may be a deep neural network.
The second sample data set is then divided into a second training data set, a second verification data set, and a second test data set.
And finally, training the initial electrocardiogram waveform segmentation model by using a machine learning method based on the second training data set, the second verification data set and the second test data set to obtain a trained electrocardiogram waveform segmentation model.
And step 2052, generating a binary image corresponding to the image to be analyzed based on the waveform segmentation result.
Here, the value of the pixel point in the generated binarized image may be a waveform pixel value for representing an electrocardiogram waveform or a background pixel value for representing a background. That is, the generated binary image is a binary image including waveform pixel points and background pixel points.
Step 2053, detecting an electrocardiogram waveform area in the binarized image.
In practice, generally, the paper electrocardiogram central electrogram waveform does not occupy the whole area. In order to improve the accuracy of analyzing the electrocardiogram and further obtaining the probability of different heart diseases, the electrocardiogram waveform area in the image to be analyzed can be detected firstly. The electrocardiogram waveform region can be characterized by various data forms. The electrocardiogram waveform region may be a set of coordinates in the image to be analyzed. The electrocardiogram waveform region may also be a shape identifier and a feature value for characterizing a coordinate range of a shape indicated by the shape identifier in an image to be analyzed. For example, an electrocardiogram waveform region may be characterized by a rectangular logo and four vertex coordinates of the rectangular electrocardiogram waveform region.
Here, the execution subject described above may detect an electrocardiogram waveform region in an image to be analyzed using various target detection algorithms.
For example, the electrocardiogram waveform region may be rectangular, and the electrocardiogram waveform region may include four vertex coordinates of the rectangular electrocardiogram waveform region.
Alternatively, step 2053 may be performed as follows: and inputting the image to be analyzed into a pre-trained electrocardiogram waveform region detection model to obtain an electrocardiogram waveform region in the image to be analyzed.
Here, the electrocardiogram waveform region detection model may be trained in advance by the following third training step:
first, a third sample data set is obtained and a model structure and model parameters of an initial electrocardiogram waveform region detection model are determined.
Here, the third sample data may include a sample electrocardiogram image and a corresponding annotated electrocardiogram waveform region. And the marked electrocardiogram waveform region is used for representing the electrocardiogram waveform region in the corresponding sample electrocardiogram image. In practice, the sample electrocardiogram image may be manually labeled to mark the electrocardiogram waveform region therein.
The sample electrocardiogram images in the third sample data set may include electrocardiogram images from different types of electrocardiographs, which detect different subjects. Optionally, the sample electrocardiogram images in the third sample data set may further include electrocardiogram images taken by various different types of cameras under different lighting environmental conditions or scanned by different types of scanners. And further ensure that the robustness of the electrocardiogram waveform region detection model obtained by training is higher.
Here, the initial electrocardiogram waveform region detection model may be various machine learning models, which are not particularly limited by the present disclosure. The initial electrocardiogram waveform region detection model is used for representing the corresponding relation between an electrocardiogram image and a waveform region. As an example, the initial electrocardiogram waveform region detection model may be a deep neural network.
The third sample data set is then divided into a third training data set, a third verification data set, and a third test data set.
And finally, training the initial electrocardiogram waveform region detection model by using a machine learning method based on the third training data set, the third verification data set and the third test data set to obtain a trained electrocardiogram waveform region detection model.
And step 2054, intercepting an image from the binary image according to the electrocardiogram waveform area.
Step 2055, generating an electrocardiogram waveform region image based on the captured image.
Here, an electrocardiogram waveform region image may be obtained based on the clipped image in accordance with the input data requirements of the electrocardiogram analysis model.
For example, the captured image may be regarded as an electrocardiogram waveform region image.
For another example, the captured image may be first scaled to a preset image size suitable for input by the ecg analysis model, and then the scaled captured image may be used as the ecg waveform region image.
For example, the clipped image may be first scaled to a preset image size suitable for input by the electrocardiogram analysis model, then the scaled clipped image may be normalized, and the normalized image may be used as the electrocardiogram waveform region image.
The electrocardiogram waveform region image obtained by adopting the optional embodiment can obtain an image which meets the data input requirement of an electrocardiogram analysis model.
Step 206, the server inputs the electrocardiogram waveform area image into a pre-trained electrocardiogram analysis model to obtain a heart disease incidence probability prediction result corresponding to the electrocardiogram waveform area image.
Here, the electrocardiogram analysis model is used for representing the corresponding relation between the electrocardiogram waveform area image and the heart disease incidence probability prediction result, wherein the heart disease incidence probability prediction result is used for representing the probability of suffering from each preset heart disease of the N preset heart diseases. Wherein N is a positive integer. Optionally, the heart disease probability prediction result may be an N-dimensional vector, N components in the N-dimensional vector correspond to N preset heart diseases one to one, and each component is used to represent a probability of having the corresponding preset heart disease of the N preset heart diseases. For example, the component in the prediction result of the heart disease probability may be a value between 0 and 1, and the closer the value is to 1, the greater the probability of having the corresponding predetermined heart disease.
In practice, heart diseases can be divided into various types. Here, the N preset cardiac diseases may be N different types of cardiac diseases. Optionally, N is a positive integer greater than or equal to 2. Therefore, the electrocardiogram analysis model can output the disease probability of more than or equal to two heart diseases.
As an example, the ecg analysis model may be a calculation formula which is pre-formulated by a technician to perform a statistical analysis on ecg waveform region images of patients diagnosed with different heart diseases of the N preset heart diseases based on a large number of actual conditions, and then calculate the ecg waveform region images to obtain the N different preset heart disease incidence probabilities.
In some alternative embodiments, the ecg analysis model may also be pre-trained by the following fourth training step:
first, a fourth sample data set is acquired and the model structure and model parameters of the initial electrocardiogram analysis model are determined.
Here, the fourth sample data may include a sample electrocardiogram waveform area image and a corresponding annotated heart disease prevalence probability prediction result. The heart disease prevalence probability prediction result is used for representing the prevalence probability of each heart disease in N preset heart diseases of the examinee corresponding to the electrocardiogram waveform area image of the corresponding sample. In practice, a doctor or a technician with professional medical knowledge can label the sample electrocardiogram waveform area image to obtain the prediction result of the heart disease prevalence rate.
The fourth sample data set is then divided into a fourth training data set, a fourth verification data set, and a fourth test data set.
And finally, training the initial electrocardiogram analysis model by using a machine learning method based on the fourth training data set, the fourth verification data set and the fourth test data set to obtain a trained electrocardiogram analysis model.
And step 207, the server performs back propagation on the obtained heart disease prevalence probability prediction result in the electrocardiogram analysis model to obtain a weight heat map.
Because the deep neural network is poor in interpretability, in order to interpret the heart disease probability prediction result given by the electrocardiogram analysis model (for example, interpreting the electrocardiogram analysis model so as to give a corresponding heart disease probability prediction result for an electrocardiogram waveform region image, because the electrocardiogram analysis model "sees" which parts in the electrocardiogram waveform region may have problems, or because the electrocardiogram analysis model "sees" which parts in the electrocardiogram waveform region may have differences from a healthy electrocardiogram and cause a certain heart disease, there may be adopted various now known or future developed weighting heat map generation methods for interpreting the deep neural network, and the internal features of the deep neural network corresponding to the heart disease probability prediction result obtained in step 206 (for example, the features output by the last layer or the full connection layer of the deep neural network corresponding to the heart disease probability prediction result) are propagated in the electrocardiogram analysis model in a reverse direction to obtain weights, which is not specifically limited by this disclosure. For example, including but not limited to the following methods: a significant map (salience map) method, a Guided Back Propagation (GBP) method, an LRP (Layer-wise Propagation) method, a Grad-CAM (Gradient-weighted Class Activation Mapping) method, and the like.
And step 208, the server superposes the weight heat map and the electrocardiogram waveform area image to obtain an internal feature visualization image.
Here, various image superimposition methods may be employed to generate the internal feature visualized image.
In practice, the weight thermal map is generally a feature image with the same size as the electrocardiogram waveform region image, the pixel value of each pixel point in the weight thermal map can be a numerical value between 0 and 1, and the closer the pixel value is to 1, the more the electrocardiogram analysis model focuses on the pixel point in the same position as the pixel point in the electrocardiogram waveform region image, or the more the possibility that the pixel point in the same position as the pixel point in the electrocardiogram waveform region image is a lesion pixel point is.
As an example, superimposing the weight heatmap and the electrocardiogram waveform area image to obtain an internal feature visualization image may be performed as follows:
first, a lesion-level feature image corresponding to the weight heat map is generated.
And each pixel point in the generated lesion degree characteristic image corresponds to a pixel point in the weight heat image one by one, and the pixel value of each pixel point in the lesion degree image is determined according to the pixel value of the pixel point at the same position in the weight heat image. For example, the smaller the pixel value of a pixel point in the weight heat map is, the closer the pixel value of the corresponding pixel point in the lesion degree feature image is to the pixel value representing the first preset color (e.g., blue); conversely, the larger the pixel value of the pixel point in the weight heat map, the closer the pixel value of the corresponding pixel point in the lesion degree feature image is to the pixel value representing the second preset color (e.g., red).
And then, covering the focus degree characteristic image on the electrocardiogram waveform region image according to the preset transparency to obtain an internal characteristic visualized image.
The degree that different pixel points in the electrocardiogram waveform region image are focus pixel points can be visually seen through the internal characteristic visual image, namely, the electrocardiogram analysis model can be intuitively explained.
In step 209, the server generates electrocardiogram image analysis result information based on the obtained heart disease probability prediction result and the internal feature visualization image.
Here, the server may generate the electrocardiographic image analysis result information based on the prediction result of the heart disease prevalence probability obtained in step 206 and the internal feature visualization image obtained in step 208 in various ways according to the needs of a specific application scenario. For example, the result of the analysis of the electrocardiogram image can be generated by directly using the predicted heart disease probability in step 206 and the internal feature visualization image in step 208.
Here, the electrocardiographic image analysis result information may be in various forms. For example, text, image, and voice data may be included, but are not limited to.
In step 210, the server sends the electrocardiogram image analysis result information to the first terminal device.
Here, the server may transmit the electrocardiographic image analysis result information generated in step 209 to the first terminal device from which the electrocardiographic image analysis request received in step 204 is received, as feedback information of the electrocardiographic image analysis request.
In step 211, the first terminal device presents electrocardiogram image analysis result information in response to receiving electrocardiogram image analysis result information transmitted by the server in response to the electrocardiogram image analysis request.
Here, the first terminal device may present the received electrocardiogram image analysis result information in various implementations.
Specifically, the electrocardiogram image analysis result information may be presented on a display device, for example, in the form of text or images. And the voice corresponding to the electrocardiogram image analysis result information can also be played on the sound playing equipment. The present disclosure is not particularly limited thereto.
Through steps 201 to 211, an image to be analyzed may be designated in the first terminal device to initiate an electrocardiogram image analysis request to the server, the image to be analyzed may be designated in the first terminal device to initiate the electrocardiogram image analysis request to the server, the server may input an electrocardiogram waveform area image obtained by preprocessing the image to be analyzed into a pre-trained electrocardiogram analysis model in a case where it is determined that the image to be analyzed is an electrocardiogram image, so as to obtain a corresponding heart disease prevalence probability prediction result, the obtained heart disease prevalence probability prediction result may be back-propagated in the electrocardiogram analysis model to obtain a weight heat map, the weight heat map and the electrocardiogram waveform area image may be superimposed to obtain an internal feature visualized image, and finally, electrocardiogram image analysis result information may be generated based on the obtained heart disease prevalence probability prediction result and the internal feature visualized image, and the electrocardiogram image analysis result information may be sent to the first terminal device for presentation. Thereby realized forming the image of waiting to analyze after first terminal equipment shoots the paper heart electrograph, carry out analysis and return heart electrograph image analysis result to the paper heart electrograph image through high in the clouds server in real time, and the analysis result that returns includes two information, an information is the probability value that has every heart disease among N kinds of heart diseases of predetermineeing, N is the positive integer, an information is the visual image of internal feature, can make things convenient for the doctor user at first terminal to obtain the region of paying close attention to of the heart electrograph analysis model of heart electrograph waveform region image after the image of waiting to analyze is preprocessed through the visual image of internal feature, above-mentioned two kinds of information can assist the doctor user of first terminal equipment to diagnose, the supplementary diagnostic effect of artificial intelligence has been embodied.
In some optional embodiments, the above timing sequence 200 may further perform the following steps 212 to 214 in case that it is determined in step 204 that the image to be analyzed is not an electrocardiogram image:
in step 212, the server generates prompt information indicating that the image to be analyzed is not an electrocardiogram image in response to determining that it is not an electrocardiogram image.
That is, if it is determined in step 204 that the image to be analyzed is not an electrocardiogram image, prompt information indicating that the image to be analyzed is not an electrocardiogram image is generated.
Step 213, the server sends the prompt message to the first terminal device.
And step 214, the first terminal device presents the prompt message.
With the alternative implementation of steps 212 to 214, in the case that the image to be analyzed uploaded by the first terminal device is not an electrocardiogram image, a prompt message may be given to remind the user of the first terminal device to upload a valid electrocardiogram image.
Based on the optional embodiment, optionally, the timing sequence flow 200 may further include the following steps 215 to 217:
step 215, the first terminal device presents the image to be diagnosed and the heart disease probability prediction result and the internal feature visualization image corresponding to the image to be diagnosed in response to receiving the further diagnosis request forwarded by the server from the second terminal device.
Here, the further diagnosis request may include an image to be diagnosed and a heart attack probability prediction result and an internal feature visualization image corresponding to the image to be diagnosed.
As an example, the second terminal device may be a terminal device used by a general user in a home environment. The common user can send the image to be diagnosed to the server for analysis by using the second terminal device. The server may input an electrocardiogram waveform area image obtained by preprocessing the image to be diagnosed into a pre-trained electrocardiogram analysis model to obtain a heart disease probability prediction result, and generate heart disease diagnosis result information based on the obtained heart disease probability prediction result, in a case where it is determined that the image to be diagnosed is an electrocardiogram image. The server may further generate an internal feature visualized image corresponding to the heart disease probability prediction result and the image to be analyzed obtained here by using the methods described in step 207 and step 208, and finally, the server sends the obtained heart disease diagnosis result information and the internal feature visualized image to the second terminal device. Then, the general user can obtain the above-mentioned cardiac disease diagnosis result information and internal feature visualization image at the second terminal device. If the general user wishes to make a further diagnosis by a doctor based on the obtained information on the diagnosis result of the cardiac disease and the internal feature visualized image, a further diagnosis operation can be initiated by using the second terminal device. Here, the further diagnosis operation may be various preset operations for triggering generation of a further diagnosis request.
For example, the second terminal device may first present a further diagnosis operation display object (e.g., an icon or button confirming submission of a further diagnosis request), and determine that a further diagnosis operation is detected if a second preset selection operation (e.g., a single click, a double click, a slide, a hover, a drag, etc.) for the above-described further diagnosis operation display object is detected. For another example, the voice data collected by the microphone of the second terminal device may be acquired in real time, and the voice data may be recognized, and if the voice recognition result includes a preset further diagnosis instruction sentence (for example, further seeking help of a doctor), it may be determined that a further diagnosis operation is detected.
After detecting the further diagnosis operation, the second terminal device may generate a further diagnosis request including the above-described image to be diagnosed, the heart disease prevalence prediction result, and the internal feature visualization image. Optionally, the further diagnosis request may further include, in addition to the image to be diagnosed, the heart disease prevalence prediction result, and the internal feature visualization image, a current user identifier of a currently logged-in and verified current user of the second terminal device. Optionally, the current user of the second terminal device may further specify, by using the second terminal device, a hospital, a specific department of the hospital, or a doctor, which is to be further diagnosed, and the further diagnosis request generated here may further include a hospital identifier of the hospital, a department identifier of the department, or a doctor identity of the doctor, which is specified by the current user. The generated further diagnosis request is sent to the server and forwarded to the first terminal device via the server.
Step 216, the first terminal device presents the information input interface and receives the diagnosis suggestion information input by the user on the information input interface.
Here, the information input interface may provide text, image, or voice input.
Namely, the doctor user can give out further diagnosis suggestion information by combining medical knowledge through the image to be diagnosed presented by the first terminal device, the heart disease probability prediction result corresponding to the image to be diagnosed and the internal feature visualization image. Here, the diagnosis advice information may be input in a text form, or may be input in an image form, or may also be input in voice, which is not particularly limited by the present disclosure.
In step 217, the first terminal device responds to the detection of the diagnosis suggestion confirmation operation aiming at the diagnosis suggestion information, and forwards the diagnosis suggestion information to the second terminal device through the server so that the second terminal device can present the diagnosis suggestion information.
Here, the diagnosis advice confirmation operation may be various preset operations for triggering the first terminal device to forward the diagnosis advice information to the second terminal device via the server.
For example, the first terminal device may first present a diagnosis suggestion confirmation operation display object (e.g., an icon or button confirming submission of diagnosis suggestion information), and determine that the diagnosis suggestion confirmation operation is detected if a third preset selection operation (e.g., a single click, a double click, a slide, a hover, a drag, etc.) for the above diagnosis suggestion confirmation operation display object is detected. For another example, the voice data collected by the microphone of the first terminal device may be obtained in real time, and the voice data may be recognized, and if the voice prediction result includes a preset diagnosis suggestion confirmation operation instruction sentence (for example, confirmation of uploading diagnosis suggestion information), it may also be determined that the diagnosis suggestion confirmation operation is detected.
Through steps 215 to 217, the doctor user can use the first terminal device to receive the further diagnosis request through the server under the condition that the ordinary user uses the second terminal device to send the further diagnosis request in the home environment, and according to the image to be diagnosed, the heart disease probability prediction result and the internal feature visualization image in the further diagnosis request, the further diagnosis suggestion is given by combining the medical knowledge of the doctor user, and the diagnosis suggestion information is forwarded to the second terminal device through the server, so that the further diagnosis suggestion is provided for the ordinary user, and the remote auxiliary diagnosis of the ordinary user is realized.
With continued reference to fig. 3, a flow 300 of one embodiment of a method for electrocardiographic image analysis in a medical environment is shown in accordance with the present disclosure. The electrocardiogram image analysis method in the medical environment is applied to first terminal equipment and comprises the following steps:
step 301, acquiring an image to be analyzed.
Step 302, in response to detecting an electrocardiogram image analysis operation for an image to be analyzed, generates an electrocardiogram image analysis request based on the image to be analyzed.
Step 303, sending an electrocardiogram image analysis request to a server.
Step 304, presenting electrocardiogram image analysis result information in response to receiving electrocardiogram image analysis result information sent by the server in response to the electrocardiogram image analysis request.
In this embodiment, the specific operations of step 301, step 302, step 303 and step 304 and the technical effects thereof are substantially the same as the operations and effects of step 201, step 202, step 203 and step 211 in the embodiment shown in fig. 2A, and are not repeated herein.
In some alternative embodiments, step 301 may proceed as follows:
in response to detecting a selection operation of a first image shot by a camera in a first terminal device, determining the first image as an image to be analyzed; or
And in response to detecting the selected operation aiming at the second image in the local album of the first terminal equipment, determining the second image as the image to be analyzed.
Here, the specific operations and technical effects of the above-mentioned alternative embodiment in step 301 are substantially the same as those described in the corresponding alternative embodiment in the embodiment shown in fig. 2A, and are not described again here.
In some optional embodiments, the method flow 300 may further include the following step 305:
step 305, in response to receiving the prompt message sent by the server for indicating that the image to be analyzed is not an electrocardiogram image, presenting the prompt message.
Here, the detailed operation of step 305 and the technical effect thereof are substantially the same as the operation and effect described in step 214 in the embodiment shown in fig. 2A, and are not repeated herein.
In some optional embodiments, the method flow 300 may further include the following step 306:
and step 306, in response to receiving a further diagnosis request from the second terminal device forwarded by the server, presenting the image to be diagnosed and the heart disease probability prediction result and the internal feature visualization image corresponding to the image to be diagnosed.
Step 307, presenting the information input interface and receiving the diagnosis suggestion information input by the user on the information input interface.
And 308, responding to the detection of the diagnosis suggestion confirmation operation aiming at the diagnosis suggestion information, forwarding the diagnosis suggestion information to the second terminal equipment through the server so as to enable the second terminal equipment to present the diagnosis suggestion information.
Here, the specific operations of step 306, step 307, and step 308 and the technical effects thereof are substantially the same as those described in step 215, step 216, and step 217 in the embodiment shown in fig. 2A, and are not repeated herein.
The method provided by the above embodiment of the present disclosure initiates an electrocardiogram analysis request on the first terminal device, and presents electrocardiogram image analysis result information received from the server and sent in response to the electrocardiogram analysis request, so that the electrocardiogram image analysis result information corresponding to an electrocardiogram image is obtained in real time on the first terminal device, and the electrocardiogram image analysis result information includes two pieces of information, one piece of information is a probability value of each heart disease in N preset heart diseases, N is a positive integer, and one piece of information is an internal feature visualization image.
With continued reference to fig. 4, a flow diagram 400 of one embodiment of a method for electrocardiographic image analysis in a medical environment is shown, in accordance with the present disclosure. The electrocardiogram image analysis method under the medical environment is applied to a server and comprises the following steps:
step 401, in response to receiving an electrocardiogram image analysis request generated based on an image to be analyzed and sent by a first terminal device, determining whether the image to be analyzed is an electrocardiogram image.
In response to the determination of being an electrocardiogram image, preprocessing the image to be analyzed to obtain an electrocardiogram waveform region image, step 402.
And step 403, inputting the electrocardiogram waveform area image into a pre-trained electrocardiogram analysis model to obtain a heart disease incidence probability prediction result corresponding to the electrocardiogram waveform area image.
Here, the heart disease incidence probability prediction result is used for representing the probability of suffering from each preset heart disease of N preset heart diseases, the electrocardiogram analysis model is used for representing the corresponding relation between the electrocardiogram waveform area image and the heart disease incidence probability prediction result, and N is a positive integer.
And step 404, reversely transmitting the obtained heart disease probability prediction result in the electrocardiogram analysis model to obtain a weight heat map.
And step 405, overlapping the weight heat map and the electrocardiogram waveform area image to obtain an internal feature visualization image.
And step 406, generating electrocardiogram image analysis result information based on the obtained heart disease probability prediction result and the internal feature visualized image.
Step 407, sending the electrocardiogram image analysis result information to the first terminal device.
In this embodiment, the specific operations of step 401, step 402, step 403, step 404, step 405, step 406, and step 407 and the technical effects thereof are substantially the same as the operations and effects of step 204, step 205, step 206, step 207, step 208, step 209, and step 210 in the embodiment shown in fig. 2A, and are not repeated herein.
In some alternative embodiments, step 402 may include steps 2051 through 2055 as shown in fig. 2B.
Here, the specific operations of the above-mentioned alternative embodiment of step 402 and the technical effects thereof are substantially the same as the operations and effects of the corresponding alternative embodiment of step 205 shown in fig. 2A, and are not described again here.
In some optional embodiments, the method flow 400 may further include the following steps 408 and 409:
in response to determining that it is not an electrocardiogram image, prompt information indicating that the image to be analyzed is not an electrocardiogram image is generated, step 408.
And step 409, sending the prompt information to the first terminal equipment.
Here, the specific operations of the above-mentioned alternative embodiments of step 408 and step 409 and the technical effects thereof are substantially the same as the operations and effects of the corresponding alternative embodiments of step 212 and step 213 shown in fig. 2A, and are not repeated herein.
According to the electrocardiogram analysis method under the medical environment provided by the embodiment of the disclosure, the electrocardiogram image analysis request sent by the first terminal device is processed on the server, and the electrocardiogram image analysis result information is fed back to the first terminal device, so that the first terminal device obtains the electrocardiogram image analysis result information corresponding to the image to be analyzed in real time, the electrocardiogram image analysis result information given by the server comprises the probability value of each heart disease in the preset N heart diseases, wherein N is a positive integer, and the electrocardiogram image analysis result information further comprises an internal characteristic visual image obtained by explaining the internal characteristics of the electrocardiogram image analysis model.
With further reference to fig. 5, as an implementation of the method shown in the above figures, the present disclosure provides an embodiment of an electrocardiogram image analysis apparatus in a medical environment, which corresponds to the embodiment of the method shown in fig. 3, and which is particularly applicable to various terminal devices.
As shown in fig. 5, the electrocardiogram image analyzing apparatus 500 of the present embodiment includes: an image acquisition unit 501, a request generation unit 502, a request transmission unit 503, and a result information presentation unit 504. Wherein, the image obtaining unit 501 is configured to obtain an image to be analyzed; a request generation unit 502 configured to generate an electrocardiogram image analysis request based on the image to be analyzed in response to detection of an electrocardiogram image analysis operation for the image to be analyzed; a request sending unit 503 configured to send the electrocardiographic image analysis request to a server, wherein the server inputs an electrocardiographic waveform area image obtained by preprocessing the image to be analyzed into a pre-trained electrocardiographic analysis model in response to determining that the image to be analyzed is an electrocardiographic image, obtains a heart disease prevalence prediction result for representing a probability of suffering from each of N preset heart diseases, where N is a positive integer, performs back propagation on the obtained heart disease prevalence prediction result in the electrocardiographic analysis model to obtain a weight heat map, superimposes the weight heat map and the electrocardiographic waveform area image to obtain an internal feature visualized image, and generates and returns electrocardiographic image analysis result information based on the obtained heart disease prevalence prediction result and the internal feature visualized image; a result information presentation unit 504 configured to present the electrocardiographic image analysis result information in response to receiving electrocardiographic image analysis result information transmitted by the server in response to the electrocardiographic image analysis request.
In this embodiment, the detailed processing and the technical effects of the image obtaining unit 501, the request generating unit 502, the request sending unit 503 and the result information presenting unit 504 of the electrocardiogram image analyzing apparatus 500 in the medical environment can respectively refer to the related descriptions of step 301, step 302, step 303 and step 304 in the corresponding embodiment of fig. 3, and are not repeated herein.
In some optional embodiments, the image acquiring unit 501 may be further configured to: in response to detecting a selected operation on a first image shot by a camera in the first terminal device, determining the first image as the image to be analyzed; or in response to detecting the selected operation of the second image in the local album of the first terminal equipment, determining the second image as the image to be analyzed.
In some optional embodiments, the apparatus 500 may further include: and a prompt information presentation unit (not shown in fig. 5) configured to present the prompt information in response to receiving the prompt information sent by the server for representing that the image to be analyzed is not an electrocardiogram image.
In some optional embodiments, the apparatus 500 may further include: an image presenting unit (not shown in fig. 5) configured to present, in response to receiving a further diagnosis request from a second terminal device forwarded via the server, the further diagnosis request including an image to be diagnosed and a cardiac disease probability prediction result and an internal feature visualization image corresponding to the image to be diagnosed, the image to be diagnosed and the cardiac disease probability prediction result and the internal feature visualization image corresponding to the image to be diagnosed; an interface presentation unit (not shown in fig. 5) configured to present an information input interface and receive diagnosis suggestion information input at the information input interface by a user; a diagnosis suggestion sending unit (not shown in fig. 5) configured to, in response to detecting a diagnosis suggestion confirmation operation for the diagnosis suggestion information, forward the diagnosis suggestion information to the second terminal device via the server, so that the second terminal device presents the diagnosis suggestion information.
It should be noted that, details of implementation and technical effects of the units in the electrocardiogram image analysis apparatus in the medical environment provided by the embodiment of the disclosure may refer to descriptions of other embodiments in the disclosure, and are not described herein again.
With further reference to fig. 6, as an implementation of the method shown in the above figures, the present disclosure provides an embodiment of an electrocardiogram image analyzing apparatus in a medical environment, which corresponds to the embodiment of the method shown in fig. 4, and which is particularly applicable to various electronic devices.
As shown in fig. 6, the electrocardiographic image analyzing apparatus 600 of the present embodiment in the medical environment includes: an image recognition unit 601, a waveform extraction unit 602, a probability prediction unit 603, a back propagation unit 604, an image generation unit 605, an analysis result generation unit 606, and an analysis result transmission unit 607. The image recognition unit 601 is configured to determine whether an image to be analyzed is an electrocardiogram image in response to receiving an electrocardiogram image analysis request generated based on the image to be analyzed and sent by a first terminal device; a waveform extraction unit 602 configured to, in response to determination of being an electrocardiogram image, preprocess the image to be analyzed to obtain an electrocardiogram waveform region image; a probability prediction unit 603 configured to input the electrocardiogram waveform area image into a pre-trained electrocardiogram analysis model, so as to obtain a heart disease probability prediction result corresponding to the electrocardiogram waveform area image, wherein the heart disease probability prediction result is used for representing the probability of suffering from each of N preset heart diseases, the electrocardiogram analysis model is used for representing the corresponding relationship between the electrocardiogram waveform area image and the heart disease probability prediction result, the electrocardiogram analysis model is a deep neural network, and N is a positive integer; a back propagation unit 604 configured to perform back propagation on the obtained prediction result of the heart disease prevalence probability in the electrocardiogram analysis model to obtain a weight heat map; an image generating unit 605 configured to obtain an internal feature visualized image by superimposing the weight heat map and the electrocardiogram waveform region image; an analysis result generation unit 606 configured to generate electrocardiogram image analysis result information based on the obtained heart disease prevalence probability prediction result and the internal feature visualized image; and an analysis result transmitting unit 607 configured to transmit the electrocardiogram image analysis result information to the first terminal device.
In this embodiment, for specific processing and technical effects of the image recognition unit 601, the waveform extraction unit 602, the probability prediction unit 603, the back propagation unit 604, the image generation unit 605, the analysis result generation unit 606 and the analysis result transmission unit 607 of the electrocardiogram image analysis apparatus 600 in the medical environment, reference may be made to relevant descriptions of step 401, step 402, step 403, step 404, step 405, step 406 and step 407 in the corresponding embodiment of fig. 4, which will not be described herein again.
In some optional embodiments, the waveform extraction unit 602 may be further configured to: performing waveform segmentation on the image to be analyzed to obtain a waveform segmentation result, wherein the waveform segmentation result is used for distinguishing a waveform from a background part in the image to be analyzed; generating a binary image corresponding to the image to be analyzed based on the waveform segmentation result; detecting an electrocardiogram waveform region in the binary image; intercepting an image in the binary image according to the electrocardiogram waveform area; an electrocardiogram waveform region image is generated based on the intercepted image.
In some optional embodiments, the apparatus 600 may further include: a prompt information generation unit (not shown in fig. 6) configured to generate prompt information indicating that the image to be analyzed is not an electrocardiogram image in response to a determination that it is not an electrocardiogram image; and a prompt information sending unit configured to send the prompt information to the first terminal device.
It should be noted that, details of implementation and technical effects of the units in the electrocardiogram image analysis apparatus in the medical environment provided by the embodiment of the disclosure may refer to descriptions of other embodiments in the disclosure, and are not described herein again.
Referring now to FIG. 7, shown is a block diagram of a computer system 700 suitable for use with a terminal device or server implementing embodiments of the present disclosure. The computer system 700 shown in fig. 7 is only an example and should not bring any limitations to the functionality or scope of use of the embodiments of the present disclosure.
As shown in fig. 7, computer system 700 may include a processing device (e.g., central processing unit, graphics processor, etc.) 701 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM) 702 or a program loaded from a storage device 708 into a Random Access Memory (RAM) 703. In the RAM 703, various programs and data necessary for the operation of the electronic apparatus 700 are also stored. The processing device 701, the ROM 702, and the RAM 703 are connected to each other by a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.
Generally, the following devices may be connected to the I/O interface 705: input devices 706 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, etc.; an output device 707 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 708 including, for example, magnetic tape, hard disk, etc.; and a communication device 709. The communications device 709 may allow the computer system 700 to communicate wirelessly or by wire with other devices to exchange data. While fig. 7 illustrates a computer system 700 having various means, it is to be understood that it is not required that all of the illustrated means be implemented or provided. More or fewer devices may alternatively be implemented or provided.
In particular, the processes described above with reference to the flow diagrams may be implemented as computer software programs, according to embodiments of the present disclosure. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such embodiments, the computer program may be downloaded and installed from a network via the communication means 709, or may be installed from the storage means 708, or may be installed from the ROM 702. The computer program, when executed by the processing device 701, performs the above-described functions defined in the methods of embodiments of the present disclosure.
It should be noted that the computer readable medium in the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present disclosure, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
The computer readable medium may be embodied in the electronic device; or may be separate and not incorporated into the electronic device.
The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to implement the method for analyzing an electrocardiogram image under medical environment as shown in the embodiment shown in fig. 3 and its optional embodiments, and/or the method for analyzing an electrocardiogram image under medical environment as shown in the embodiment shown in fig. 4 and its optional embodiments.
Computer program code for carrying out operations for aspects of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in the embodiments of the present disclosure may be implemented by software or hardware. The name of a unit does not in some cases constitute a limitation of the unit itself, and for example, an image acquisition unit may also be described as a "unit that acquires an image to be analyzed".
The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the disclosure herein is not limited to the particular combination of features described above, but also encompasses other embodiments in which any combination of the features described above or their equivalents does not depart from the spirit of the disclosure. For example, the above features and (but not limited to) the features disclosed in this disclosure having similar functions are replaced with each other to form the technical solution.

Claims (13)

1. An electrocardiogram image analysis method in a medical environment is applied to a first terminal device, and the method comprises the following steps:
acquiring an image to be analyzed;
in response to detecting an electrocardiogram image analysis operation for the image to be analyzed, generating an electrocardiogram image analysis request based on the image to be analyzed;
sending the electrocardiogram image analysis request to a server, wherein the server responds to the fact that the image to be analyzed is determined to be an electrocardiogram image, an electrocardiogram waveform area image obtained by preprocessing the image to be analyzed is input into a pre-trained electrocardiogram analysis model, a heart disease incidence probability prediction result used for representing the probability of suffering from each preset heart disease in N preset heart diseases is obtained, N is a positive integer, the obtained heart disease incidence probability prediction result is subjected to back propagation in the electrocardiogram analysis model to obtain a weight heat map, the weight heat map and the electrocardiogram waveform area image are superposed to obtain an internal feature visualized image, and electrocardiogram image analysis result information is generated and returned based on the obtained heart disease incidence prediction result and the internal feature visualized image;
presenting the electrocardiogram image analysis result information in response to receiving electrocardiogram image analysis result information transmitted by the server in response to the electrocardiogram image analysis request.
2. The method of claim 1, wherein the acquiring an image to be analyzed comprises:
in response to detecting a selected operation on a first image shot by a camera in the first terminal device, determining the first image as the image to be analyzed; or
And in response to detecting the selected operation of a second image in the local album of the first terminal equipment, determining the second image as the image to be analyzed.
3. The method of claim 1, wherein the method further comprises:
and presenting the prompt information in response to receiving the prompt information which is sent by the server and used for representing that the image to be analyzed is not an electrocardiogram image.
4. The method of claim 1, wherein the method further comprises:
in response to receiving a further diagnosis request from a second terminal device forwarded by the server, wherein the further diagnosis request comprises an image to be diagnosed and a heart disease probability prediction result and an internal feature visualization image corresponding to the image to be diagnosed, presenting the image to be diagnosed and the heart disease probability prediction result and the internal feature visualization image corresponding to the image to be diagnosed;
presenting an information input interface and receiving diagnosis suggestion information input by a user on the information input interface;
in response to detecting a diagnosis suggestion confirmation operation for the diagnosis suggestion information, forwarding the diagnosis suggestion information to the second terminal device via the server for the second terminal device to present the diagnosis suggestion information.
5. An electrocardiogram image analysis method in a medical environment is applied to a server, and the method comprises the following steps:
in response to receiving an electrocardiogram image analysis request which is sent by a first terminal device and generated based on an image to be analyzed, determining whether the image to be analyzed is an electrocardiogram image;
in response to the fact that the image to be analyzed is an electrocardiogram image, preprocessing the image to be analyzed to obtain an electrocardiogram waveform area image;
inputting the electrocardiogram waveform area image into a pre-trained electrocardiogram analysis model to obtain a heart disease incidence probability prediction result corresponding to the electrocardiogram waveform area image, wherein the heart disease incidence probability prediction result is used for representing the probability of suffering from each preset heart disease of N preset heart diseases, the electrocardiogram analysis model is used for representing the corresponding relation between the electrocardiogram waveform area image and the heart disease incidence probability prediction result, the electrocardiogram analysis model is a deep neural network, and N is a positive integer;
reversely transmitting the obtained heart disease probability prediction result in the electrocardiogram analysis model to obtain a weight thermal map;
superposing the weight heat map and the electrocardiogram waveform area image to obtain an internal feature visual image;
generating electrocardiogram image analysis result information based on the obtained heart disease probability prediction result and the internal characteristic visual image; and
and sending the electrocardiogram image analysis result information to the first terminal equipment.
6. The method of claim 5, wherein the preprocessing the image to be analyzed to obtain an electrocardiogram waveform region image comprises:
performing waveform segmentation on the image to be analyzed to obtain a waveform segmentation result, wherein the waveform segmentation result is used for distinguishing a waveform from a background part in the image to be analyzed;
generating a binary image corresponding to the image to be analyzed based on the waveform segmentation result;
detecting an electrocardiogram waveform region in the binary image;
intercepting an image in the binary image according to the electrocardiogram waveform region;
an electrocardiogram waveform region image is generated based on the intercepted image.
7. The method of claim 5, wherein the method further comprises:
in response to determining not to be an electrocardiogram image, generating prompt information indicating that the image to be analyzed is not an electrocardiogram image;
and sending the prompt information to the first terminal equipment.
8. An electrocardiogram image analysis device in a medical environment is applied to a first terminal device, and the device comprises:
an image acquisition unit configured to acquire an image to be analyzed;
a request generation unit configured to generate an electrocardiogram image analysis request based on the image to be analyzed in response to detection of an electrocardiogram image analysis operation for the image to be analyzed;
a request sending unit configured to send the electrocardiogram image analysis request to a server, wherein the server inputs an electrocardiogram waveform area image obtained by preprocessing the image to be analyzed into a pre-trained electrocardiogram analysis model in response to determining that the image to be analyzed is an electrocardiogram image, obtains a heart disease prevalence prediction result for representing a probability of suffering from each of N preset heart diseases, where N is a positive integer, performs back propagation on the obtained heart disease prevalence prediction result in the electrocardiogram analysis model to obtain a weight heat map, superimposes the weight heat map and the electrocardiogram waveform area image to obtain an internal feature visualized image, and generates and returns electrocardiogram image analysis result information based on the obtained heart disease prevalence prediction result and the internal feature visualized image;
a result information presentation unit configured to present the electrocardiogram image analysis result information in response to receiving the electrocardiogram image analysis result information transmitted by the server in response to the electrocardiogram image analysis request.
9. An electrocardiogram image analysis device in a medical environment, which is applied to a server, the device comprises:
an image recognition unit configured to determine whether an image to be analyzed is an electrocardiogram image in response to receiving an electrocardiogram image analysis request generated based on the image to be analyzed and transmitted by a first terminal device;
a waveform extraction unit configured to preprocess the image to be analyzed to obtain an electrocardiogram waveform region image in response to a determination that it is an electrocardiogram image;
a probability prediction unit configured to input the electrocardiogram waveform area image into a pre-trained electrocardiogram analysis model to obtain a heart disease incidence probability prediction result corresponding to the electrocardiogram waveform area image, wherein the heart disease incidence probability prediction result is used for representing the probability of suffering from each of N preset heart diseases, the electrocardiogram analysis model is used for representing the corresponding relation between the electrocardiogram waveform area image and the heart disease incidence probability prediction result, the electrocardiogram analysis model is a deep neural network, and N is a positive integer;
a back propagation unit configured to perform back propagation on the obtained heart disease probability prediction result in the electrocardiogram analysis model to obtain a weight heat map;
an image generation unit configured to superimpose the weight heat map and the electrocardiogram waveform region image to obtain an internal feature visualized image;
an analysis result generation unit configured to generate electrocardiogram image analysis result information based on the obtained heart disease probability prediction result and the internal feature visualized image; and
an analysis result transmitting unit configured to transmit the electrocardiogram image analysis result information to the first terminal device.
10. A terminal device, comprising:
one or more processors;
a storage device having one or more programs stored thereon,
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method recited in any of claims 1-4.
11. A server, comprising:
one or more processors;
a storage device having one or more programs stored thereon,
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of claims 5-7.
12. A computer readable storage medium having a computer program stored thereon, wherein the computer program, when executed by one or more processors, implements the method of any of claims 1-4 and/or the method of any of claims 5-7.
13. An electrocardiogram image analysis system in a medical environment, comprising a terminal device according to claim 10 and a server according to claim 11.
CN202211482855.3A 2022-11-24 2022-11-24 Electrocardiogram image analysis method and device in medical environment and terminal equipment Pending CN115579109A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211482855.3A CN115579109A (en) 2022-11-24 2022-11-24 Electrocardiogram image analysis method and device in medical environment and terminal equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211482855.3A CN115579109A (en) 2022-11-24 2022-11-24 Electrocardiogram image analysis method and device in medical environment and terminal equipment

Publications (1)

Publication Number Publication Date
CN115579109A true CN115579109A (en) 2023-01-06

Family

ID=84590586

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211482855.3A Pending CN115579109A (en) 2022-11-24 2022-11-24 Electrocardiogram image analysis method and device in medical environment and terminal equipment

Country Status (1)

Country Link
CN (1) CN115579109A (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107220644A (en) * 2017-04-18 2017-09-29 天津大学 A kind of ecg scanning image gradient bearing calibration
CN110504029A (en) * 2019-08-29 2019-11-26 腾讯医疗健康(深圳)有限公司 A kind of medical image processing method, medical image recognition method and device
CN110875092A (en) * 2018-08-31 2020-03-10 福州依影健康科技有限公司 Health big data service method and system based on remote fundus screening
CN111898622A (en) * 2019-05-05 2020-11-06 阿里巴巴集团控股有限公司 Information processing method, information display method, model training method, information display system, model training system and equipment
CN114648049A (en) * 2022-05-20 2022-06-21 合肥心之声健康科技有限公司 Method, device and system for constructing and classifying electrocardio image classification model
WO2022179645A2 (en) * 2022-06-13 2022-09-01 合肥心之声健康科技有限公司 Electrocardiogram analysis method and apparatus, electronic device and storage medium
WO2022220649A1 (en) * 2021-04-16 2022-10-20 서울대학교병원 System and method for electrocardiogram image-based patient evaluation
CN115272112A (en) * 2022-07-20 2022-11-01 清华大学 Paper electrocardiogram digitization method and device

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107220644A (en) * 2017-04-18 2017-09-29 天津大学 A kind of ecg scanning image gradient bearing calibration
CN110875092A (en) * 2018-08-31 2020-03-10 福州依影健康科技有限公司 Health big data service method and system based on remote fundus screening
CN111898622A (en) * 2019-05-05 2020-11-06 阿里巴巴集团控股有限公司 Information processing method, information display method, model training method, information display system, model training system and equipment
CN110504029A (en) * 2019-08-29 2019-11-26 腾讯医疗健康(深圳)有限公司 A kind of medical image processing method, medical image recognition method and device
WO2022220649A1 (en) * 2021-04-16 2022-10-20 서울대학교병원 System and method for electrocardiogram image-based patient evaluation
CN114648049A (en) * 2022-05-20 2022-06-21 合肥心之声健康科技有限公司 Method, device and system for constructing and classifying electrocardio image classification model
WO2022179645A2 (en) * 2022-06-13 2022-09-01 合肥心之声健康科技有限公司 Electrocardiogram analysis method and apparatus, electronic device and storage medium
CN115272112A (en) * 2022-07-20 2022-11-01 清华大学 Paper electrocardiogram digitization method and device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
张弘，李嘉锋, 西安电子科技大学出版社 *

Similar Documents

Publication Publication Date Title
CN107680684B (en) Method and device for acquiring information
US10937164B2 (en) Medical evaluation machine learning workflows and processes
CN110738263B (en) Image recognition model training method, image recognition method and image recognition device
CN110504029B (en) Medical image processing method, medical image identification method and medical image identification device
CN109887077B (en) Method and apparatus for generating three-dimensional model
CN107729929B (en) Method and device for acquiring information
US11900594B2 (en) Methods and systems for displaying a region of interest of a medical image
CN113177928B (en) Image identification method and device, electronic equipment and storage medium
CN108388889B (en) Method and device for analyzing face image
US20190147346A1 (en) Database systems and interactive user interfaces for dynamic conversational interactions
CN115994902A (en) Medical image analysis method, electronic device and storage medium
CN110517771B (en) Medical image processing method, medical image identification method and device
CN113360611A (en) AI diagnosis method, device, storage medium and equipment based on inspection result
CN115736939A (en) Atrial fibrillation disease probability generation method and device, electronic equipment and storage medium
Lampreave et al. Towards assisted electrocardiogram interpretation using an AI-enabled Augmented Reality headset
CN115579109A (en) Electrocardiogram image analysis method and device in medical environment and terminal equipment
CN115517686A (en) Family environment electrocardiogram image analysis method, device, equipment, medium and system
US20230238151A1 (en) Determining a medical professional having experience relevant to a medical procedure
US20230033263A1 (en) Information processing system, information processing method, information terminal, and non-transitory computer-readable medium
US20240062857A1 (en) Systems and methods for visualization of medical records
JEMAA et al. Digital Twin For A Human Heart Using Deep Learning and Stream Processing Platforms
RU2788482C2 (en) Training of neural network model
CN116784863A (en) Atrial fibrillation disease probability generation method, atrial fibrillation disease probability generation device, electronic equipment and storage medium
CN115376198A (en) Gaze direction estimation method, gaze direction estimation device, electronic apparatus, medium, and program product
Hayette Hadjar et al. Video-based emotion detection analyzing facial expressions and contactless vital signs for psychosomatic monitoring

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