WO2021189900A1 - 医学图像分析方法、装置、电子设备及可读存储介质 - Google Patents

医学图像分析方法、装置、电子设备及可读存储介质 Download PDF

Info

Publication number
WO2021189900A1
WO2021189900A1 PCT/CN2020/131977 CN2020131977W WO2021189900A1 WO 2021189900 A1 WO2021189900 A1 WO 2021189900A1 CN 2020131977 W CN2020131977 W CN 2020131977W WO 2021189900 A1 WO2021189900 A1 WO 2021189900A1
Authority
WO
WIPO (PCT)
Prior art keywords
image set
feature
cluster
index
image
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.)
Ceased
Application number
PCT/CN2020/131977
Other languages
English (en)
French (fr)
Inventor
章古月
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.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen 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 Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2021189900A1 publication Critical patent/WO2021189900A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/136Segmentation; Edge detection involving thresholding
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/23Clustering techniques
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/70Denoising; Smoothing
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • G06V10/44Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00ICT specially adapted for the handling or processing of medical images
    • G16H30/20ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10072Tomographic images
    • G06T2207/10081Computed x-ray tomography [CT]
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10072Tomographic images
    • G06T2207/10088Magnetic resonance imaging [MRI]

Definitions

  • This application relates to the field of image detection, and in particular to a medical image analysis method, device, electronic equipment, and computer-readable storage medium.
  • Medical image analysis is to diagnose whether there are lesions through medical images, and to quantify the severity of the lesions and so on. Automatically identifying the lesion area and normal tissues and organs in the image is the basic task of medical image analysis.
  • the inventor realizes that currently, medical image segmentation technology is generally used to identify the lesion area in the image.
  • the partitions are not obvious due to the mutual fit of parts of the lesion area, which will lead to inaccurate segmentation results and affect the quantitative results of the picture.
  • the existing image analysis does not have a unified judgment standard, which will Affect the analysis result of the image.
  • a medical image analysis method including:
  • preset parameter indexes compare and extract the target images in the target image set to obtain the extraction result, and quantify the extraction result to obtain the detection index set of the target object;
  • an analysis result of the medical image set is obtained.
  • a medical image analysis device includes:
  • An image acquisition module for acquiring a medical image set of a target object and acquiring a standard index set corresponding to the target object, and performing denoising processing on the medical image set to obtain a denoising image set;
  • An image processing module configured to perform feature extraction on the denoising image set to obtain a feature image set, and perform feature clustering on the feature image set to obtain a cluster feature image set;
  • the image cutting module cuts the feature image set according to the cluster feature image set to obtain a target image set
  • the image analysis module is used to compare and extract the target images in the target image group according to preset parameter indexes to obtain the extraction result, and quantify the extraction result to obtain the detection index set of the target object, according to the The index set and the standard index set are detected, and the analysis result of the medical image set is obtained.
  • An electronic device comprising: at least one processor; and,
  • a memory communicatively connected with the at least one processor; wherein,
  • the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute the following steps:
  • preset parameter indexes compare and extract the target images in the target image set to obtain the extraction result, and quantify the extraction result to obtain the detection index set of the target object;
  • an analysis result of the medical image set is obtained.
  • a computer-readable storage medium includes a storage data area and a storage program area, wherein the storage data area stores created data, and the storage program area stores a computer program; wherein, when the computer program is executed by a processor To achieve the following steps:
  • preset parameter indexes compare and extract the target images in the target image set to obtain the extraction result, and quantify the extraction result to obtain the detection index set of the target object;
  • an analysis result of the medical image set is obtained.
  • This application solves the problems of the medical image segmentation technology that the slice image partition is not clear and there is no standard for image data analysis.
  • FIG. 1 is a schematic flowchart of a medical image analysis method provided by an embodiment of the application
  • FIG. 2 is a schematic diagram of a detailed implementation flow of one of the steps in the medical image analysis method provided in FIG. 1;
  • FIG. 3 is a schematic diagram of modules of a medical image analysis device provided by an embodiment of the application.
  • FIG. 4 is a schematic diagram of the internal structure of an electronic device for implementing a medical image analysis method provided by an embodiment of the application;
  • the embodiment of the application provides a medical image analysis method.
  • the execution subject of the medical image analysis method includes, but is not limited to, a server, a terminal, etc., which can be configured to execute at least one of the electronic devices of the method provided in the embodiments of the present application.
  • the medical image analysis method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform.
  • the server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.
  • the medical image analysis method includes:
  • the target may include human tissues and organs such as the brain, lung, and heart.
  • the medical image set may include a computerized tomography (CT) image set, a magnetic resonance (MRI) image set, and the like.
  • the standard index set corresponding to the target refers to the index standard of healthy human tissues summarized according to a large number of medical studies.
  • the maximum inner diameter of the anterior horn of the lateral ventricle of the brain of a healthy human is about 0.3 compared with that of the upper skull.
  • the embodiment of the present application can use MySQL to construct a distributed blockchain node and connect to a pre-built database management system; through the database management system, the standard index set of the target is stored to the blockchain node When the target is detected, the target keyword information is obtained, and the standard index set of the target is obtained by querying the blockchain node.
  • the denoising processing is a process of removing noise on the medical image set.
  • performing denoising processing on a medical image set to obtain a denoising image includes:
  • Step a Perform salt and pepper noise filtering on the medical image set to obtain a semi-denoised image set
  • the salt and pepper noise is a kind of white or black spots randomly appearing in the image, which is caused by the sudden strong interference of the image signal. It is named after the salt and pepper, which can cause misjudgments in medical image analysis, especially in the brain. Analysis of fine organs.
  • the filtering is an operation of filtering out frequencies of a specific band in the medical image.
  • the local invention embodiment can use classical filtering, that is, according to Fourier transform and analysis, to split the medical image set into countless waves, and eliminate The salt and pepper wave is recombined.
  • a median filtering method can be used to eliminate the salt and pepper noise for filtering.
  • the median filtering method is a nonlinear digital filter technology.
  • Step b Perform Gaussian noise filtering on the semi-denoised image set to obtain the denoised image set.
  • the Gaussian noise refers to a type of noise whose probability density function obeys Gaussian distribution.
  • the embodiment of the present application can use the wavelet transform (wavelet transform, WT) method to filter the Gaussian noise appearing in the semi-denoised image.
  • WT wavelet transform
  • the wavelet transform inherits the idea of localization of the short-time Fourier transform, and at the same time overcomes the window size. It does not change with frequency and other shortcomings, which can provide a "time-frequency" window that changes with frequency. Using the time-frequency" window of the semi-denoised image can effectively remove the Gaussian noise in the semi-denoised image.
  • the embodiment of the present application may perform feature extraction on the denoised image through a convolutional neural network (CNN).
  • CNN convolutional neural network
  • the convolutional neural network is a type of feedforward neural network that includes convolution calculations and has a deep structure, and is used to extract image features.
  • performing feature extraction on the denoising image set to obtain a feature image set includes:
  • the denoising image set is convolved to obtain a feature mapping matrix set; the feature mapping matrix set is pooled and activated, and the feature image set is generated through a pre-built fully connected layer.
  • the clustering of the feature image set in the embodiment of the present application can ensure that the feature image set can maintain a clear boundary during the following cutting process.
  • performing feature clustering on the feature image set to obtain a cluster feature image set includes:
  • Step a Calculate the standard cluster center of the target object in each feature image in the feature image set to obtain a cluster center set
  • Step b Use a pre-built clustering algorithm to perform edge clustering on the feature image according to the cluster center set to obtain the cluster feature image
  • the step a includes:
  • the calculation method of the loss value is:
  • c i represents the data coordinates in the data cluster
  • x t is the cluster center coordinates of the data cluster
  • ⁇ i is the data coordinates other than x t in the data cluster
  • K is the number of data clusters.
  • the preset threshold may be set to 0.01.
  • the loss value when the loss value is greater than 0.01, it is determined that the cluster is unstable or does not converge, and the cluster center needs to be reselected.
  • the loss value is less than 0.01, it is determined that the clustering process has reached the desired result, and all cluster centers corresponding to the loss value less than 0.01 are retained to the cluster center set.
  • the edge clustering of the feature image may be performed by the K-Means clustering algorithm, so that the edge of the feature image is clearer.
  • the cluster image feature may be segmented by a threshold segmentation method to obtain a target image set.
  • S5. Perform comparison and extraction operations on the target images in the target image set according to the preset parameter index to obtain an extraction result, and quantify the extraction result to obtain a detection index set of the target object.
  • the embodiment of the present application takes brain tissue as an example, and the preset parameter indicators may include: Evans index, Ralphman index, third ventricle transverse diameter C, fourth ventricle transverse diameter D, and sulcus width F. in:
  • Evans index anterior horn of lateral ventricle A/maximum inner diameter of skull E;
  • the maximum inner diameter E of the skull As an example, in the embodiment of the present application, all the target images in the target image set are compared, the target image with the largest cranial cavity diameter is extracted, and the target image with the largest cranial cavity diameter is quantified, which is convenient The maximum cranial cavity diameter is then used to calculate the detection index.
  • the analysis result of the medical image set is obtained by comparing the detection index set with the standard index set.
  • the S6 includes:
  • the embodiment of the present application takes the above-mentioned Evans index of brain tissue as an example.
  • the detected Evans index is X Evans and the standard Evans index is Y Evans
  • the difference between the detected Evans index and the standard Evans index is calculated as
  • the detection index when the difference between the detection index of the tissue structure and the standard index is less than or equal to a prescribed threshold, the detection index is determined to be a normal detection index, and if the difference is greater than the threshold, it is determined
  • the detection index is an abnormal detection index.
  • X Evans -Y Evans 0.1, the preset threshold is 0.05, and 0.1 is greater than 0.05, indicating that there may be hydrocephalus in the medical image .
  • the embodiment of the application performs denoising processing on a medical image set to obtain a denoised image set, eliminating the noise factor in the image, performing feature extraction on the denoising image set to obtain a feature image set, and combining the feature image set Perform feature clustering.
  • Feature clustering can make the partition between the target and non-target more obvious and facilitate the subsequent cutting process.
  • the preset parameter index the target image is compared, extracted and quantified in the target image.
  • the detection index set is compared with the standard index set, and the analysis result is obtained, so that the analysis result standard is unified, and the time for medical staff to diagnose the disease can be reduced, and the efficiency can be improved.
  • FIG. 3 it is a schematic diagram of the modules of the medical image analysis device of the present application.
  • the medical image analysis device 100 described in this application can be installed in an electronic device. According to the implemented functions, the medical image analysis device 100 may include an image acquisition module 101, an image processing module 102, a cutting module 103, and an image analysis module 104.
  • the module described in the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can complete fixed functions, and are stored in the memory of the electronic device.
  • each module/unit is as follows:
  • the image acquisition module 101 is configured to acquire a medical image set of a target object, and perform denoising processing on the medical image set to obtain a denoising image set.
  • the target may include human tissues and organs such as the brain, lung, and heart.
  • the medical image set may include a computerized tomography (CT) image set, a magnetic resonance (MRI) image set, and the like.
  • the standard index set corresponding to the target refers to the index standard of healthy human tissues summarized according to a large number of medical researches.
  • the anterior horn of the lateral ventricle of a healthy human brain is about 0.3 of the maximum inner diameter of the upper skull.
  • the image acquisition module 101 described in the embodiment of the present application can use MySQL to construct a distributed blockchain node and connect to a pre-built database management system; through the database management system, the standard index set of the target object is stored to The blockchain node; when the target is detected, the target keyword information is obtained, and the standard index set of the target is obtained by querying the blockchain node.
  • the denoising processing is a process of removing noise on the medical image set.
  • the image acquisition module 101 performs denoising processing on a medical image set to obtain a denoised image, including:
  • Step a Perform salt and pepper noise filtering on the medical image set to obtain a semi-denoised image set
  • the salt and pepper noise is a kind of white or black spots randomly appearing in the image, which is caused by the sudden strong interference of the image signal. It is named after the salt and pepper, which can cause misjudgments in medical image analysis, especially in the brain. Analysis of fine organs.
  • the filtering is an operation of filtering out frequencies of a specific band in the medical image.
  • the local invention embodiment can use classical filtering, that is, according to Fourier transform and analysis, to split the medical image set into countless waves, and eliminate The salt and pepper wave is recombined.
  • the image acquisition module 101 of the preferred embodiment of the present application may use a median filtering method to eliminate the salt and pepper noise for filtering.
  • the median filtering method is a non-linear digital filter technology.
  • Step b Perform Gaussian noise filtering on the semi-denoised image set to obtain the denoised image set.
  • the Gaussian noise refers to a type of noise whose probability density function obeys Gaussian distribution.
  • the image acquisition module 101 described in the embodiment of the present application can use a wavelet transform (wavelet transform, WT) method to filter the Gaussian noise appearing in the semi-denoised image.
  • WT wavelet transform
  • the wavelet transform inherits the idea of localization of the short-time Fourier transform.
  • it overcomes the shortcomings that the window size does not change with frequency, and can provide a "time-frequency" window that changes with frequency.
  • the semi-denoised image can effectively remove the inside of the semi-denoised image. Gaussian noise.
  • the image processing module 102 is configured to perform feature extraction on the denoising image set to obtain a feature image set, and to converge the features of the target object according to the feature image set to obtain a convergent feature image set.
  • the image processing module 102 described in the embodiment of the present application may perform feature extraction on the denoised image through a convolutional neural network (CNN).
  • CNN convolutional neural network
  • the convolutional neural network is a type of feedforward neural network that includes convolution calculations and has a deep structure, and is used to extract image features.
  • the image processing module 102 performs feature extraction on the denoising image set to obtain a feature image set, including:
  • the denoising image set is convolved to obtain a feature mapping matrix set; the feature mapping matrix set is pooled and activated, and the feature image set is generated through a pre-built fully connected layer.
  • the image processing module 102 converges the features of the target object according to the feature image set to obtain a convergent feature image set, including:
  • Step a Calculate the standard cluster center of the target object in each feature image in the feature image set to obtain a cluster center set
  • Step b Use a pre-built clustering algorithm to perform edge clustering on the feature image according to the cluster center set to obtain the cluster feature image
  • the step a includes:
  • the calculation method of the loss value is:
  • c i represents the data coordinates in the data cluster
  • x t is the cluster center coordinates of the data cluster
  • ⁇ i is the data coordinates other than x t in the data cluster
  • K is the number of data clusters.
  • the preset threshold may be set to 0.01.
  • the image processing module 102 determines that the cluster is unstable or does not converge, and the cluster center needs to be reselected; when the loss value is less than 0.01, the image processing module 102 determines that the clustering process has reached the desired result, and reserves all cluster centers corresponding to the loss value less than 0.01 to the cluster center set.
  • the image processing module 102 described in the embodiment of the present application may perform edge clustering on the feature image through the K-Means clustering algorithm, so that the edge of the feature image is clearer.
  • the cutting module 103 is configured to cut the target object according to the convergent feature image set to obtain a target image set;
  • the cutting module 103 described in the embodiment of the present application may segment the cluster image features by a threshold segmentation method to obtain a target image set.
  • the image analysis module 104 is configured to perform comparison and extraction operations on the target images in the target images according to preset parameter indexes to obtain the extraction result, and quantify the extraction result to obtain the target image A detection index set, and an analysis result of the medical image set is obtained according to the detection index set and the standard index set.
  • the embodiment of the present application takes brain tissue as an example, and the preset parameter indicators may include: Evans index, Ralphman index, third ventricle transverse diameter C, fourth ventricle transverse diameter D, and sulcus width F. in:
  • Evans index anterior horn of lateral ventricle A/maximum inner diameter of skull E;
  • the image analysis module 104 compares all target images in the target image set, and extracts the target image with the largest cranial cavity diameter and the one with the largest cranial cavity diameter.
  • the target image is quantified to facilitate the subsequent calculation of the detection index using the maximum cranial cavity diameter.
  • the image analysis module 104 in this embodiment of the application obtains the analysis result of the medical image set according to the detection index set and the standard index set.
  • the image analysis module 104 obtains the analysis result of the medical image set by comparing the detection index set with the standard index set.
  • the image analysis module 104 obtains the analysis result of the medical image set according to the detection index set and the standard index set, including:
  • the embodiment of the application takes the above-mentioned Evans index of brain tissue as an example.
  • the detected Evans index is X Evans and the standard Evans index is Y Evans
  • the difference between the detected Evans index and the standard Evans index is calculated as
  • the image analysis module 104 determines that the detection index is normal, and when the difference value is not within the preset threshold value range, the image analysis module 104 The module 104 determines that the detection index is abnormal.
  • the image analysis module 104 determines that the detection index is normal, and if the difference is greater than the Threshold, the image analysis module 104 determines that the detection index is abnormal.
  • the preset threshold value is 0.05, and 0.1 is greater than 0.05, then the medical image indicates that the brain may appear brain Stagnant water.
  • the image analysis module 104 uses all normal detection indicators to construct a normal index set, and uses all abnormal detection indicators to construct an abnormal index set. According to the normal index set and the abnormal index set, Get the analysis result.
  • the image analysis module 104 may also only integrate the abnormal detection indicators, construct an abnormal indicator set, summarize the corresponding problems of each abnormal indicator, and obtain the analysis result.
  • FIG. 4 it is a schematic diagram of the structure of an electronic device implementing the medical image analysis method of the present application.
  • the electronic device 1 may include a processor 10, a memory 11, and a bus, and may also include a computer program stored in the memory 11 and running on the processor 10, such as a medical image analysis program 12.
  • the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, CD etc.
  • the memory 11 may be an internal storage unit of the electronic device 1 in some embodiments, for example, a mobile hard disk of the electronic device 1.
  • the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart media card (SMC), and a secure digital (Secure Digital) equipped on the electronic device 1. , SD) card, flash card (Flash Card), etc.
  • the memory 11 may also include both an internal storage unit of the electronic device 1 and an external storage device.
  • the memory 11 can be used not only to store application software and various data installed in the electronic device 1, such as the code of the medical image analysis program 12, etc., but also to temporarily store data that has been output or will be output.
  • the processor 10 may be composed of integrated circuits in some embodiments, for example, may be composed of a single packaged integrated circuit, or may be composed of multiple integrated circuits with the same function or different functions, including one or more Combinations of central processing unit (CPU), microprocessor, digital processing chip, graphics processor, and various control chips, etc.
  • the processor 10 is the control unit of the electronic device, which uses various interfaces and lines to connect the various components of the entire electronic device, and runs or executes programs or modules stored in the memory 11 (for example, executing Medical image analysis programs, etc.), and call data stored in the memory 11 to execute various functions of the electronic device 1 and process data.
  • the bus may be a peripheral component interconnect standard (PCI) bus or an extended industry standard architecture (EISA) bus, etc.
  • PCI peripheral component interconnect standard
  • EISA extended industry standard architecture
  • the bus can be divided into address bus, data bus, control bus and so on.
  • the bus is configured to implement connection and communication between the memory 11 and at least one processor 10 and the like.
  • FIG. 4 only shows an electronic device with components. Those skilled in the art can understand that the structure shown in FIG. 4 does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure. Components, or a combination of certain components, or different component arrangements.
  • the electronic device 1 may also include a power source (such as a battery) for supplying power to various components.
  • the power source may be logically connected to the at least one processor 10 through a power management device, thereby controlling power
  • the device implements functions such as charge management, discharge management, and power consumption management.
  • the power supply may also include one or more DC or AC power supplies, recharging devices, power failure detection circuits, power converters or inverters, power supply status indicators and other arbitrary components.
  • the electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.
  • the electronic device 1 may also include a network interface.
  • the network interface may include a wired interface and/or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is usually used in the electronic device 1 Establish a communication connection with other electronic devices.
  • the electronic device 1 may also include a user interface.
  • the user interface may be a display (Display) and an input unit (such as a keyboard (Keyboard)).
  • the user interface may also be a standard wired interface or a wireless interface.
  • the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode, organic light-emitting diode) touch device, etc.
  • the display can also be appropriately called a display screen or a display unit, which is used to display the information processed in the electronic device 1 and to display a visualized user interface.
  • the medical image analysis program 12 stored in the memory 11 in the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can realize:
  • preset parameter indexes compare and extract the target images in the target image set to obtain the extraction result, and quantify the extraction result to obtain the detection index set of the target object;
  • an analysis result of the medical image set is obtained.
  • the integrated module/unit of the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the computer-readable storage medium may be volatile or non-volatile, and the computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, Mobile hard disks, magnetic disks, optical disks, computer memory, read-only memory (ROM, Read-Only Memory).
  • the computer usable storage medium may mainly include a storage program area and a storage data area, where the storage program area may store an operating system, a computer program required by at least one function, etc.; the storage data area may store a block chain node When the computer program is executed by the processor, it can realize:
  • preset parameter indexes compare and extract the target images in the target image set to obtain the extraction result, and quantify the extraction result to obtain the detection index set of the target object;
  • an analysis result of the medical image set is obtained.
  • all the above-mentioned data can also be stored in a node of a blockchain.
  • a node of a blockchain For example, medical image sets, feature image sets or target image sets, etc., these data can all be stored in the blockchain node.
  • modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • the functional modules in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit may be implemented in the form of hardware, or may be implemented in the form of hardware plus software functional modules.
  • the blockchain referred to in this application is a new application mode of computer technology such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm.
  • Blockchain essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information for verification. The validity of the information (anti-counterfeiting) and the generation of the next block.
  • the blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Radiology & Medical Imaging (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Multimedia (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Primary Health Care (AREA)
  • Public Health (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Image Analysis (AREA)

Abstract

一种医学图像分析方法、医学图像分析装置、设备及存储介质,该方法包括:获取目标物的医学图像集和标准指标集,对所述医学图像集去噪,得到去噪图像集(S1);从去噪图像集中提取特征,得到特征图像集(S2);对特征图像集进行特征聚类,得到聚类特征图像集(S3);根据聚类特征图像集对所述特征图像集进行切割,得到目标图像集(S4);将目标图像集中目标物进行对比、提取并量化所述提取的结果,得到所述目标物的检测指标集(S5);根据所述检测指标集与所述标准指标集得到分析结果(S6)。所述标准指标集可以存储在区块链节点中。该方法解决了切片图像分区不明确、图像数据分析没有标准的问题。

Description

医学图像分析方法、装置、电子设备及可读存储介质
本申请要求于2020年10月14日提交中国专利局、申请号为CN202011096338.3,发明名称为“医学图像分析方法、装置、电子设备及可读存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及图像检测领域,尤其涉及一种医学图像分析方法、装置、电子设备及计算机可读存储介质。
背景技术
随着医学成像技术和计算机技术的不断发展和进步,医学图像分析已成为医学研究、临床疾病诊断和治疗中一个不可或缺的工具和技术手段。医学图像分析是通过医学图像诊断是否有病灶、对病灶的轻重程度进行量化分级等。自动识别图像中的病灶区域和正常组织器官区域是医学图像分析的基本任务。
发明人意识到目前通常采用医学图像分割技术识别图像中的病灶区域。然而,在执行图像分割时,由于病灶区域部分位置相互贴合导致分区不明显,会导致分割结果不精确,影响图片的量化结果,此外,现有的图像分析并没有统一的评判标准,从而会影响图像的分析结果。
发明内容
一种医学图像分析方法,包括:
获取目标物的医学图像集及获取所述目标物对应的标准指标集,对所述医学图像集进行去噪处理,得到去噪图像集;
对所述去噪图像集进行特征提取,得到特征图像集;
将所述特征图像集进行特征聚类,得到聚类特征图像集;
根据所述聚类特征图像集,对所述特征图像集进行切割,得到目标图像集;
按照预设参数指标,将所述目标图像集中目标图像进行对比及提取操作,得到提取结果,并量化所述提取结果,得到所述目标物的检测指标集;
根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
一种医学图像分析装置,所述装置包括:
图像获取模块,用于获取目标物的医学图像集及获取所述目标物对应的标准指标集,对所述医学图像集进行去噪处理,得到去噪图像集;
图像处理模块,用于对所述去噪图像集进行特征提取,得到特征图像集,将所述特征图像集进行特征聚类,得到聚类特征图像集;
图像切割模块根据所述聚类特征图像集,对所述特征图像集进行切割,得到目标图像集;
图像分析模块,用于按照预设参数指标,将所述目标图像集中目标图像进行对比及提取操作,得到提取结果,并量化所述提取结果,得到所述目标物的检测指标集,根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
一种电子设备,所述电子设备包括:至少一个处理器;以及,
与所述至少一个处理器通信连接的存储器;其中,
所述存储器存储有可被所述至少一个处理器执行的指令,所述指令被所述至少一个处理器执行,以使所述至少一个处理器能够执行如下步骤:
获取目标物的医学图像集及获取所述目标物对应的标准指标集,对所述医学图像集进 行去噪处理,得到去噪图像集;
对所述去噪图像集进行特征提取,得到特征图像集;
将所述特征图像集进行特征聚类,得到聚类特征图像集;
根据所述聚类特征图像集,对所述特征图像集进行切割,得到目标图像集;
按照预设参数指标,将所述目标图像集中目标图像进行对比及提取操作,得到提取结果,并量化所述提取结果,得到所述目标物的检测指标集;
根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
一种计算机可读存储介质,包括存储数据区和存储程序区,其中,所述存储数据区存储创建的数据,所述存储程序区存储有计算机程序;其中,所述计算机程序被处理器执行时实现如下步骤:
获取目标物的医学图像集及获取所述目标物对应的标准指标集,对所述医学图像集进行去噪处理,得到去噪图像集;
对所述去噪图像集进行特征提取,得到特征图像集;
将所述特征图像集进行特征聚类,得到聚类特征图像集;
根据所述聚类特征图像集,对所述特征图像集进行切割,得到目标图像集;
按照预设参数指标,将所述目标图像集中目标图像进行对比及提取操作,得到提取结果,并量化所述提取结果,得到所述目标物的检测指标集;
根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
本申请解决了医学图像分割技术存在的切片图像分区不明确、图像数据分析没有标准的问题。
附图说明
图1为本申请一实施例提供的医学图像分析方法的流程示意图;
图2为图1提供的医学图像分析方法中其中一个步骤的详细实施流程示意图;
图3为本申请一实施例提供的医学图像分析装置的模块示意图;
图4为本申请一实施例提供的实现医学图像分析方法的电子设备的内部结构示意图;
本申请目的的实现、功能特点及优点将结合实施例,参照附图做进一步说明。
具体实施方式
应当理解,此处所描述的具体实施例仅仅用以解释本申请,并不用于限定本申请。
本申请实施例提供一种医学图像分析方法。所述医学图像分析方法的执行主体包括但不限于服务端、终端等能够被配置为执行本申请实施例提供的该方法的电子设备中的至少一种。换言之,所述医学图像分析方法可以由安装在终端设备或服务端设备的软件或硬件来执行,所述软件可以是区块链平台。所述服务端包括但不限于:单台服务器、服务器集群、云端服务器或云端服务器集群等。
参照图1所示的本申请实施例提供的医学图像分析方法的流程示意图,所述医学图像分析方法包括:
S1、获取目标物的医学图像集及获取所述目标物对应的标准指标集,对所述医学图像集进行去噪处理,得到去噪图像集。
本申请实施例中,所述目标物可以包括:大脑、肺、心脏等人体组织器官。所述医学图像集可以包括电子计算机断层(简称CT)图像集、磁共振(MRI)图像集等。
所述目标物对应的标准指标集是指依据大量的医学研究总结出来的健康的人体组织的指标标准,例如,健康人体大脑的侧脑室前角比上颅骨最大内径约为0.3。
进一步地,本申请实施例可以利用MySQL构建分布式的区块链节点并连接预构建的数据库管理系统;通过所述数据库管理系统,存储所述目标物的标准指标集至所述区块链 节点;当检测所述目标物时,获取目标物关键字信息,从所述区块链节点查询得到所述目标物的标准指标集。
本申请实施例中,所述去噪处理是将所述医学图像集上的噪声除去的过程。详细地,本申请较佳实施例中,所述对医学图像集进行去噪处理,得到去噪图像,包括:
步骤a、对所述医学图像集进行椒盐噪声滤波,得到半消噪图像集;
所述椒盐噪声为一种图像中随机出现的白点或者黑点,是影像讯号受到突如其来的强烈干扰而产生,相似椒盐而得名,在医学图像分析中会产生误判,尤其是在大脑等精细器官的分析中。
所述滤波为将所述医学图像中特定波段频率滤除的操作,本地发明实施例可以通过经典滤波,即根据傅里叶变换与分析,将所述医学图像集拆分成无数的波,剔除椒盐波再重新组合。
本申请较佳实施例可以利用中值滤波方法消掉所述椒盐噪声进行滤波,所述中值滤波方法一种非线性数字滤波器技术。
步骤b、对所述半消噪图像集进行高斯噪声滤波,得到所述去噪图像集。
所述高斯噪声是指概率密度函数服从高斯分布的一类噪声。
优选地,本申请实施例可采用小波变换(wavelet transform,WT)法对半消噪图像中出现的高斯噪声进行滤波,小波变换是继承短时傅立叶变换局部化的思想,同时又克服了窗口大小不随频率变化等缺点,能够提供一个随频率改变的“时间-频率”窗口,利用该时间-频率”窗口所述半消噪图像,可有效去除所述半消噪图像内的高斯噪声。
S2、对所述去噪图像集进行特征提取,得到特征图像集。
本申请实施例可以通过卷积神经网络(CNN)对所述去噪图像进行特征提取。所述卷积神经网络(Convolutional Neural Networks,CNN)是一类包含卷积计算且具有深度结构的前馈神经网络,用于提取图像特征。
详细地,本申请实施例中,所述对所述去噪图像集进行特征提取,得到特征图像集,包括:
将所述去噪图像集进行卷积,得到特征映射矩阵集;将所述特征映射矩阵集进行池化及激活操作,通过预构建的全连接层,生成所述特征图像集。
S3、将所述特征图像集进行特征聚类,得到聚类特征图像集。
本申请实施例对所述特征图像集进行聚类,可以保证所述特征图像集在下述切割过程中能保持边界清晰。
详细地,本申请实施例中,所述将所述特征图像集进行特征聚类,得到聚类特征图像集,包括:
步骤a、计算所述特征图像集中每一张特征图像所述目标物的标准簇心,得到簇心集;
步骤b、根据所述簇心集,利用预构建的聚类算法,对所述特征图像进行边缘聚类,得到所述聚类特征图像
进一步的,本申请实施例中,所述步骤a,包括:
将所述特征图像中所述目标物作为一个簇,初始化所述簇,随机得到所述簇对应的初始簇心;
计算所述簇的损失值,比较所述损失值与预设阈值,当所述损失值大于所述阈值时,重新初始化所述初始簇心,当所述损失值小于或等于所述阈值时,确定所述初始簇心为标准簇心。
其中,所述损失值的计算方法为:
Figure PCTCN2020131977-appb-000001
其中,c i表示数据簇中的数据坐标,x t为所述数据簇的簇心坐标,μ i为所述数据簇内除x t之外的其他数据坐标,K为数据簇的数量。
本申请其中一个实施例中,所述预设阈值可以设定为0.01。
本申请实施例中,当所述损失值大于0.01,判定为所述簇不稳定或不收敛,需要重新选择簇心。当所述损失值小于0.01,判定为聚类过程已经达到期望的结果,保留所有所述小于0.01的损失值对应的簇心至所述簇心集。
进一步地,本申请实施例可以通过K-Means聚类算法对所述特征图像进行边缘聚类,使得所述特征图像的边缘更加清晰。
S4、根据所述聚类特征图像集,对所述特征图像集进行切割,得到目标图像集。
本申请实施例可以通过阈值分割法对所述聚类图像特征进行分割,得到目标图像集。
S5、按照预设参数指标,将所述目标图像集中目标图像进行对比及提取操作,得到提取结果,并量化所述提取结果,得到所述目标物的检测指标集。
本申请实施例以大脑组织为例,则所述预设参数指标可以包括:Evans指数、Hackman指数及第三脑室横径C、第四脑室横径D、脑沟宽度F。其中:
Evans指数=侧脑室前角A/颅骨最大内径E;
Hackman指数=侧脑室前角A+尾状核B;
进一步以测量颅骨最大内径E为例,本申请实施例中,将所述目标图像集中所有目标图像进行对比,提取出颅腔直径最大的目标图像并对所述颅腔直径最大的目标图像进行量化,方便后续利用所述最大颅腔直径进行运算所述检测指标。
S6、根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
本申请实施例通过将所述检测指标集与所述标准指标集进行比较,得到所述医学图像集的分析结果。
详细地,如图2所示,本申请实施例中,所述S6,包括:
S61、将所述检测指标集中每一个检测指标与所述标准指标集中对应的标准指标进行对比,并计算所述检测指标与对应的所述标准指标的差值。
本申请实施例以大脑组织的上述Evans指数为例,当检测Evans指数为X Evans时,标准Evans指数为Y Evans时,计算所述检测Evans指数与标准Evans指数的差值为|X Evans-Y Evans|。
S62、当所述差值大于预设阈值时,判定所述检测指标为异常指标。
本申请实施例中,当所述组织结构的检测指标与所述标准指标的差值小于或等于规定阈值,判定所述检测指标为正常的检测指标,若所述差值大于所述阈值,判定所述检测指标为异常的检测指标。以上述Evans指数为例,当X Evans=0.2,Y Evans=0.3,则|X Evans-Y Evans=0.1,预设阈值为0.05,0.1大于0.05,则表明所述医学图像中可能存在大脑积水。
S63、根据所有异常的检测指标构建异常指标集,得到所述分析结果
利用所有正常的检测指标构建的正常指标集,及利用所有异常的检测指标构建的异常指标集,根据所述正常指标集及所述异常指标集,得到分析结果。
本申请其他实施例中,也可以只整合所述出现异常的检测指标,构建异常指标集,总结各个所述异常指标的对应问题,可以得到分析结果。
本申请实施例对医学图像集进行去噪处理,得到去噪图像集,消除了图像中的噪音因素,对所述去噪图像集进行特征提取,得到特征图像集,并将所述特征图像集进行特征聚类,特征聚类可以使得目标物与非目标物间的分区更加明显方便后续切割过程,按照预设参数指标,将所述目标图像集中目标图像进行对比、提取并量化所述提取的结果,并将检测指标集与标准指标集进行比较,得到分析结果,使得分析结果标准统一,并可以减少医护人员的对疾病的诊断时间,提高效率。
如图3所示,是本申请医学图像分析装置的模块示意图。
本申请所述医学图像分析装置100可以安装于电子设备中。根据实现的功能,所述医学图像分析装置100可以包括图像获取模块101、图像处理模块102、切割模块103及图像分析模块104。本发所述模块也可以称之为单元,是指一种能够被电子设备处理器所执行,并且能够完成固定功能的一系列计算机程序段,其存储在电子设备的存储器中。
在本实施例中,关于各模块/单元的功能如下:
所述图像获取模块101,用于获取目标物的医学图像集,对所述医学图像集进行去噪处理,得到去噪图像集。
本申请实施例中,所述目标物可以包括:大脑、肺、心脏等人体组织器官。所述医学图像集可以包括电子计算机断层(简称CT)图像集、磁共振(MRI)图像集等。
所述目标物对应的标准指标集是指依据大量的医学研究总结出来的健康的人体组织的指标标准,例如,健康人体大脑的侧脑室前角比上颅骨最大内径约为0.3。
进一步地,本申请实施例所述图像获取模块101可以利用MySQL构建分布式的区块链节点并连接预构建的数据库管理系统;通过所述数据库管理系统,存储所述目标物的标准指标集至所述区块链节点;当检测所述目标物时,获取目标物关键字信息,从所述区块链节点查询得到所述目标物的标准指标集。
本申请实施例中,所述去噪处理是将所述医学图像集上的噪声除去的过程。详细地,本申请较佳实施例中,所述图像获取模块101对医学图像集进行去噪处理,得到去噪图像,包括:
步骤a、对所述医学图像集进行椒盐噪声滤波,得到半消噪图像集;
所述椒盐噪声为一种图像中随机出现的白点或者黑点,是影像讯号受到突如其来的强烈干扰而产生,相似椒盐而得名,在医学图像分析中会产生误判,尤其是在大脑等精细器官的分析中。
所述滤波为将所述医学图像中特定波段频率滤除的操作,本地发明实施例可以通过经典滤波,即根据傅里叶变换与分析,将所述医学图像集拆分成无数的波,剔除椒盐波再重新组合。
本申请较佳实施例所述图像获取模块101可以利用中值滤波方法消掉所述椒盐噪声进行滤波,所述中值滤波方法一种非线性数字滤波器技术。
步骤b、对所述半消噪图像集进行高斯噪声滤波,得到所述去噪图像集。
所述高斯噪声是指概率密度函数服从高斯分布的一类噪声。
优选地,本申请实施例所述图像获取模块101可采用小波变换(wavelet transform,WT)法对半消噪图像中出现的高斯噪声进行滤波,小波变换是继承短时傅立叶变换局部化的思想,同时又克服了窗口大小不随频率变化等缺点,能够提供一个随频率改变的“时间-频率”窗口,利用该时间-频率”窗口所述半消噪图像,可有效去除所述半消噪图像内的高斯噪声。
所述图像处理模块102,用于对所述去噪图像集进行特征提取,得到特征图像集,及根据所述特征图像集对所述目标物的特征进行收敛,得到收敛特征图像集。
本申请实施例所述图像处理模块102可以通过卷积神经网络(CNN)对所述去噪图像进行特征提取。所述卷积神经网络(Convolutional Neural Networks,CNN)是一类包含卷积计算且具有深度结构的前馈神经网络,用于提取图像特征。
详细地,本申请实施例中,所述图像处理模块102对所述去噪图像集进行特征提取,得到特征图像集,包括:
将所述去噪图像集进行卷积,得到特征映射矩阵集;将所述特征映射矩阵集进行池化及激活操作,通过预构建的全连接层,生成所述特征图像集。
详细地,本申请实施例中,所述图像处理模块102根据所述特征图像集对所述目标物 的特征进行收敛,得到收敛特征图像集,包括:
步骤a、计算所述特征图像集中每一张特征图像所述目标物的标准簇心,得到簇心集;
步骤b、根据所述簇心集,利用预构建的聚类算法,对所述特征图像进行边缘聚类,得到所述聚类特征图像
进一步的,本申请实施例中,所述步骤a,包括:
将所述特征图像中所述目标物作为一个簇,初始化所述簇,随机得到所述簇对应的初始簇心;
计算所述簇的损失值,比较所述损失值与预设阈值,当所述损失值大于所述阈值时,重新初始化所述初始簇心,当所述损失值小于或等于所述阈值时,确定所述初始簇心为标准簇心。
其中,所述损失值的计算方法为:
Figure PCTCN2020131977-appb-000002
其中,c i表示数据簇中的数据坐标,x t为所述数据簇的簇心坐标,μ i为所述数据簇内除x t之外的其他数据坐标,K为数据簇的数量。
本申请其中一个实施例中,所述预设阈值可以设定为0.01。
本申请实施例中,当所述损失值大于0.01,所述图像处理模块102判定为所述簇不稳定或不收敛,需要重新选择簇心;当所述损失值小于0.01,所述图像处理模块102判定为聚类过程已经达到期望的结果,保留所有所述小于0.01的损失值对应的簇心至所述簇心集。
进一步地,本申请实施例所述图像处理模块102可以通过K-Means聚类算法对所述特征图像进行边缘聚类,使得所述特征图像的边缘更加清晰。
所述切割模块103,用于根据所述收敛特征图像集对所述目标物进行切割,得到目标图像集;
本申请实施例所述切割模块103可以通过阈值分割法对所述聚类图像特征进行分割,得到目标图像集。
所述图像分析模块104,用于对所述特征按照预设参数指标,将所述目标图像集中目标图像进行对比及提取操作,得到提取结果,并量化所述提取结果,得到所述目标物的检测指标集,及根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
本申请实施例以大脑组织为例,则所述预设参数指标可以包括:Evans指数、Hackman指数及第三脑室横径C、第四脑室横径D、脑沟宽度F。其中:
Evans指数=侧脑室前角A/颅骨最大内径E;
Hackman指数=侧脑室前角A+尾状核B;
进一步以测量颅骨最大内径E为例,本申请实施例中,所述图像分析模块104将所述目标图像集中所有目标图像进行对比,提取出颅腔直径最大的目标图像并对所述颅腔直径最大的目标图像进行量化,方便后续利用所述最大颅腔直径进行运算所述检测指标。
本申请实施例所述图像分析模块104根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
本申请实施例中,所述图像分析模块104通过将所述检测指标集与所述标准指标集进行比较,得到所述医学图像集的分析结果。
详细地,所述图像分析模块104根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果,包括:
将所述检测指标集中每一个检测指标与所述标准指标集中对应的标准指标进行对比,并计算所述检测指标与对应的所述标准指标的差值。
本申请实施例以大脑组织的上述Evans指数为例,当检测Evans指数为X Evans时,标 准Evans指数为Y Evans,则计算所述检测Evans指数与标准Evans指数的差值为|X Evans-Y Evans|。
当所述差值在所述预设阈值范围之内,则所述图像分析模块104判断所述检测指标为正常,及当所述差值不在所述预设阈值范围之内,所述图像分析模块104判断所述检测指标为异常。
本申请实施例中,当所述组织结构的检测指标与所述标准指标的差值小于或等于规定阈值,所述图像分析模块104判定所述检测指标为正常,若所述差值大于所述阈值,所述图像分析模块104判定所述检测指标为异常。以上述Evans指数为例,当X Evans=0.2,Y Evans=0.3,则|X Evans-Y Evans|=0.1,预设阈值为0.05,0.1大于0.05,则所述医学图像表明大脑可能会出现脑积水。
本申请实施例中,所述图像分析模块104利用所有正常的检测指标构建的正常指标集,及利用所有异常的检测指标构建的异常指标集,根据所述正常指标集及所述异常指标集,得到分析结果。
本申请其他实施例中,图像分析模块104也可以只整合所述出现异常的检测指标,构建异常指标集,总结各个所述异常指标的对应问题,可以得到分析结果。
如图4所示,是本申请实现医学图像分析方法的电子设备的结构示意图。
所述电子设备1可以包括处理器10、存储器11和总线,还可以包括存储在所述存储器11中并可在所述处理器10上运行的计算机程序,如医学图像分析程序12。
其中,所述存储器11至少包括一种类型的可读存储介质,所述可读存储介质包括闪存、移动硬盘、多媒体卡、卡型存储器(例如:SD或DX存储器等)、磁性存储器、磁盘、光盘等。所述存储器11在一些实施例中可以是电子设备1的内部存储单元,例如该电子设备1的移动硬盘。所述存储器11在另一些实施例中也可以是电子设备1的外部存储设备,例如电子设备1上配备的插接式移动硬盘、智能存储卡(Smart Media Card,SMC)、安全数字(Secure Digital,SD)卡、闪存卡(Flash Card)等。进一步地,所述存储器11还可以既包括电子设备1的内部存储单元也包括外部存储设备。所述存储器11不仅可以用于存储安装于电子设备1的应用软件及各类数据,例如医学图像分析程序12的代码等,还可以用于暂时地存储已经输出或者将要输出的数据。
所述处理器10在一些实施例中可以由集成电路组成,例如可以由单个封装的集成电路所组成,也可以是由多个相同功能或不同功能封装的集成电路所组成,包括一个或者多个中央处理器(Central Processing unit,CPU)、微处理器、数字处理芯片、图形处理器及各种控制芯片的组合等。所述处理器10是所述电子设备的控制核心(Control Unit),利用各种接口和线路连接整个电子设备的各个部件,通过运行或执行存储在所述存储器11内的程序或者模块(例如执行医学图像分析程序等),以及调用存储在所述存储器11内的数据,以执行电子设备1的各种功能和处理数据。
所述总线可以是外设部件互连标准(peripheral component interconnect,简称PCI)总线或扩展工业标准结构(extended industry standard architecture,简称EISA)总线等。该总线可以分为地址总线、数据总线、控制总线等。所述总线被设置为实现所述存储器11以及至少一个处理器10等之间的连接通信。
图4仅示出了具有部件的电子设备,本领域技术人员可以理解的是,图4示出的结构并不构成对所述电子设备1的限定,可以包括比图示更少或者更多的部件,或者组合某些部件,或者不同的部件布置。
例如,尽管未示出,所述电子设备1还可以包括给各个部件供电的电源(比如电池),优选地,电源可以通过电源管理装置与所述至少一个处理器10逻辑相连,从而通过电源管理装置实现充电管理、放电管理、以及功耗管理等功能。电源还可以包括一个或一个以上的直流或交流电源、再充电装置、电源故障检测电路、电源转换器或者逆变器、电源状 态指示器等任意组件。所述电子设备1还可以包括多种传感器、蓝牙模块、Wi-Fi模块等,在此不再赘述。
进一步地,所述电子设备1还可以包括网络接口,可选地,所述网络接口可以包括有线接口和/或无线接口(如WI-FI接口、蓝牙接口等),通常用于在该电子设备1与其他电子设备之间建立通信连接。
可选地,该电子设备1还可以包括用户接口,用户接口可以是显示器(Display)、输入单元(比如键盘(Keyboard)),可选地,用户接口还可以是标准的有线接口、无线接口。可选地,在一些实施例中,显示器可以是LED显示器、液晶显示器、触控式液晶显示器以及OLED(Organic Light-Emitting Diode,有机发光二极管)触摸器等。其中,显示器也可以适当的称为显示屏或显示单元,用于显示在电子设备1中处理的信息以及用于显示可视化的用户界面。
应该了解,所述实施例仅为说明之用,在专利申请范围上并不受此结构的限制。
所述电子设备1中的所述存储器11存储的医学图像分析程序12是多个指令的组合,在所述处理器10中运行时,可以实现:
获取目标物的医学图像集及获取所述目标物对应的标准指标集,对所述医学图像集进行去噪处理,得到去噪图像集;
对所述去噪图像集进行特征提取,得到特征图像集;
将所述特征图像集进行特征聚类,得到聚类特征图像集;
根据所述聚类特征图像集,对所述特征图像集进行切割,得到目标图像集;
按照预设参数指标,将所述目标图像集中目标图像进行对比及提取操作,得到提取结果,并量化所述提取结果,得到所述目标物的检测指标集;
根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
进一步地,所述电子设备1集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。所述计算机可读存储介质可以是易失性的,也可以是非易失性的,所述计算机可读介质可以包括:能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(ROM,Read-Only Memory)。
进一步地,所述计算机可用存储介质可主要包括存储程序区和存储数据区,其中,存储程序区可存储操作系统、至少一个功能所需的计算机程序等;存储数据区可存储根据区块链节点的使用所创建的数据等,所述计算机程序被处理器执行时,可以实现:
获取目标物的医学图像集及获取所述目标物对应的标准指标集,对所述医学图像集进行去噪处理,得到去噪图像集;
对所述去噪图像集进行特征提取,得到特征图像集;
将所述特征图像集进行特征聚类,得到聚类特征图像集;
根据所述聚类特征图像集,对所述特征图像集进行切割,得到目标图像集;
按照预设参数指标,将所述目标图像集中目标图像进行对比及提取操作,得到提取结果,并量化所述提取结果,得到所述目标物的检测指标集;
根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
本申请之计算机可读存储介质的具体实施方式与上述医学图像分析方法的具体实施方式大致相同,在此不再赘述。
在另一个实施例中,本申请所提供的医学图像分析方法,为进一步保证上述所有出现的数据的私密和安全性,上述所有数据还可以存储于一区块链的节点中。例如医学图像集、特征图像集或目标图像集等等,这些数据均可存储在区块链节点中。
在本申请所提供的几个实施例中,应该理解到,所揭露的设备,装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述模块的 划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式。
所述作为分离部件说明的模块可以是或者也可以不是物理上分开的,作为模块显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。
另外,在本申请各个实施例中的各功能模块可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用硬件加软件功能模块的形式实现。
对于本领域技术人员而言,显然本申请不限于上述示范性实施例的细节,而且在不背离本申请的精神或基本特征的情况下,能够以其他的具体形式实现本申请。
因此,无论从哪一点来看,均应将实施例看作是示范性的,而且是非限制性的,本申请的范围由所附权利要求而不是上述说明限定,因此旨在将落在权利要求的等同要件的含义和范围内的所有变化涵括在本申请内。不应将权利要求中的任何附关联图表记视为限制所涉及的权利要求。
本申请所指区块链是分布式数据存储、点对点传输、共识机制、加密算法等计算机技术的新型应用模式。区块链(Blockchain),本质上是一个去中心化的数据库,是一串使用密码学方法相关联产生的数据块,每一个数据块中包含了一批次网络交易的信息,用于验证其信息的有效性(防伪)和生成下一个区块。区块链可以包括区块链底层平台、平台产品服务层以及应用服务层等。
此外,显然“包括”一词不排除其他单元或步骤,单数不排除复数。系统权利要求中陈述的多个单元或装置也可以由一个单元或装置通过软件或者硬件来实现。第二等词语用来表示名称,而并不表示任何特定的顺序。
最后应说明的是,以上实施例仅用以说明本申请的技术方案而非限制,尽管参照较佳实施例对本申请进行了详细说明,本领域的普通技术人员应当理解,可以对本申请的技术方案进行修改或等同替换,而不脱离本申请技术方案的精神和范围。

Claims (20)

  1. 一种医学图像分析方法,其中,所述方法包括:
    获取目标物的医学图像集及获取所述目标物对应的标准指标集,对所述医学图像集进行去噪处理,得到去噪图像集;
    对所述去噪图像集进行特征提取,得到特征图像集;
    将所述特征图像集进行特征聚类,得到聚类特征图像集;
    根据所述聚类特征图像集,对所述特征图像集进行切割,得到目标图像集;
    按照预设参数指标,将所述目标图像集中目标图像进行对比及提取操作,得到提取结果,并量化所述提取结果,得到所述目标物的检测指标集;
    根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
  2. 如权利要求1所述医学图像分析方法,其中,根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果,包括:
    将所述检测指标集中每一个检测指标与所述标准指标集中对应的标准指标进行对比,并计算所述检测指标与对应的所述标准指标的差值;
    当所述差值大于预设阈值时,判定所述检测指标为异常指标;
    根据所有异常的检测指标构建异常指标集,得到所述分析结果。
  3. 如权利要求1所述医学图像分析方法,其中,所述将所述特征图像集进行特征聚类,得到聚类特征图像集,包括:
    计算所述特征图像集中每一张特征图像中目标物的标准簇心,得到簇心集;
    根据所述簇心集,利用预构建的聚类算法,对所述特征图像进行边缘聚类,得到所述聚类特征图像。
  4. 如权利要求3所述医学图像分析方法,其中,所述计算所述特征图像集中每一张特征图像的标准簇心,包括:
    将所述特征图像中所述目标物作为一个簇,初始化所述簇,随机得到所述簇对应的初始簇心;
    计算所述簇的损失值,比较所述损失值与预设阈值,当所述损失值大于所述阈值时,重新初始化所述初始簇心,当所述损失值小于或等于所述阈值时,确定所述初始簇心为标准簇心。
  5. 如权利要求1所述医学图像分析方法,其中,所述对所述去噪图像集进行特征提取,得到特征图像集,包括:
    将所述去噪图像集进行卷积,得到特征映射矩阵集;
    将所述特征映射矩阵集进行池化及激活操作,通过预构建的全连接层,生成所述特征图像集。
  6. 如权利要求1所述医学图像分析方法,其中,所述对所述医学图像集进行去噪处理,得到去噪图像集,包括:
    对所述医学图像集进行椒盐噪声滤波,得到半消噪图像集;
    对所述半消噪图像集进行高斯噪声滤波,得到所述去噪图像集。
  7. 如权利要求1至6中任意一项所述医学图像分析方法,其中,所述获取所述目标物对应的标准指标集,包括;
    构建分布式的区块链节点并连接预构建的数据库管理系统;
    通过所述数据库管理系统,存储所述目标物的标准指标集至所述区块链节点;
    当检测所述目标物时,获取目标物关键字信息,查询得到所述目标物的标准指标集。
  8. 一种医学图像分析装置,其中,所述装置包括:
    图像获取模块,用于获取目标物的医学图像集及获取所述目标物对应的标准指标集, 对所述医学图像集进行去噪处理,得到去噪图像集;
    图像处理模块,用于对所述去噪图像集进行特征提取,得到特征图像集,将所述特征图像集进行特征聚类,得到聚类特征图像集;
    图像切割模块根据所述聚类特征图像集,对所述特征图像集进行切割,得到目标图像集;
    图像分析模块,用于按照预设参数指标,将所述目标图像集中目标图像进行对比及提取操作,得到提取结果,并量化所述提取结果,得到所述目标物的检测指标集,根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
  9. 一种电子设备,其中,所述电子设备包括:
    至少一个处理器;以及,
    与所述至少一个处理器通信连接的存储器;其中,
    所述存储器存储有能够被所述至少一个处理器执行的计算机程序指令,所述计算机程序指令被所述至少一个处理器执行,以使所述至少一个处理器能够执行如下步骤:
    获取目标物的医学图像集及获取所述目标物对应的标准指标集,对所述医学图像集进行去噪处理,得到去噪图像集;
    对所述去噪图像集进行特征提取,得到特征图像集;
    将所述特征图像集进行特征聚类,得到聚类特征图像集;
    根据所述聚类特征图像集,对所述特征图像集进行切割,得到目标图像集;
    按照预设参数指标,将所述目标图像集中目标图像进行对比及提取操作,得到提取结果,并量化所述提取结果,得到所述目标物的检测指标集;
    根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
  10. 如权利要求9所述电子设备,其中,根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果,包括:
    将所述检测指标集中每一个检测指标与所述标准指标集中对应的标准指标进行对比,并计算所述检测指标与对应的所述标准指标的差值;
    当所述差值大于预设阈值时,判定所述检测指标为异常指标;
    根据所有异常的检测指标构建异常指标集,得到所述分析结果。
  11. 如权利要求9所述电子设备,其中,所述将所述特征图像集进行特征聚类,得到聚类特征图像集,包括:
    计算所述特征图像集中每一张特征图像中目标物的标准簇心,得到簇心集;
    根据所述簇心集,利用预构建的聚类算法,对所述特征图像进行边缘聚类,得到所述聚类特征图像。
  12. 如权利要求11所述电子设备,其中,所述计算所述特征图像集中每一张特征图像的标准簇心,包括:
    将所述特征图像中所述目标物作为一个簇,初始化所述簇,随机得到所述簇对应的初始簇心;
    计算所述簇的损失值,比较所述损失值与预设阈值,当所述损失值大于所述阈值时,重新初始化所述初始簇心,当所述损失值小于或等于所述阈值时,确定所述初始簇心为标准簇心。
  13. 如权利要求9所述电子设备,其中,所述对所述去噪图像集进行特征提取,得到特征图像集,包括:
    将所述去噪图像集进行卷积,得到特征映射矩阵集;
    将所述特征映射矩阵集进行池化及激活操作,通过预构建的全连接层,生成所述特征图像集。
  14. 如权利要求9所述电子设备,其中,所述对所述医学图像集进行去噪处理,得到 去噪图像集,包括:
    对所述医学图像集进行椒盐噪声滤波,得到半消噪图像集;
    对所述半消噪图像集进行高斯噪声滤波,得到所述去噪图像集。
  15. 如权利要求9至14中任意一项所述电子设备,其中,所述获取所述目标物对应的标准指标集,包括;
    构建分布式的区块链节点并连接预构建的数据库管理系统;
    通过所述数据库管理系统,存储所述目标物的标准指标集至所述区块链节点;
    当检测所述目标物时,获取目标物关键字信息,查询得到所述目标物的标准指标集。
  16. 一种计算机可读存储介质,包括存储数据区和存储程序区,其中,所述存储数据区存储创建的数据,所述存储程序区存储有计算机程序;其中,所述计算机程序被处理器执行时实现如下步骤:
    获取目标物的医学图像集及获取所述目标物对应的标准指标集,对所述医学图像集进行去噪处理,得到去噪图像集;
    对所述去噪图像集进行特征提取,得到特征图像集;
    将所述特征图像集进行特征聚类,得到聚类特征图像集;
    根据所述聚类特征图像集,对所述特征图像集进行切割,得到目标图像集;
    按照预设参数指标,将所述目标图像集中目标图像进行对比及提取操作,得到提取结果,并量化所述提取结果,得到所述目标物的检测指标集;
    根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果。
  17. 如权利要求16所述计算机可读存储介质,其中,根据所述检测指标集与所述标准指标集,得到所述医学图像集的分析结果,包括:
    将所述检测指标集中每一个检测指标与所述标准指标集中对应的标准指标进行对比,并计算所述检测指标与对应的所述标准指标的差值;
    当所述差值大于预设阈值时,判定所述检测指标为异常指标;
    根据所有异常的检测指标构建异常指标集,得到所述分析结果。
  18. 如权利要求16所述计算机可读存储介质,其中,所述将所述特征图像集进行特征聚类,得到聚类特征图像集,包括:
    计算所述特征图像集中每一张特征图像中目标物的标准簇心,得到簇心集;
    根据所述簇心集,利用预构建的聚类算法,对所述特征图像进行边缘聚类,得到所述聚类特征图像。
  19. 如权利要求18所述计算机可读存储介质,其中,所述计算所述特征图像集中每一张特征图像的标准簇心,包括:
    将所述特征图像中所述目标物作为一个簇,初始化所述簇,随机得到所述簇对应的初始簇心;
    计算所述簇的损失值,比较所述损失值与预设阈值,当所述损失值大于所述阈值时,重新初始化所述初始簇心,当所述损失值小于或等于所述阈值时,确定所述初始簇心为标准簇心。
  20. 如权利要求16所述计算机可读存储介质,其中,所述对所述去噪图像集进行特征提取,得到特征图像集,包括:
    将所述去噪图像集进行卷积,得到特征映射矩阵集;
    将所述特征映射矩阵集进行池化及激活操作,通过预构建的全连接层,生成所述特征图像集。
PCT/CN2020/131977 2020-10-14 2020-11-26 医学图像分析方法、装置、电子设备及可读存储介质 Ceased WO2021189900A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011096338.3A CN112102351B (zh) 2020-10-14 2020-10-14 医学图像分析方法、装置、电子设备及可读存储介质
CN202011096338.3 2020-10-14

Publications (1)

Publication Number Publication Date
WO2021189900A1 true WO2021189900A1 (zh) 2021-09-30

Family

ID=73784221

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/131977 Ceased WO2021189900A1 (zh) 2020-10-14 2020-11-26 医学图像分析方法、装置、电子设备及可读存储介质

Country Status (2)

Country Link
CN (1) CN112102351B (zh)
WO (1) WO2021189900A1 (zh)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114121236A (zh) * 2021-11-22 2022-03-01 上海商汤智能科技有限公司 图像检测方法及相关装置、电子设备和存储介质
CN114495056A (zh) * 2022-01-14 2022-05-13 广州小鹏自动驾驶科技有限公司 停车场柱子检测方法、检测装置、车辆及存储介质
CN115601684A (zh) * 2022-11-08 2023-01-13 深圳市计通智能技术有限公司(Cn) 突发事件预警方法、装置、电子设备及存储介质
CN116205940A (zh) * 2023-05-05 2023-06-02 武汉大学人民医院(湖北省人民医院) 基于医学检验的数字图像目标检测方法及系统
CN116705250A (zh) * 2023-06-07 2023-09-05 北京海上升科技有限公司 一种医疗影像大数据的低耗优化和智能存储方法及系统
CN117152106A (zh) * 2023-09-13 2023-12-01 厦门良医真选健康科技有限公司 基于深度学习的生物活性检测方法及装置

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114822752A (zh) * 2022-04-28 2022-07-29 上海商汤智能科技有限公司 医学报告生成方法以及相关装置、设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101295309A (zh) * 2008-05-22 2008-10-29 江苏大学 一种医学图像识别的方法
CN108681731A (zh) * 2018-04-03 2018-10-19 中山大学 一种甲状腺癌症超声图片自动标注方法及系统
CN110728673A (zh) * 2019-10-21 2020-01-24 上海联影医疗科技有限公司 一种目标部位分析方法、装置、计算机设备及存储介质
CN111259979A (zh) * 2020-02-10 2020-06-09 大连理工大学 一种基于标签自适应策略的深度半监督图像聚类方法
CN111652845A (zh) * 2020-04-27 2020-09-11 平安科技(深圳)有限公司 异常细胞自动标注方法、装置、电子设备及存储介质
CN111753831A (zh) * 2020-06-28 2020-10-09 上海联影医疗科技有限公司 一种图像分析方法、装置、图像采集设备及存储介质

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103617618B (zh) * 2013-12-03 2017-06-13 西安电子科技大学 基于特征提取与聚类集成的sar图像分割方法
CN110111323B (zh) * 2019-05-13 2021-08-10 上海交通大学医学院附属上海儿童医学中心 髋关节检测方法及装置
CN110675412B (zh) * 2019-09-27 2023-08-01 腾讯科技(深圳)有限公司 图像分割方法、图像分割模型的训练方法、装置及设备

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101295309A (zh) * 2008-05-22 2008-10-29 江苏大学 一种医学图像识别的方法
CN108681731A (zh) * 2018-04-03 2018-10-19 中山大学 一种甲状腺癌症超声图片自动标注方法及系统
CN110728673A (zh) * 2019-10-21 2020-01-24 上海联影医疗科技有限公司 一种目标部位分析方法、装置、计算机设备及存储介质
CN111259979A (zh) * 2020-02-10 2020-06-09 大连理工大学 一种基于标签自适应策略的深度半监督图像聚类方法
CN111652845A (zh) * 2020-04-27 2020-09-11 平安科技(深圳)有限公司 异常细胞自动标注方法、装置、电子设备及存储介质
CN111753831A (zh) * 2020-06-28 2020-10-09 上海联影医疗科技有限公司 一种图像分析方法、装置、图像采集设备及存储介质

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114121236A (zh) * 2021-11-22 2022-03-01 上海商汤智能科技有限公司 图像检测方法及相关装置、电子设备和存储介质
CN114495056A (zh) * 2022-01-14 2022-05-13 广州小鹏自动驾驶科技有限公司 停车场柱子检测方法、检测装置、车辆及存储介质
CN115601684A (zh) * 2022-11-08 2023-01-13 深圳市计通智能技术有限公司(Cn) 突发事件预警方法、装置、电子设备及存储介质
CN116205940A (zh) * 2023-05-05 2023-06-02 武汉大学人民医院(湖北省人民医院) 基于医学检验的数字图像目标检测方法及系统
CN116205940B (zh) * 2023-05-05 2023-08-01 武汉大学人民医院(湖北省人民医院) 基于医学检验的数字图像目标检测方法及系统
CN116705250A (zh) * 2023-06-07 2023-09-05 北京海上升科技有限公司 一种医疗影像大数据的低耗优化和智能存储方法及系统
CN117152106A (zh) * 2023-09-13 2023-12-01 厦门良医真选健康科技有限公司 基于深度学习的生物活性检测方法及装置

Also Published As

Publication number Publication date
CN112102351B (zh) 2024-10-22
CN112102351A (zh) 2020-12-18

Similar Documents

Publication Publication Date Title
WO2021189900A1 (zh) 医学图像分析方法、装置、电子设备及可读存储介质
CN110232383B (zh) 一种基于深度学习模型的病灶图像识别方法及病灶图像识别系统
Albalawi et al. Classification of breast cancer mammogram images using convolution neural network
ES2905660T3 (es) Procedimiento y sistema para el triaje asistido por ordenador
Rajini et al. Computer aided detection of ischemic stroke using segmentation and texture features
JP7082993B2 (ja) 医療画像処理装置、方法及びプログラム、診断支援装置、方法及びプログラム並びに医療支援システム及び方法
CN102165454B (zh) 用于提高计算机辅助诊断对图像处理不确定性的鲁棒性的方法
Ahirwar Study of techniques used for medical image segmentation and computation of statistical test for region classification of brain MRI
CN105431089B (zh) 根据肝脏扫描确定肝功能的系统和方法
WO2021082691A1 (zh) 眼部oct图像病灶区域的分割方法、装置及终端设备
WO2021189909A1 (zh) 病灶检测分析方法、装置、电子设备及计算机存储介质
KR20170046104A (ko) 질환 모델 기반의 의료 정보 서비스 제공 방법 및 장치
CN115131343B (zh) 一种慢性鼻窦炎分型方法、装置、设备及可读存储介质
WO2022036968A1 (zh) 心电图数据增广方法、装置、电子设备及介质
Berkley et al. Clinical capability of modern brain tumor segmentation models
Sivanantham Deep learning-based convolutional neural network with cuckoo search optimization for MRI brain tumour segmentation
CN113610841A (zh) 一种血管异常图像识别方法、装置、电子设备和存储介质
CN116712080A (zh) 基于单导联心电图心肌梗死检测与定位方法、装置及介质
Priya et al. Detection and grading of diabetic retinopathy in retinal images using deep intelligent systems: a comprehensive review
CN114387287A (zh) 脑部核心梗死区分割方法、装置和计算机设备
Ghaderi et al. Marker-controlled watershed algorithm and fuzzy C-means clustering machine learning: automated segmentation of glioblastoma from MRI images in a case series
JP2019213785A (ja) 医用画像処理装置、方法およびプログラム
CN112633405A (zh) 模型训练方法、医学影像分析方法及装置、设备、介质
CN105726052B (zh) 一种获取浸润性病灶概率的方法及系统
CN115429291B (zh) 结合多尺度散布熵与多标签k近邻的癫痫检测系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20927387

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20927387

Country of ref document: EP

Kind code of ref document: A1