CN116563358B - Data alignment preprocessing method for liver enhancement multi-stage CT data AI training - Google Patents
Data alignment preprocessing method for liver enhancement multi-stage CT data AI training Download PDFInfo
- Publication number
- CN116563358B CN116563358B CN202310827251.6A CN202310827251A CN116563358B CN 116563358 B CN116563358 B CN 116563358B CN 202310827251 A CN202310827251 A CN 202310827251A CN 116563358 B CN116563358 B CN 116563358B
- Authority
- CN
- China
- Prior art keywords
- data
- liver
- slice
- arterial
- phase
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 210000004185 liver Anatomy 0.000 title claims abstract description 438
- 238000000034 method Methods 0.000 title claims abstract description 70
- 238000012549 training Methods 0.000 title claims abstract description 36
- 238000007781 pre-processing Methods 0.000 title claims description 19
- 238000001514 detection method Methods 0.000 claims abstract description 32
- 230000003111 delayed effect Effects 0.000 claims description 16
- 238000013473 artificial intelligence Methods 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 8
- 238000004891 communication Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 238000010339 medical test Methods 0.000 claims description 5
- 238000003384 imaging method Methods 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 7
- 238000002203 pretreatment Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 16
- 239000004973 liquid crystal related substance Substances 0.000 description 8
- 238000004590 computer program Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000002591 computed tomography Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 206010019695 Hepatic neoplasm Diseases 0.000 description 1
- 206010028980 Neoplasm Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- XDDAORKBJWWYJS-UHFFFAOYSA-N glyphosate Chemical compound OC(=O)CNCP(O)(O)=O XDDAORKBJWWYJS-UHFFFAOYSA-N 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 208000014018 liver neoplasm Diseases 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/38—Registration of image sequences
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06T7/0014—Biomedical image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30056—Liver; Hepatic
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The application provides a data alignment pretreatment method for liver enhancement multi-stage CT data AI training. The method comprises the following steps: acquiring arterial liver CT data, portal liver CT data, venous liver CT data and delay liver CT data; acquiring a plurality of groups of arterial phase CT slice data, a plurality of groups of portal phase CT slice data, a plurality of groups of venous phase CT slice data and a plurality of groups of delay phase CT slice data based on four-phase liver CT data and the grouping number of the four-phase liver CT data; and training the artificial intelligent medical detection model based on the combined data of the arterial phase CT slice data, the portal phase CT slice data, the venous phase CT slice data and the delay phase CT slice data to obtain a trained artificial intelligent medical detection model. The data alignment pretreatment method for the liver enhancement multi-stage CT data AI training can improve the training effect of the model and improve the precision of the model.
Description
Technical Field
The application belongs to the field of medical detection, relates to a data preprocessing method, and particularly relates to a data alignment preprocessing method for liver enhancement multi-stage CT data AI training.
Background
In the training process of the liver tumor artificial intelligence medical detection model, multi-stage CT (Computed Tomography, electronic computer tomography) data are often required as input, the CT data comprise arterial liver CT data, portal liver CT data, venous liver CT data and delay liver CT data, the layer thickness value of each stage data is generally 5mm, but when the tumor volume is smaller, thinner layer data such as 1.25mm are generally required. Since it takes a long time for the CT machine to scan the slice data, the above four-phase data can only be used for one phase of slice data, such as the gate phase, and the thickness of the slice data of the other phases is still 5mm.
When the layer thickness values of the multi-stage data are inconsistent, the problem that the number of multi-stage data is asymmetric is caused, for example, the layer number of the arterial liver CT data is 50, the layer thickness value is 5mm, the layer number of the portal liver CT data is 200, the layer thickness value is 1.25mm, the layer number of the venous liver CT data is 50, the layer thickness value is 5mm, the layer number of the delay liver CT data is 100, the layer thickness value is 2.5mm, and when the multi-stage data with asymmetric number are used as input data for training of an artificial intelligent medical detection model, the training effect is affected, and therefore the model accuracy is not high.
Disclosure of Invention
The application aims to provide a data alignment preprocessing method for liver enhancement multi-stage CT data AI training, which is used for solving the problems of poor training effect and low model precision when the existing multi-stage data is used as input data for artificial intelligent medical detection model training.
In a first aspect, the present application provides a data alignment preprocessing method for AI training of liver-enhanced multi-stage CT data, the method comprising: acquiring arterial liver CT data, portal liver CT data, venous liver CT data and delay liver CT data; acquiring the group number of the arterial liver CT data, the group number of the portal liver CT data, the group number of the venous liver CT data and the group number of the delay liver CT data based on the layer thickness value of the arterial liver CT data, the layer thickness value of the portal liver CT data, the layer thickness value of the venous liver CT data and the layer thickness value of the delay liver CT data; acquiring a plurality of groups of arterial phase CT slice data, a plurality of groups of portal phase CT slice data, a plurality of groups of venous phase CT slice data and a plurality of groups of delay phase CT slice data based on the four-phase liver CT data and the grouping number of the four-phase liver CT data, wherein the number of slices of the arterial phase CT slice data, the number of slices of the portal phase CT slice data, the number of slices of the venous phase CT slice data and the number of slices of the delay phase CT slice data are mutually aligned; and training the artificial intelligent medical detection model based on the combined data of the arterial phase CT slice data, the portal phase CT slice data, the venous phase CT slice data and the delay phase CT slice data to obtain a trained artificial intelligent medical detection model.
According to the method, the arterial phase CT slice data, the portal phase CT slice data, the venous phase CT slice data and the delayed phase CT slice data are combined data as the input of the artificial intelligent medical detection model, so that the model input is the slice data with the same four-phase slice number, namely the model input is the slice data with the four-phase symmetrical number, the training effect of the model is improved, and the model precision is improved.
In an embodiment of the application, the method further comprises: preprocessing the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delay liver CT data to obtain preprocessed arterial liver CT data, preprocessed portal liver CT data, preprocessed venous liver CT data and preprocessed delay liver CT data, wherein each layer of preprocessed arterial liver CT data, each layer of preprocessed portal liver CT data, each layer of preprocessed venous liver CT data and each layer of preprocessed delay liver CT data all comprise liver information.
In an embodiment of the present application, the method for obtaining the number of packets of arterial liver CT data, the number of packets of portal liver CT data, the number of packets of venous liver CT data, and the number of packets of delay liver CT data includes: acquiring a first layer thickness multiple, a second layer thickness multiple, a third layer thickness multiple, a fourth layer thickness multiple and a fifth layer thickness multiple based on the layer thickness value of the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delayed liver CT data, wherein the fifth layer thickness multiple is the minimum value of the first layer thickness multiple, the second layer thickness multiple, the third layer thickness multiple and the fourth layer thickness multiple; and acquiring the grouping number of the arterial liver CT data, the grouping number of the portal liver CT data, the grouping number of the venous liver CT data and the grouping number of the delay liver CT data based on the first layer thickness multiple, the second layer thickness multiple, the third layer thickness multiple, the fourth layer thickness multiple and the fifth layer thickness multiple.
In an embodiment of the present application, the first layer thickness multiple is:
；
the second layer thickness multiple is:
；
the third layer thickness multiple is:
；
the fourth layer thickness multiple is:
;
wherein, the liquid crystal display device comprises a liquid crystal display device,layer thickness values representing the arterial liver CT data,layer thickness values representing the portal liver CT data,layer thickness values representing the venous liver CT data,layer thickness values representing the delay period liver CT data,indicating a multiple of the first layer thickness,indicating a multiple of the thickness of the second layer,indicating a multiple of the thickness of the third layer,indicating a multiple of the fourth layer thickness,representing the fifth layer thickness multiple, the fifth layer thickness multiple represented as:。
in one embodiment of the present application, the number of packets of arterial liver CT data is expressed as:
；
the number of packets of portal liver CT data is expressed as:
；
the number of packets of venous liver CT data is expressed as:
；
the number of packets of the delay period liver CT data is expressed as:
；
wherein, the liquid crystal display device comprises a liquid crystal display device,a packet number representing the arterial liver CT data,a packet number representing the portal liver CT data,a number of packets representing the venous liver CT data,a number of packets representing the delay period liver CT data, Representing a multiple of the thickness of the fifth layer,as a round-up function.
In an embodiment of the present application, a method for obtaining a plurality of sets of arterial phase CT slice data, a plurality of sets of portal phase CT slice data, a plurality of sets of venous phase CT slice data, and a plurality of sets of delay phase CT slice data includes: acquiring a slice sequence of arterial liver CT data, a slice sequence of portal liver CT data, a slice sequence of venous liver CT data and a slice sequence of delay liver CT data based on the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delay liver CT data; and performing slice screening processing on the four-stage liver CT data based on the slice sequence of the four-stage liver CT data and the grouping number of the four-stage liver CT data so as to acquire the arterial stage CT slice data, the portal stage CT slice data, the venous stage CT slice data and the delay stage CT slice data.
In one embodiment of the application, the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delayed liver CT data are stored in a medical digital imaging and communication file.
In an embodiment of the application, the artificial intelligence medical examination model is a neural network model for medical examination.
In a second aspect, the present application provides a data alignment preprocessing method for AI training of liver-enhanced multi-stage CT data, which specifically includes: acquiring arterial liver CT data to be detected, portal liver CT data to be detected, venous liver CT data to be detected and delay liver CT data to be detected; and detecting the arterial liver CT data to be detected, the portal liver CT data to be detected, the venous liver CT data to be detected and the delay liver CT data to be detected by using an artificial intelligent medical detection model, wherein the artificial intelligent medical detection model is trained by adopting the method according to any one of the first aspect.
In an embodiment of the present application, the implementation method for detecting the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delayed liver CT data by using an artificial intelligence medical detection model includes: processing the arterial liver CT data to be detected, the portal liver CT data to be detected, the venous liver CT data to be detected and the delay liver CT data to be detected by using the artificial intelligent medical detection model so as to obtain an intersection ratio; and obtaining a prediction result based on the intersection ratio.
As described above, the data alignment pretreatment method for liver enhancement multi-stage CT data AI training of the present application has the following beneficial effects:
according to the method, the arterial phase CT slice data, the portal phase CT slice data, the venous phase CT slice data and the delayed phase CT slice data are combined data as the input of the artificial intelligent medical detection model, so that the model input is the slice data with the same four-phase slice number, namely the model input is the slice data with the four-phase symmetrical number, the training effect of the model is improved, and the model precision is improved.
Drawings
FIG. 1 is a schematic diagram of a model training system according to an embodiment of the present application.
Fig. 2 is a flowchart of a data alignment preprocessing method for AI training of liver-enhanced multi-stage CT data according to an embodiment of the present application.
Fig. 3 is a flowchart showing an implementation method of acquiring the number of packets of the arterial liver CT data, the number of packets of the portal liver CT data, the number of packets of the venous liver CT data, and the number of packets of the delay liver CT data according to an embodiment of the present application.
Fig. 4 is a flowchart of an implementation method for acquiring a plurality of sets of arterial phase CT slice data, a plurality of sets of portal phase CT slice data, a plurality of sets of venous phase CT slice data, and a plurality of sets of delay phase CT slice data according to an embodiment of the present application.
Fig. 5 is a flowchart of a data alignment preprocessing method for AI training of liver-enhanced multi-stage CT data according to an embodiment of the present application.
Fig. 6 is a flowchart of an implementation method for detecting the arterial liver CT data to be detected, the portal liver CT data to be detected, the venous liver CT data to be detected, and the delayed liver CT data to be detected by using an artificial intelligence medical detection model according to an embodiment of the present application.
Description of element numbers: 10. the model training system comprises a model training system, a 110 memory, a 120 processor, a 130 display, S11-S14 steps, S21-S22 steps, S31-S32 steps, S41-S42 steps and S51-S52 steps.
Detailed Description
Other advantages and effects of the present application will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present application with reference to specific examples. The application may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present application. It should be noted that the following embodiments and features in the embodiments may be combined with each other without conflict.
It should be noted that the illustrations provided in the following embodiments merely illustrate the basic concept of the present application by way of illustration, and only the components related to the present application are shown in the drawings and are not drawn according to the number, shape and size of the components in actual implementation, and the form, number and proportion of the components in actual implementation may be arbitrarily changed, and the layout of the components may be more complicated.
The following describes the technical solution in the embodiment of the present application in detail with reference to the drawings in the embodiment of the present application.
As shown in fig. 1, an embodiment of the present application provides a model training system 10, the model training system 10 comprising: the device comprises a memory 110, a processor 120 and a display 130, wherein the memory 110 is used for storing arterial liver CT data, portal liver CT data, venous liver CT data, delay liver CT data and the like, and the processor 120 is used for providing calculation force for a slice screening process of arterial liver CT data, a slice screening process of portal liver CT data, a slice screening process of venous liver CT data, a slice screening process of delay liver CT data and a training process of an artificial intelligent medical detection model. The display 130 is used to display an associated GUI (Graphical User Interface ) interactive interface.
As shown in fig. 2, the present embodiment provides a data alignment preprocessing method for liver enhancement multi-stage CT data AI training, which can be implemented by a processor of a computer device, and the method includes:
s11, arterial liver CT data, portal liver CT data, venous liver CT data and delay liver CT data are acquired.
Optionally, the arterial liver CT data refers to image data obtained by scanning arterial liver when the liver is examined using a computed tomography technique. The portal liver CT data, the venous liver CT data and the delay liver CT data are respectively image data obtained by scanning the portal, venous and delay phases. The arterial liver CT data, the portal liver CT data, the venous liver CT data and the delay liver CT data can be stored in a DICOM (Digital Imaging and Communications in Medicine, digital imaging and communication) file, the number of layers and the layer thickness value of the arterial liver CT data, the number of layers and the layer thickness value of the portal liver CT data, the number of layers and the layer thickness value of the venous liver CT data and the number of layers and the layer thickness value of the delay liver CT data can be directly obtained from the DICOM file, and the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delay liver CT data can be displayed in an image mode through a DICOM reader. When the number of layers of the arterial liver CT data is 250, the arterial liver CT data may include 250 arterial CT slices, where the arterial CT slices are two-dimensional images for presenting different slice structures of a body part of a patient, and the arterial CT slices are arterial CT slices for presenting a liver of the patient in this embodiment, and similarly, the portal CT slices, the venous CT slices, and the delay CT slices are not repeated herein. The thickness of the arterial liver CT data may be 1mm, and the thickness of the arterial liver CT data may refer to the thickness of the arterial CT slice in the arterial liver CT data. When the number of layers of the portal liver CT data is 200, the portal liver CT data may include 200 portal CT slices, the thickness of the portal liver CT data may be 1.25mm, when the number of layers of the venous liver CT data is 50, the venous liver CT data may include 50 venous CT slices, the thickness of the venous liver CT data may be 5mm, when the number of layers of the delay liver CT data is 50, the delay liver CT data may include 50 delay CT slices, and the thickness of the delay liver CT data may be 5mm.
Optionally, the method further comprises: preprocessing the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delay liver CT data to obtain preprocessed arterial liver CT data, preprocessed portal liver CT data, preprocessed venous liver CT data and preprocessed delay liver CT data, wherein each layer of preprocessed arterial liver CT data, each layer of preprocessed portal liver CT data, each layer of preprocessed venous liver CT data and each layer of preprocessed delay liver CT data all comprise liver information. When the arterial liver CT data contains 250 arterial liver CT slices before pretreatment, the arterial liver CT data contains 238 arterial liver CT slices after pretreatment, each layer of arterial liver CT data contains liver information, which may mean that each arterial CT slice in the arterial liver CT data after pretreatment contains liver information, each layer of portal liver CT data after pretreatment contains liver information, each portal CT slice in the portal liver CT data after pretreatment contains liver information, each layer of venous liver CT data after pretreatment contains liver information, which may mean that each venous CT slice in the venous liver CT data after pretreatment contains liver information, and each layer of delay liver CT data after pretreatment contains liver information, which may mean that each delay liver CT slice in the delay liver CT data after pretreatment contains liver information. The 12 arterial CT slices before and after pretreatment are arterial CT slices which do not contain liver information, and the liver information can be the development information of the liver in the image. The preprocessing process of the portal liver CT data, the venous liver CT data and the delay liver CT data is consistent with the process of the arterial liver CT data, and the embodiment will not be repeated here. The pretreatment of the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delay liver CT data can ensure that the pretreated four-stage liver CT data contain liver information, thereby ensuring the quality of the pretreated four-stage liver CT data.
Optionally, the implementation method for preprocessing the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delay liver CT data includes: the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delay liver CT data are subjected to segmentation processing through a convolutional neural network, so that the preprocessed arterial liver CT data, the preprocessed portal liver CT data, the preprocessed venous liver CT data and the delay liver CT data are obtained, the convolutional neural network can be a U-Net neural network, and the U-Net neural network can be used for segmenting images. Taking the arterial liver CT data as an example, the segmentation process may refer to segmenting each arterial CT slice in the arterial liver CT data, determining whether each segmented region includes liver information, and similarly, the portal liver CT data, the venous liver CT data, and the delay liver CT data are not described herein.
S12, acquiring the grouping number of the arterial liver CT data, the grouping number of the portal liver CT data, the grouping number of the venous liver CT data and the grouping number of the delay liver CT data based on the layer thickness value of the arterial liver CT data, the layer thickness value of the portal liver CT data, the layer thickness value of the venous liver CT data and the layer thickness value of the delay liver CT data.
Optionally, the number of packets of the arterial liver CT data is the final number of packets of the arterial liver CT data, for example, the arterial liver CT data includes 250 arterial CT slices, the number of packets of the arterial liver CT data may be 5, the arterial liver CT data may be finally divided into 5 sets of data, and the number of arterial CT slices in each set is 50. In addition, when the arterial liver CT data is 238 arterial CT slices, the number of packets of the arterial liver CT data is 4, the arterial liver CT data can be finally divided into 4 sets of data, the number of arterial CT slices in each set is 59, and two arterial liver CT slices are not in the packets, and the two arterial CT slices are directly discarded. Similarly, the principle that the number of packets of the portal liver CT data, the number of packets of the venous liver CT data and the number of packets of the delay liver CT data are similar to the number of packets of the arterial liver CT data is not described in detail herein.
S13, acquiring a plurality of groups of arterial phase CT slice data, a plurality of groups of portal phase CT slice data, a plurality of groups of venous phase CT slice data and a plurality of groups of delay phase CT slice data based on the four-phase liver CT data and the grouping number of the four-phase liver CT data, wherein the number of slices of the arterial phase CT slice data, the number of slices of the portal phase CT slice data, the number of slices of the venous phase CT slice data and the number of slices of the delay phase CT slice data are mutually aligned.
Optionally, the four-stage liver CT data are the arterial stage liver CT data, the portal stage liver CT data, the venous stage liver CT data, and the delay stage liver CT data. The number of groups of arterial phase CT slice data is the number of groups of arterial phase liver CT data, and the same as the number of groups of portal phase CT slice data, the number of groups of venous phase CT slice data and the number of groups of delay phase CT slice data are not described in detail herein.
Optionally, the arterial phase CT slice data may be a data set of arterial phase CT slices, and may include a plurality of arterial phase CT slices, and similarly, the portal phase CT slice data, the venous phase CT slice data, and the delay phase CT slice data are not described herein. For example, arterial liver CT data contains 250 arterial CT slices, which are eventually divided into 5 sets of arterial CT slice data, each set of arterial CT slice data contains 50 arterial CT slices, with a slice number of 50. The portal liver CT data comprise 200 portal CT slices which are finally divided into 4 groups of portal CT slice data, wherein each group of portal CT slice data comprises 50 portal CT slices, and the number of the slices is 50. The venous phase liver CT data comprise 50 venous phase CT slices which are finally divided into 1 group of venous phase CT slice data, and each group of venous phase CT slice data comprise 50 venous phase CT slices with the number of slices being 50. The delay period liver CT data comprises 50 delay period CT slices, and is finally divided into 1 group of 50 delay period CT slice data, wherein each group of delay period CT slice data comprises 50 delay period CT slices, and the number of the slices is 50. The number of slices of the arterial phase CT slice data, the number of slices of the portal phase CT slice data, the number of slices of the venous phase CT slice data, and the number of slices of the delay phase CT slice data being aligned with one another may mean that the number of slices in each set of arterial phase CT slice data is the same as the number of slices of each set of portal phase CT slice data, the number of slices of each set of venous phase CT slice data, and the number of slices of each set of delay phase CT slice data.
S14, training the artificial intelligent medical detection model based on the combined data of the arterial phase CT slice data, the portal phase CT slice data, the venous phase CT slice data and the delay phase CT slice data to obtain a trained artificial intelligent medical detection model.
Alternatively, the combined data of the arterial phase CT slice data, the portal phase CT slice data, the venous phase CT slice data and the delay phase CT slice data may be composed of arterial phase CT slice data of different groups, portal phase CT slice data of different groups, venous phase CT slice data of different groups and delay phase CT slice data of different groups, for example, the arterial phase liver CT data is divided into A, B, C sets of the arterial phase CT slice data, the portal phase liver CT data is divided into D, E two sets of the portal phase CT slice data, the venous phase CT slice data is divided into F sets of the venous phase CT slice data, the delay phase liver CT data is divided into G sets of the delay phase CT slice data, and the combined data may include six sets of four phase CT slice data, that is, (a, D, F, G) four phase CT slice data, (a, E, F, G) four phase CT slice data, (B, D, F, G) four phase CT slice data, (B, F) four phase CT slice data, (C, F) four phase CT slice data, (G) four phase CT slice data, and (C, F) four phase CT slice data, C, F phase CT data, four phase CT slice data, and the same number of each phase CT slice data.
Alternatively, the artificial intelligence medical test model may be an initialized artificial intelligence medical test model, which may refer to an artificial intelligence model for medical testing, which may be a neural network model.
From the above description, the method comprises the following steps: acquiring arterial liver CT data, portal liver CT data, venous liver CT data and delay liver CT data; acquiring the group number of the arterial liver CT data, the group number of the portal liver CT data, the group number of the venous liver CT data and the group number of the delay liver CT data based on the layer thickness value of the arterial liver CT data, the layer thickness value of the portal liver CT data, the layer thickness value of the venous liver CT data and the layer thickness value of the delay liver CT data; acquiring a plurality of groups of arterial phase CT slice data, a plurality of groups of portal phase CT slice data, a plurality of groups of venous phase CT slice data and a plurality of groups of delay phase CT slice data based on the four-phase liver CT data and the grouping number of the four-phase liver CT data, wherein the number of slices of the arterial phase CT slice data, the number of slices of the portal phase CT slice data, the number of slices of the venous phase CT slice data and the number of slices of the delay phase CT slice data are mutually aligned; and training the artificial intelligent medical detection model based on the combined data of the arterial phase CT slice data, the portal phase CT slice data, the venous phase CT slice data and the delay phase CT slice data to obtain a trained artificial intelligent medical detection model.
According to the method, the arterial phase CT slice data, the portal phase CT slice data, the venous phase CT slice data and the delayed phase CT slice data are combined data as the input of the artificial intelligent medical detection model, so that the model input is the slice data with the same four-phase slice number, namely the model input is the slice data with the four-phase symmetrical number, the training effect of the model is improved, and the model precision is improved. And the method can make maximum use of the arterial liver CT data, the portal liver CT data, the venous liver CT data, and the delayed liver CT data.
As shown in fig. 3, the implementation method for acquiring the packet number of the arterial liver CT data, the packet number of the portal liver CT data, the packet number of the venous liver CT data, and the packet number of the delay liver CT data in this embodiment includes:
s21, acquiring a first layer thickness multiple, a second layer thickness multiple, a third layer thickness multiple, a fourth layer thickness multiple and a fifth layer thickness multiple based on the layer thickness value of the arterial liver CT data, the layer thickness value of the portal liver CT data, the layer thickness value of the venous liver CT data and the layer thickness value of the delay liver CT data, wherein the fifth layer thickness multiple is the minimum value of the first layer thickness multiple, the second layer thickness multiple, the third layer thickness multiple and the fourth layer thickness multiple.
Optionally, the first layer thickness multiple is:
；
the second layer thickness multiple is:
；
the third layer thickness multiple is:
；
the fourth layer thickness multiple is:
;
wherein, the liquid crystal display device comprises a liquid crystal display device,layer thickness values representing the arterial liver CT data,layer thickness values representing the portal liver CT data,layer thickness values representing the venous liver CT data,layer thickness values representing the delay period liver CT data,indicating a multiple of the first layer thickness,indicating a multiple of the thickness of the second layer,indicating a multiple of the thickness of the third layer,indicating a multiple of the fourth layer thickness,representing the fifth layer thickness multiple, the fifth layer thickness multiple represented as:。
s22, acquiring the grouping number of the arterial liver CT data, the grouping number of the portal liver CT data, the grouping number of the venous liver CT data and the grouping number of the delay liver CT data based on the first layer thickness multiple, the second layer thickness multiple, the third layer thickness multiple, the fourth layer thickness multiple and the fifth layer thickness multiple.
Optionally, the number of packets of arterial liver CT data is expressed as:
；
the number of packets of portal liver CT data is expressed as:
；
the number of packets of venous liver CT data is expressed as:
；
The number of packets of the delay period liver CT data is expressed as:
；
wherein, the liquid crystal display device comprises a liquid crystal display device,a packet number representing the arterial liver CT data,a packet number representing the portal liver CT data,a number of packets representing the venous liver CT data,a number of packets representing the delay period liver CT data,representing a multiple of the thickness of the fifth layer,as a round-up function. The roundup (4.1) can be regarded as 5. For example, when the first layer thickness multiple is 1, the second layer thickness multiple is 0.8, the third layer thickness multiple is 0.2, and the fourth layer thickness multiple is 0.2, the number of packets of the arterial liver CT data is 5, the number of packets of the portal liver CT data is 4, and the number of packets of the venous liver CT data is 1.
As shown in fig. 4, the embodiment provides a method for obtaining a plurality of sets of arterial phase CT slice data, a plurality of sets of portal phase CT slice data, a plurality of sets of venous phase CT slice data, and a plurality of sets of delay phase CT slice data, which includes:
s31, acquiring the slice sequence of the arterial liver CT data, the slice sequence of the portal liver CT data, the slice sequence of the venous liver CT data and the slice sequence of the delay liver CT data based on the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delay liver CT data.
Alternatively, the slice order of the arterial liver CT data may refer to the arrangement order of the arterial CT slices in the arterial liver CT data, for example, the arterial liver CT data includes 200 arterial CT slices, the slice order of the arterial liver CT data may be increased by 1 according to the slice number order of the arterial CT slices, for example, the slice order of the first arterial CT slice is 1, the slice order of the second arterial CT slice is 2, and so on, the slice order of the second hundred arterial CT slice data is 200. Similarly, the slice order of the portal liver CT data, the slice order of the venous liver CT data and the slice order of the delay liver CT data are similar to the slice order principle of the arterial liver CT data, and will not be described again here.
S32, performing slice screening processing on the four-stage liver CT data based on the slice sequence of the four-stage liver CT data and the grouping number of the four-stage liver CT data so as to acquire the arterial phase CT slice data, the portal phase CT slice data, the venous phase CT slice data and the delay phase CT slice data.
Alternatively, the slice screening process may refer to a process of dividing the four-stage liver CT data into four-stage CT slice data of different groups, for example, when the number of groups of the arterial liver CT data is 5 and the number of layers of the arterial liver CT data is 200, the arterial liver CT data may be divided into 5 groups of the arterial CT slice data, I, II, III, IV, V groups may be respectively, and the arterial CT slice data of group I may be in a slice order The difference value of the slice sequence of the adjacent arterial phase CT slices in the arterial phase CT slice data is 5, the difference value is the grouping number of the arterial phase liver CT data, and the other data groups of the II, III, IV and V groups are not repeated in the embodiment, and are similar to the data structures in the I group. In addition, the data structures of the portal CT slice data, the venous CT slice data, and the delay CT slice data are similar to the principle of the arterial CT packet data, so that description is omitted for the sake of saving the description space, and the description is omitted.
Optionally, the order difference is an absolute value of a difference between slice orders of two adjacent slices, and the two adjacent slices in the arterial CT slice data may refer to two slices with the smallest absolute value of the difference between slice orders in the arterial CT slice data. For example, the arterial phase CT slice data includes an arterial phase CT slice with a slice order of 1, an arterial phase CT slice with a slice order of 6, and an arterial phase CT slice with a slice order of 11, the arterial phase CT slice with a slice order of 1 and the arterial phase CT slice with a slice order of 6 are two adjacent slices, and the arterial phase CT slice with a slice order of 6 and the arterial phase CT slice with a slice order of 11 are two adjacent slices.
Referring to fig. 5, the present embodiment provides a data alignment preprocessing method for AI training of liver-enhanced multi-stage CT data, including:
s41, acquiring arterial liver CT data to be detected, portal liver CT data to be detected, venous liver CT data to be detected and delay liver CT data to be detected.
Optionally, the arterial liver CT data to be detected, the portal liver CT data to be detected, the venous liver CT data to be detected, and the delayed liver CT data to be detected are arterial liver CT data to be detected, portal liver CT data to be detected, venous liver CT data to be detected, and delayed liver CT data to be detected.
S42, detecting the arterial liver CT data to be detected, the portal liver CT data to be detected, the venous liver CT data to be detected and the delay liver CT data to be detected by using an artificial intelligent medical detection model, wherein the artificial intelligent medical detection model is trained by adopting the method shown in FIG. 2.
Referring to fig. 6, the embodiment provides a method for detecting arterial liver CT data, portal liver CT data, venous liver CT data and delayed liver CT data by using an artificial intelligent medical detection model, including:
S51, the artificial intelligent medical detection model is utilized to process the arterial liver CT data to be detected, the portal liver CT data to be detected, the venous liver CT data to be detected and the delay liver CT data to be detected so as to obtain an intersection ratio.
Alternatively, the intersection ratio may refer to an index for measuring the degree of overlap of the prediction frame of the artificial intelligent hospital detection model with the actual target frame of the four-phase liver CT data.
S52, obtaining a prediction result based on the intersection ratio.
Optionally, the implementation method for obtaining the prediction result based on the intersection ratio value includes: and if the cross ratio is larger than the cross ratio threshold, the prediction result is correct, and if the cross ratio is not larger than the cross ratio threshold, the prediction result is wrong. The cross ratio threshold may be flexibly set according to practical situations, which is not limited in this embodiment.
The protection scope of the data alignment preprocessing method for liver enhancement multi-stage CT data AI training in the embodiment of the application is not limited to the step execution sequence listed in the embodiment, and all the schemes realized by step increase and decrease and step replacement of the prior art according to the principles of the application are included in the protection scope of the application.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus or method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of modules/units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple modules or units may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or modules or units, which may be in electrical, mechanical or other forms.
The modules/units illustrated as separate components may or may not be physically separate, and components shown as modules/units may or may not be physical modules, i.e., may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules/units may be selected according to actual needs to achieve the objectives of the embodiments of the present application. For example, functional modules/units in various embodiments of the application may be integrated into one processing module, or each module/unit may exist alone physically, or two or more modules/units may be integrated into one module/unit.
Those of ordinary skill would further appreciate that the elements and algorithm steps of the examples described in connection with the embodiments disclosed herein may be embodied in electronic hardware, in computer software, or in a combination of the two, and that the elements and steps of the examples have been generally described in terms of function in the foregoing description to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The embodiment provides an electronic device, which comprises a memory, wherein a computer program is stored in the memory; and the processor is in communication connection with the memory and executes the method shown in fig. 2 when the computer program is called. And a display communicatively coupled to the processor and the memory for displaying a GUI interactive interface associated with the method of fig. 2.
The embodiment of the application also provides a computer readable storage medium. Those of ordinary skill in the art will appreciate that all or part of the steps in the method implementing the above embodiments may be implemented by a program to instruct a processor, where the program may be stored in a computer readable storage medium, where the storage medium is a non-transitory (non-transitory) medium, such as a random access memory, a read only memory, a flash memory, a hard disk, a solid state disk, a magnetic tape (magnetic tape), a floppy disk (floppy disk), an optical disk (optical disk), and any combination thereof. The storage media may be any available media that can be accessed by a computer or a data storage device such as a server, data center, or the like that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a digital video disc (digital video disc, DVD)), or a semiconductor medium (e.g., a Solid State Disk (SSD)), or the like.
Embodiments of the present application may also provide a computer program product comprising one or more computer instructions. When the computer instructions are loaded and executed on a computing device, the processes or functions in accordance with embodiments of the present application are fully or partially developed. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center by a wired (e.g., coaxial cable, fiber optic, digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.).
The computer program product is executed by a computer, which performs the method according to the preceding method embodiment. The computer program product may be a software installation package, which may be downloaded and executed on a computer in case the aforementioned method is required.
The descriptions of the processes or structures corresponding to the drawings have emphasis, and the descriptions of other processes or structures may be referred to for the parts of a certain process or structure that are not described in detail.
The above embodiments are merely illustrative of the principles of the present application and its effectiveness, and are not intended to limit the application. Modifications and variations may be made to the above-described embodiments by those skilled in the art without departing from the spirit and scope of the application. Accordingly, it is intended that all equivalent modifications and variations of the application be covered by the claims, which are within the ordinary skill of the art, be within the spirit and scope of the present disclosure.
Claims (7)
1. A data alignment preprocessing method for AI training of liver-enhanced multi-stage CT data, comprising:
acquiring arterial liver CT data, portal liver CT data, venous liver CT data and delay liver CT data;
the method for obtaining the number of packets of the arterial liver CT data, the number of packets of the portal liver CT data and the number of packets of the delay liver CT data based on the layer thickness value of the arterial liver CT data, the layer thickness value of the portal liver CT data, the layer thickness value of the venous liver CT data and the layer thickness value of the delay liver CT data includes: acquiring a first layer thickness multiple, a second layer thickness multiple, a third layer thickness multiple, a fourth layer thickness multiple and a fifth layer thickness multiple based on the layer thickness value of the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delayed liver CT data, wherein the fifth layer thickness multiple is the minimum value of the first layer thickness multiple, the second layer thickness multiple, the third layer thickness multiple and the fourth layer thickness multiple; based on the first layer thickness multiple, the second layer thickness multiple, the third layer thickness multiple, the fourth layer thickness multiple, and the fifth layer thickness multiple, obtaining a packet number of the arterial liver CT data, a packet number of the portal liver CT data, a packet number of the venous liver CT data, and a packet number of the delay liver CT data, the first layer thickness multiple is:
The second layer thickness multiple is:
the third layer thickness multiple is:
the fourth layer thickness multiple is:
wherein la represents a layer thickness value of the arterial liver CT data, lb represents a layer thickness value of the portal liver CT data, lc represents a layer thickness value of the venous liver CT data, ld represents a layer thickness value of the delayed liver CT data, ya represents the first layer thickness multiple, yb represents the second layer thickness multiple, yc represents the third layer thickness multiple, yd represents the fourth layer thickness multiple, z represents the fifth layer thickness multiple, and the fifth layer thickness multiple is represented as: z=min (ya, yb, yc, yd), the number of packets of arterial liver CT data is expressed as:
the number of packets of portal liver CT data is expressed as:
the number of packets of venous liver CT data is expressed as:
the number of packets of the delay period liver CT data is expressed as:
wherein za represents the packet number of the arterial liver CT data, zb represents the packet number of the portal liver CT data, zc represents the packet number of the venous liver CT data, zd represents the packet number of the delay liver CT data, z represents the fifth layer thickness multiple, and round up () is an upward rounding function;
Acquiring a plurality of groups of arterial phase CT slice data, a plurality of groups of portal phase CT slice data, a plurality of groups of venous phase CT slice data and a plurality of groups of delay phase CT slice data based on the four-phase liver CT data and the group number of the four-phase liver CT data, wherein the number of slices of arterial phase CT slice data, the number of slices of portal phase CT slice data, the number of slices of venous phase CT slice data and the number of slices of delay phase CT slice data are mutually aligned, and the number of slices of arterial phase CT slice data, the number of slices of portal phase CT slice data, the number of slices of venous phase CT slice data and the number of slices of delay phase CT slice data are mutually aligned and mean that the number of slices in each group of arterial phase CT slice data, the number of slices of venous phase CT slice data and the number of delay phase CT slice data are identical;
and training the artificial intelligent medical detection model based on the combined data of the arterial phase CT slice data, the portal phase CT slice data, the venous phase CT slice data and the delay phase CT slice data to obtain a trained artificial intelligent medical detection model.
2. The method as recited in claim 1, further comprising: preprocessing the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delay liver CT data to obtain preprocessed arterial liver CT data, preprocessed portal liver CT data, preprocessed venous liver CT data and preprocessed delay liver CT data, wherein each layer of preprocessed arterial liver CT data, each layer of preprocessed portal liver CT data, each layer of preprocessed venous liver CT data and each layer of preprocessed delay liver CT data all comprise liver information.
3. The method of claim 1, wherein the method of obtaining sets of arterial phase CT slice data, sets of portal phase CT slice data, sets of venous phase CT slice data, and sets of delay phase CT slice data comprises:
acquiring a slice sequence of arterial liver CT data, a slice sequence of portal liver CT data, a slice sequence of venous liver CT data and a slice sequence of delay liver CT data based on the arterial liver CT data, the portal liver CT data, the venous liver CT data and the delay liver CT data;
And performing slice screening processing on the four-stage liver CT data based on the slice sequence of the four-stage liver CT data and the grouping number of the four-stage liver CT data so as to acquire the arterial stage CT slice data, the portal stage CT slice data, the venous stage CT slice data and the delay stage CT slice data.
4. The method of claim 1, wherein the arterial liver CT data, the portal liver CT data, the venous liver CT data, and the delayed liver CT data are stored in a medical digital imaging and communication file.
5. The method of claim 1, wherein the artificial intelligence medical test model is a neural network model for medical testing.
6. A data alignment preprocessing method for AI training of liver-enhanced multi-stage CT data, comprising:
acquiring arterial liver CT data to be detected, portal liver CT data to be detected, venous liver CT data to be detected and delay liver CT data to be detected;
detecting the arterial liver CT data to be detected, the portal liver CT data to be detected, the venous liver CT data to be detected and the delayed liver CT data to be detected by using an artificial intelligent medical detection model, wherein the artificial intelligent medical detection model is trained by adopting the method of any one of claims 1 to 5.
7. The method according to claim 6, wherein the implementation method for detecting the arterial liver CT data to be detected, the portal liver CT data to be detected, the venous liver CT data to be detected and the delayed liver CT data to be detected by using an artificial intelligence medical detection model comprises:
processing the arterial liver CT data to be detected, the portal liver CT data to be detected, the venous liver CT data to be detected and the delay liver CT data to be detected by using the artificial intelligent medical detection model so as to obtain an intersection ratio;
and obtaining a prediction result based on the intersection ratio.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310827251.6A CN116563358B (en) | 2023-07-07 | 2023-07-07 | Data alignment preprocessing method for liver enhancement multi-stage CT data AI training |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310827251.6A CN116563358B (en) | 2023-07-07 | 2023-07-07 | Data alignment preprocessing method for liver enhancement multi-stage CT data AI training |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116563358A CN116563358A (en) | 2023-08-08 |
| CN116563358B true CN116563358B (en) | 2023-09-22 |
Family
ID=87491874
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310827251.6A Active CN116563358B (en) | 2023-07-07 | 2023-07-07 | Data alignment preprocessing method for liver enhancement multi-stage CT data AI training |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116563358B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110513089A (en) * | 2019-09-06 | 2019-11-29 | 中国海洋石油集团有限公司 | A kind of Offshore Heavy Oil Field thermal recovery, which is handled up, develops production capacity multiple and determines method |
| CN111402207A (en) * | 2020-03-02 | 2020-07-10 | 中山大学附属第一医院 | Ultrasonic angiography video data analysis method based on composite neural network |
| CN112785605A (en) * | 2021-01-26 | 2021-05-11 | 西安电子科技大学 | Multi-temporal CT image liver tumor segmentation method based on semantic migration |
| CN112951381A (en) * | 2020-05-28 | 2021-06-11 | 福州宜星大数据产业投资有限公司 | Registration method of liver CT and MRI multi-stage enhancement examination images |
| CN113706644A (en) * | 2021-03-04 | 2021-11-26 | 腾讯科技（深圳）有限公司 | Image processing method, image processing apparatus, and storage medium |
| CN113935951A (en) * | 2021-09-13 | 2022-01-14 | 南京邮电大学 | Three-channel cascade SEU-Nets liver tumor segmentation method |
| CN114511599A (en) * | 2022-01-20 | 2022-05-17 | 推想医疗科技股份有限公司 | Model training method and device, medical image registration method and device |
| CN115546149A (en) * | 2022-10-09 | 2022-12-30 | 推想医疗科技股份有限公司 | Liver segmentation method and device, electronic device and storage medium |
| CN115619794A (en) * | 2022-05-11 | 2023-01-17 | 复旦大学 | Neural network-based primary hepatocellular carcinoma lesion segmentation system and method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007070709A2 (en) * | 2005-12-15 | 2007-06-21 | The Trustees The University Of Pennesylvania | Skeletal site -specific characterization of orofacial and iliac crest human bone marrow tromal cells in same individuals |
-
2023
- 2023-07-07 CN CN202310827251.6A patent/CN116563358B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110513089A (en) * | 2019-09-06 | 2019-11-29 | 中国海洋石油集团有限公司 | A kind of Offshore Heavy Oil Field thermal recovery, which is handled up, develops production capacity multiple and determines method |
| CN111402207A (en) * | 2020-03-02 | 2020-07-10 | 中山大学附属第一医院 | Ultrasonic angiography video data analysis method based on composite neural network |
| CN112951381A (en) * | 2020-05-28 | 2021-06-11 | 福州宜星大数据产业投资有限公司 | Registration method of liver CT and MRI multi-stage enhancement examination images |
| CN112785605A (en) * | 2021-01-26 | 2021-05-11 | 西安电子科技大学 | Multi-temporal CT image liver tumor segmentation method based on semantic migration |
| CN113706644A (en) * | 2021-03-04 | 2021-11-26 | 腾讯科技（深圳）有限公司 | Image processing method, image processing apparatus, and storage medium |
| CN113935951A (en) * | 2021-09-13 | 2022-01-14 | 南京邮电大学 | Three-channel cascade SEU-Nets liver tumor segmentation method |
| CN114511599A (en) * | 2022-01-20 | 2022-05-17 | 推想医疗科技股份有限公司 | Model training method and device, medical image registration method and device |
| CN115619794A (en) * | 2022-05-11 | 2023-01-17 | 复旦大学 | Neural network-based primary hepatocellular carcinoma lesion segmentation system and method |
| CN115546149A (en) * | 2022-10-09 | 2022-12-30 | 推想医疗科技股份有限公司 | Liver segmentation method and device, electronic device and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| Deep Learning-Based Radiomics Models for Early Recurrence Prediction of Hepatocellular Carcinoma with Multi-phase CT Images and Clinical Data;Weibin WANG et al.;《2019 IEEE》;第4881-4884页 * |
| Pulmonary Nodule Volumetric Measurement Variability as a Function of CT Slice Thickness and Nodule Morphology;Myria Petrou et al.;《American Roentgen Ray Society》;第306-312页 * |
| 基于边缘文理特征的多时相肝脏图像配准研究;谭晓敏;《中国优秀硕士学位论文全文数据库 医药卫生科技辑》;E064-34 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116563358A (en) | 2023-08-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10991100B2 (en) | Disease detection algorithms trainable with small number of positive samples | |
| JP7350878B2 (en) | Image analysis method, device, program | |
| US20170358079A1 (en) | Systems, methods and devices for analyzing quantitative information obtained from radiological images | |
| CN112868020A (en) | System and method for improved analysis and generation of medical imaging reports | |
| EP3791325A1 (en) | Systems and methods for detecting an indication of a visual finding type in an anatomical image | |
| US20200194108A1 (en) | Object detection in medical image | |
| US11776243B2 (en) | Systems and methods for automated detection of visual objects in medical images | |
| US10789712B2 (en) | Method and system for image analysis to detect cancer | |
| US10290101B1 (en) | Heat map based medical image diagnostic mechanism | |
| Lee et al. | The current state of artificial intelligence in medical imaging and nuclear medicine | |
| US20220051396A1 (en) | Cross modality training of machine learning models | |
| CN106326416B (en) | DICOM image display method and device | |
| WO2017164204A1 (en) | Method and apparatus for extracting diagnosis object from medical document | |
| CN113658175A (en) | Method and device for determining symptom data | |
| KR20230021665A (en) | Systems and methods for processing electronic images to determine salient information in digital pathology | |
| CN116563358B (en) | Data alignment preprocessing method for liver enhancement multi-stage CT data AI training | |
| WO2019109410A1 (en) | Fully convolutional network model training method for splitting abnormal signal region in mri image | |
| Li et al. | Contrast enhancement for cDNA microarray image based on fourth-order moment | |
| Robben et al. | The state of applying artificial intelligence to tissue imaging for cancer research and early detection | |
| Chang et al. | Primer for image informatics in personalized medicine | |
| US11315248B2 (en) | Identifying invalid medical images | |
| CN116071555B (en) | Method for establishing WMHs segmentation model, WMHs segmentation method and device | |
| US20240145068A1 (en) | Medical image analysis platform and associated methods | |
| US20230196583A1 (en) | Systems and methods to process electronic images to identify abnormal morphologies | |
| US20220245427A1 (en) | Identifying a finding in a dataset using a machine learning model ensemble |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |