CN115527657A - Image and image multi-mode reconstruction, imaging and labeling based on medical digital imaging and communication - Google Patents
Image and image multi-mode reconstruction, imaging and labeling based on medical digital imaging and communication Download PDFInfo
- Publication number
- CN115527657A CN115527657A CN202211262218.5A CN202211262218A CN115527657A CN 115527657 A CN115527657 A CN 115527657A CN 202211262218 A CN202211262218 A CN 202211262218A CN 115527657 A CN115527657 A CN 115527657A
- Authority
- CN
- China
- Prior art keywords
- image
- imaging
- information
- communication
- labeling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003384 imaging method Methods 0.000 title claims abstract description 51
- 238000002372 labelling Methods 0.000 title claims abstract description 35
- 238000004891 communication Methods 0.000 title claims abstract description 29
- 238000012545 processing Methods 0.000 claims abstract description 19
- 230000005540 biological transmission Effects 0.000 claims abstract description 12
- 230000006835 compression Effects 0.000 claims abstract description 5
- 238000007906 compression Methods 0.000 claims abstract description 5
- 238000012800 visualization Methods 0.000 claims abstract description 5
- 238000006243 chemical reaction Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 44
- 230000000007 visual effect Effects 0.000 claims description 18
- 230000009466 transformation Effects 0.000 claims description 17
- 238000001914 filtration Methods 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 7
- 230000003902 lesion Effects 0.000 claims description 7
- 238000005259 measurement Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 238000009877 rendering Methods 0.000 claims description 4
- 238000011160 research Methods 0.000 claims description 4
- 229910000831 Steel Inorganic materials 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 230000015556 catabolic process Effects 0.000 claims description 3
- 238000013499 data model Methods 0.000 claims description 3
- 238000013500 data storage Methods 0.000 claims description 3
- 238000006731 degradation reaction Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 238000003709 image segmentation Methods 0.000 claims description 3
- 238000012067 mathematical method Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000004451 qualitative analysis Methods 0.000 claims description 3
- 238000004445 quantitative analysis Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000002441 reversible effect Effects 0.000 claims description 3
- 238000003045 statistical classification method Methods 0.000 claims description 3
- 239000010959 steel Substances 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 230000002708 enhancing effect Effects 0.000 claims description 2
- 238000000513 principal component analysis Methods 0.000 claims 2
- 238000003745 diagnosis Methods 0.000 abstract description 3
- 230000011218 segmentation Effects 0.000 abstract description 3
- 230000010365 information processing Effects 0.000 abstract description 2
- 210000000056 organ Anatomy 0.000 abstract description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/10—Image enhancement or restoration using non-spatial domain filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T9/00—Image coding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/42—Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Radiology & Medical Imaging (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention relates to the technical field of medical information processing, and discloses multi-modal reconstruction of images and images based on medical digital imaging and communication, which comprises the following steps: s1: the image acquisition mainly comprises the steps of image taking, photoelectric conversion, digitization and the like by converting an analog image or an image into a digital signal suitable for being processed by a computer or digital equipment. The image, the image multi-modal reconstruction, the imaging and the labeling based on the medical digital imaging and the communication are realized by collecting the image to perform a series of processing such as compression, enhancement, restoration and segmentation on the image, so that the image is displayed in a three-dimensional form after being reconstructed, the transmission stability of the image is improved, and meanwhile, the diagnosis and treatment of doctors can be assisted, a three-dimensional geometric model can be constructed according to image sequences such as CR CT DR MR MRI and the like through medical image visualization, invisible human organs are truly displayed in a three-dimensional form, and the images can be arbitrarily amplified, reduced, rotated, contrasted and adjusted and the like.
Description
Technical Field
The invention relates to the technical field of medical information processing, in particular to multi-modal reconstruction, imaging and labeling of images and images based on medical digital imaging and communication.
Background
In recent years, the large-scale growth of digital medical images provides a data base for promoting research and application of medical image processing technology represented by a deep neural network, but a large-scale and reliable reference data set is lacked in the field of medical images so as to influence the development of deep learning for medical image data.
The reason why the accurate and reliable reference data set is lacking in the field of the current medical images is that due to the specialty and complexity of the medical images, different experts have great diversity when manually labeling the medical images, and labeling results often have differences, so that the manual labeling of the medical images becomes more difficult and expensive, and visibly.
In the field of medical imaging, lesion delineation (lesion segmentation) is a key preprocessing mode for automatic measurement and diagnostic analysis of lesions by using machine learning, and is currently applied to various medical works in a large quantity, however, the current lesion delineation mode has obvious disadvantages: the manual labeling needs an experienced doctor to judge and trace the boundary, the process is complicated and time-consuming, the doctor is difficult to extract time to complete labeling work in the heavy work, and great challenges are brought to data set collection work, so that the problems are solved by image and image multi-mode reconstruction, imaging and labeling based on medical digital imaging and communication.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the prior art, the invention provides the multi-mode reconstruction, imaging and labeling of images and images based on medical digital imaging and communication, has the advantages of high labeling efficiency, stable transmission of the images and the like, and solves the problems of low labeling efficiency and quality and unstable transmission of the images.
(II) technical scheme
In order to achieve the purposes of high labeling efficiency and stable image transmission, the invention provides the following technical scheme: the multi-mode image and image reconstruction based on medical digital imaging and communication comprises the following steps:
s1: the image file processing device comprises an image file processing device, an image shooting device, a computer and a digital device, wherein the image file processing device is used for acquiring an image file, converting the image file into a digital signal suitable for being processed by the computer or the digital device, mainly comprising the steps of image shooting, photoelectric conversion, digitization and the like, and establishing communication connection with the image shooting device, so that data transmission can be realized between the image file processing device and the image shooting device based on medical digital imaging and a communication protocol;
s2: the image compression and coding achieves the purpose of reducing the data storage capacity by deleting redundant or unnecessary information, and the main methods of the image coding comprise redundancy removing coding, transformation coding, wavelet transformation coding, neural network coding, model base coding and the like;
s3: image enhancement: enhancing the information of interest in the image, removing or attenuating the unnecessary information, and facilitating the target differentiation or object interpretation;
s4: image restoration: the imaging system is influenced by various factors, so that the image quality is reduced, the basic degradation expression is image blurring, and the blurring and noise interference can be removed by a restoration method;
s5: image transformation, which can be performed by fourier transform, walsh-hadamard transform, discrete cosine transform, hotelling transform, discrete kaffner-lewy transform, hadamard transform, etc. by performing two-dimensional linear reversible transform on an original image with an orthogonal function or an orthogonal matrix;
s6: image segmentation: dividing the image into mutually disjoint areas with characteristics, extracting an interested target and providing a basis for quantitative and qualitative analysis;
s7: the image recognition method comprises the steps of processing, analyzing and understanding an image, wherein the recognition method can be roughly divided into a statistical classification method, a syntax (structure) recognition method and a fuzzy recognition method, the statistical recognition method mainly extracts image feature research and can be realized by a Bayesian classifier, a neural network and a support vector machine, and the syntax recognition is used for analyzing the structure of an image mode and is realized by syntax analysis or a corresponding automatic machine; the fuzzy recognition method introduces a fuzzy mathematical method into image recognition;
s8: image reconstruction, namely reconstructing an image by data acquired by detecting an object, wherein the data for reconstructing the image is generally acquired in time-sharing and step-by-step mode, and performing multi-mode reconstruction on the image by three steps of a transmission model (light and x-ray), an emission model (MRI, PET and the like) and a reflection model (photoelectron, radar and ultrasonic wave), and the processing result can be stored in a storage space after reconstruction is finished;
s9: image labeling: acquiring a target CT MR MRI DR image to be marked; marking operation on the boundary of the target ultrasonic image, and displaying boundary marking information; determining a target image area based on the boundary marking information; extracting regional characteristic information of the target image region, and extracting global characteristic information of the target ultrasonic image; outputting focus marking information based on the region characteristic information and the global characteristic information, wherein the focus marking information is used for distinguishing focus regions and non-focus regions in the target ultrasonic image;
s10: and (3) lesion measurement arbitration: by identifying a target CT MR MRI DR image; drawing the target focus area, and establishing a boundary marking area; the method comprises the steps of carrying out regional measurement, steel amount, length, precision and dimensionality through a selected regional frame, establishing a data test model, forming regional layer thickness calculation through the data model, forming conclusions of the volume, area, layer thickness, layer height and other embodied areas of a focus through an artificial intelligence fusion calculation method, and providing a service flow for advanced arbitration, auditing and approval.
Preferably, the main method in step S3 includes: histogram enhancement, spatial and frequency domain enhancement, pseudo color enhancement, and the like.
Preferably, the restoration implementation method in step S4 includes wiener filtering, inverse filtering, homomorphic filtering, least constrained quadratic filtering, and the like.
Preferably, the main purpose of the three-dimensional visualization in step S8 is to use computer graphics technology to intuitively represent three-dimensional effects, so as to provide structural information that cannot be obtained by conventional means, and the algorithms thereof include two major categories, i.e., surface rendering and volume rendering.
Another technical problem to be solved by the present invention is to provide an imaging and labeling method based on medical digital imaging and communication, which comprises the following steps:
1) For the imaging local feature extraction, a subset of a plurality of randomly selected secondary images in an image database is used as a training sample of a dictionary, a certain number of NxN-sized image subblocks (namely Patch) are extracted from the selected images in a random or grid mode, and the extracted image subblocks are arranged into a 1xN image subblock according to rows or columns 2 The feature vector of (2);
2) Secondly, in order to remove noise and reduce calculated amount, PCA dimensionality reduction is carried out on the normalized Patch features, feature vectors corresponding to a plurality of feature values with the largest values are selected as principal elements, and the principal elements are used as the basis of features in an image expression stage;
3) The further operation is that the obtained local features are clustered to construct a visual dictionary, the position information and the visual information of the Patch are directly combined into the local features, and then the K mean value is clustered to generate the visual dictionary, so that a good classification and labeling effect is achieved;
4) After the visual dictionary is obtained, the image feature labels of each image can be generated, similar to the first step when the Patch is extracted, but dense grids are adopted to select the Patch, and the visual word corresponding to each Patch is calculated.
Preferably, in step 1), in order to eliminate noise and enhance information included in Patch, the gray-level value of the Patch feature is normalized to have a mean value of 0 and a variance of 1.
Preferably, in step 4), in order to utilize the spatial information of each Patch, a spatial pyramid feature representation method is adopted to obtain a high-dimensional histogram of the visual words.
(III) advantageous effects
Compared with the prior art, the invention provides multi-mode reconstruction, imaging and labeling of images and images based on medical digital imaging and communication, and the invention has the following beneficial effects:
1. the image, the image multi-modal reconstruction, the imaging and the labeling based on the medical digital imaging and the communication are realized by collecting the image to perform a series of processing such as compression, enhancement, restoration and segmentation on the image, so that the image is displayed in a three-dimensional form after being reconstructed, the transmission stability of the image is improved, and meanwhile, the diagnosis and treatment of doctors can be assisted, a three-dimensional geometric model can be constructed according to image sequences such as CR CT DR MR MRI and the like through medical image visualization, invisible human organs are truly displayed in a three-dimensional form, and the images can be arbitrarily amplified, reduced, rotated, contrasted and adjusted and the like.
2. The image and image multi-mode reconstruction, imaging and labeling based on medical digital imaging and communication show good classification performance through effectiveness of the Patch combined with position information in medical images, so that the characteristic extraction and labeling speed is high, and the information labeling efficiency and quality of the medical images are improved.
Drawings
FIG. 1 is a flow chart of a method for multi-modal reconstruction, imaging and labeling of images and images based on medical digital imaging and communication according to the present invention;
fig. 2 is a frame diagram of multi-modal image and image reconstruction, imaging and labeling based on medical digital imaging and communication according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The multi-mode image and image reconstruction based on medical digital imaging and communication comprises the following steps:
s1: the image file processing device comprises an image file processing device, an image shooting device, a computer and a digital device, wherein the image file processing device is used for acquiring an image file, converting the image file into a digital signal suitable for being processed by the computer or the digital device, mainly comprising the steps of image shooting, photoelectric conversion, digitization and the like, and establishing communication connection with the image shooting device, so that data transmission can be realized between the image file processing device and the image shooting device based on medical digital imaging and a communication protocol;
s2: the image compression and coding achieves the purpose of reducing the data storage capacity by deleting redundant or unnecessary information, and the main methods of the image coding comprise redundancy removing coding, transformation coding, wavelet transformation coding, neural network coding, model base coding and the like;
s3: image enhancement: the method is characterized in that the information of interest in the image is enhanced, and the unnecessary information is removed or attenuated, so that the target distinguishing or the object interpretation is facilitated, and the main methods are as follows: histogram enhancement, spatial and frequency domain enhancement, pseudo-color enhancement and the like;
s4: image restoration: the imaging system is influenced by various factors, so that the image quality is reduced, the basic degradation expression is image blurring, blurring and noise interference can be removed by a restoration method, and the restoration realization method comprises wiener filtering, inverse filtering, homomorphic filtering, minimum constraint quadratic filtering and the like;
s5: image transformation, which can be performed by fourier transform, walsh-hadamard transform, discrete cosine transform, hotelling transform, discrete kaffner-lewy transform, hadamard transform, etc. by performing two-dimensional linear reversible transform on an original image with an orthogonal function or an orthogonal matrix;
s6: image segmentation: dividing the image into mutually disjoint areas with characteristics, extracting an interested target and providing a basis for quantitative and qualitative analysis;
s7: the image recognition method comprises the steps of processing, analyzing and understanding an image, wherein the recognition method can be roughly divided into a statistical classification method, a syntactic (structure) recognition method and a fuzzy recognition method, the statistical recognition method mainly extracts image feature research and can be realized by a Bayes classifier, a neural network and a support vector machine, and the syntactic recognition is used for analyzing the structure of an image mode and is realized by syntactic analysis or a corresponding automatic machine; the fuzzy recognition method introduces a fuzzy mathematical method into image recognition;
s8: the method comprises the steps of image reconstruction, image reconstruction and image processing, wherein the image reconstruction data are generally acquired in a time-sharing and step-by-step mode, multimode reconstruction is carried out on the image through three steps of a transmission model (light and x rays), an emission model (MRI, PET and the like) and a reflection model (photoelectron, radar and ultrasonic waves), a processing result can be stored into a storage space after reconstruction is finished, and the main purpose of three-dimensional visualization is to use a computer graphics technology to intuitively express a three-dimensional effect, so that structural information which cannot be acquired by the traditional means is provided, and an algorithm of the method comprises two major categories of a surface drawing method and a body drawing method;
s9: image labeling: acquiring a target CT MR MRI DR image to be marked; marking operation on the boundary of the target ultrasonic image, and displaying boundary marking information; determining a target image area based on the boundary marking information; extracting regional characteristic information of the target image region, and extracting global characteristic information of the target ultrasonic image; outputting focus marking information based on the regional characteristic information and the global characteristic information, wherein the focus marking information is used for distinguishing a focus region and a non-focus region in the target ultrasonic image;
s10: and (3) lesion measurement arbitration: by identifying a target CT MR MRI DR image; drawing the target focus area, and establishing a boundary marking area; the method comprises the steps of carrying out regional measurement, steel amount, length, precision and dimensionality through a selected regional frame, establishing a data test model, forming regional layer thickness calculation through the data model, forming conclusions of the volume, area, layer thickness, layer height and other embodied areas of a focus through an artificial intelligence fusion calculation method, and providing a service flow for advanced arbitration, auditing and approval.
Imaging and labeling based on medical digital imaging and communication, comprising the steps of:
1) For local feature extraction, a subset of randomly selected sub-images in an image database is used as a training sample of a dictionary, a certain number of NxN image sub-blocks (i.e., patch) are extracted from the selected images in a random or grid manner, and the extracted sub-blocks are arranged into a 1xN image sub-block according to rows or columns 2 In order to eliminate noise and enhance the information contained in Patch, the gray value of the Patch feature is normalized to mean 0 and variance 1;
2) Secondly, in order to remove noise and reduce calculated amount, PCA dimensionality reduction is carried out on the normalized Patch features, feature vectors corresponding to a plurality of feature values with the largest values are selected as principal elements, and the principal elements are used as the basis of features in an image expression stage;
3) The further operation is that the obtained local features are clustered to construct a visual dictionary, the position information and the visual information of the Patch are directly combined into the local features, and then the K mean value is clustered to generate the visual dictionary, so that a good classification and labeling effect is achieved;
4) After the visual dictionary is obtained, generating image feature labels for each image, extracting Patch, selecting Patch by adopting dense grids, calculating a visual word corresponding to each Patch, and obtaining a high-dimensional histogram of the visual word by adopting a spatial pyramid feature representation method in order to utilize spatial information of each Patch.
The invention has the beneficial effects that: the image is compressed, enhanced, restored and segmented by collecting the image, so that the image is displayed in a three-dimensional form after being reconstructed, the transmission stability of the image is improved, and the diagnosis and treatment of doctors can be assisted.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (7)
1. The multi-mode image and image reconstruction based on medical digital imaging and communication is characterized by comprising the following steps of:
s1, obtaining an image, converting an analog image or an image into a digital signal suitable for being processed by a computer or a digital device, mainly comprising the steps of image shooting, photoelectric conversion, digitalization and the like, and establishing communication connection with an image shooting device to enable data transmission between the image file processing device and the image shooting device to be realized based on medical digital imaging and a communication protocol;
s2: the image compression and coding achieves the purpose of reducing the data storage capacity by deleting redundant or unnecessary information, and the main methods of the image coding comprise redundancy removing coding, transformation coding, wavelet transformation coding, neural network coding, model base coding and the like;
s3: image enhancement: enhancing the information of interest in the image, removing or attenuating the unwanted information, facilitating the object discrimination or object interpretation;
s4: image restoration: the imaging system is influenced by various factors, so that the image quality is reduced, the basic degradation expression is image blurring, and the blurring and noise interference can be removed by a restoration method;
s5, image transformation, namely performing two-dimensional linear reversible transformation on the original image by using an orthogonal function or an orthogonal matrix, wherein the image transformation can be performed by Fourier transformation, walsh, hadamard transformation, discrete cosine transformation, hotelling transformation, discrete Kaffner-Lewy transformation, hadamard transformation and the like;
s6, image segmentation: dividing the image into mutually disjoint areas with characteristics, extracting an interested target and providing a basis for quantitative and qualitative analysis;
s7: the image recognition method comprises the steps of processing, analyzing and understanding an image, wherein the recognition method can be roughly divided into a statistical classification method, a syntax (structure) recognition method and a fuzzy recognition method, the statistical recognition method mainly extracts image feature research and can be realized by a Bayesian classifier, a neural network and a support vector machine, and the syntax recognition is used for analyzing the structure of an image mode and is realized by syntax analysis or a corresponding automatic machine; the fuzzy recognition method introduces a fuzzy mathematical method into image recognition;
s8: reconstructing an image, namely reconstructing the image by using data acquired by detecting an object, wherein the data for reconstructing the image is generally acquired in a time-sharing and step-by-step manner, performing multi-mode reconstruction on the image by using a transmission model (light and x-ray), an emission model (MRI, PET and the like) and a reflection model (photoelectron, radar and ultrasonic), and storing a processing result into a storage space after reconstruction is finished;
s9: image labeling: acquiring a target CT MR MRI DR image to be marked; marking operation on the boundary of the target ultrasonic image, and displaying boundary marking information; determining a target image area based on the boundary marking information; extracting regional characteristic information of the target image region, and extracting global characteristic information of the target ultrasonic image; outputting focus marking information based on the region characteristic information and the global characteristic information, wherein the focus marking information is used for distinguishing focus regions and non-focus regions in the target ultrasonic image;
s10: and (3) lesion measurement arbitration: identifying a target CT MR MRI DR image; drawing the target focus area, and establishing a boundary marking area; the method comprises the steps of carrying out regional measurement, steel amount, length, precision and dimensionality through a selected regional frame, establishing a data test model, forming regional layer thickness calculation through the data model, forming conclusions of the volume, area, layer thickness, layer height and other performance areas of a focus through an artificial intelligence fusion calculation method, and providing a business process for high-grade arbitration, auditing and approval.
2. The medical digital imaging and communication based imagery, image multi-modal reconstruction, imaging and labeling of claim 1, wherein said step S3 comprises the steps of: histogram enhancement, spatial and frequency domain enhancement, pseudo color enhancement, and the like.
3. The imaging, image multi-modal reconstruction, imaging and labeling based on medical digital imaging and communication as claimed in claim 1, wherein said restoration implementation methods in step S4 include wiener filtering, inverse filtering, homomorphic filtering, least constrained dyadic filtering, etc.
4. The multi-modal image, video and image reconstruction, imaging and labeling based on digital imaging and communication as claimed in claim 1, wherein the main purpose of the three-dimensional visualization in step S8 is to use computer graphics technology to visually represent three-dimensional effects, thereby providing structural information that cannot be obtained by conventional means, and the algorithms thereof include two categories, surface rendering and volume rendering.
5. Imaging and labeling based on medical digital imaging and communication, comprising the steps of:
1) For local imaging feature extraction, a subset of randomly selected sub-images in an image database is used as a training sample of a dictionary, a certain number of NxN image subblocks (i.e., patch) are extracted from the selected images in a random or grid mode and are arranged into a row or a column1xN 2 The feature vector of (2);
2) Secondly, in order to remove noise and reduce calculated amount, PCA (principal component analysis) dimensionality reduction is carried out on the normalized Patch characteristic, characteristic vectors corresponding to a plurality of characteristic values with the largest values are selected as principal elements, and the principal elements are used as bases of characteristic features in an image expression stage;
3) The further operation is that the obtained local features are clustered to construct a visual dictionary, the position information of the Patch and the visual information are directly combined to form the local features, and then the K mean value is clustered to generate the visual dictionary, so that a good classifying and labeling effect is achieved;
4) After the visual dictionary is obtained, the image feature labels of each image can be generated, and the process of extracting Patch is similar to the first step, but dense grids are adopted to select Patch, and the visual words corresponding to each Patch are calculated.
6. An imaging and labeling method based on medical digital imaging and communication as claimed in claim 5, wherein in step 1) the gray values of the Patch features are normalized to mean 0 and variance 1 in order to eliminate noise and enhance the information contained in the Patch.
7. An imaging and labeling method based on medical digital imaging and communication as claimed in claim 5, wherein in step 4), in order to utilize the spatial information of each Patch, we use the spatial pyramid feature representation method to obtain a high-dimensional histogram of visual words.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211262218.5A CN115527657A (en) | 2022-10-14 | 2022-10-14 | Image and image multi-mode reconstruction, imaging and labeling based on medical digital imaging and communication |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211262218.5A CN115527657A (en) | 2022-10-14 | 2022-10-14 | Image and image multi-mode reconstruction, imaging and labeling based on medical digital imaging and communication |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115527657A true CN115527657A (en) | 2022-12-27 |
Family
ID=84701452
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211262218.5A Pending CN115527657A (en) | 2022-10-14 | 2022-10-14 | Image and image multi-mode reconstruction, imaging and labeling based on medical digital imaging and communication |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115527657A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115995287A (en) * | 2023-03-23 | 2023-04-21 | 山东远程分子互联网医院有限公司 | Cloud image data receiving and transmitting system and method |
| CN116485778A (en) * | 2023-04-27 | 2023-07-25 | 鄂东医疗集团市中心医院 | Imaging detection method, imaging detection system, computer equipment and storage medium |
-
2022
- 2022-10-14 CN CN202211262218.5A patent/CN115527657A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115995287A (en) * | 2023-03-23 | 2023-04-21 | 山东远程分子互联网医院有限公司 | Cloud image data receiving and transmitting system and method |
| CN116485778A (en) * | 2023-04-27 | 2023-07-25 | 鄂东医疗集团市中心医院 | Imaging detection method, imaging detection system, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106056595B (en) | Based on the pernicious assistant diagnosis system of depth convolutional neural networks automatic identification Benign Thyroid Nodules | |
| Gu et al. | MedSRGAN: medical images super-resolution using generative adversarial networks | |
| CN112529894B (en) | Thyroid nodule diagnosis method based on deep learning network | |
| CN115527657A (en) | Image and image multi-mode reconstruction, imaging and labeling based on medical digital imaging and communication | |
| Hossain | Microc alcification segmentation using modified u-net segmentation network from mammogram images | |
| CN112150428A (en) | Medical image segmentation method based on deep learning | |
| CN110782427B (en) | Magnetic resonance brain tumor automatic segmentation method based on separable cavity convolution | |
| EP2401719B1 (en) | Methods for segmenting images and detecting specific structures | |
| CN112508953B (en) | Meningioma rapid segmentation qualitative method based on deep neural network | |
| CN112070785A (en) | Medical image analysis method based on computer vision | |
| CN114241077B (en) | CT image resolution optimization method and device | |
| CN113592794A (en) | Spine image segmentation method of 2D convolutional neural network based on mixed attention mechanism | |
| WO2021183765A1 (en) | Automated detection of tumors based on image processing | |
| Shan et al. | SCA-Net: A spatial and channel attention network for medical image segmentation | |
| CN116309647B (en) | Method for constructing craniocerebral lesion image segmentation model, image segmentation method and device | |
| CN116883341A (en) | Liver tumor CT image automatic segmentation method based on deep learning | |
| CN116524315A (en) | Mask R-CNN-based lung cancer pathological tissue section identification and segmentation method | |
| Judice et al. | A Novel assessment of various bio-imaging methods for lung tumor detection and treatment by using 4-D and 2-D CT images | |
| Srilakshmi et al. | Performance enhancement of content based medical image retrieval for MRI brain images based on hybrid approach | |
| CN112967295B (en) | Image processing method and system based on residual network and attention mechanism | |
| Patibandla et al. | CT Image Precise Denoising Model with Edge Based Segmentation with Labeled Pixel Extraction Using CNN Based Feature Extraction for Oral Cancer Detection | |
| CN117476219B (en) | Auxiliary method and auxiliary system for positioning CT (computed tomography) tomographic image based on big data analysis | |
| Sayyad et al. | Data Conditioning for Medical Imaging | |
| CN116051508A (en) | Brain tumor weak supervision segmentation method and system based on adjacent slice reasoning | |
| Dubey | Classification of the Magnetic Resonance Imaging of the Brain Tumor Using the Residual Neural Network Framework |
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 |