CN108922601A - A kind of medical image processing system - Google Patents
A kind of medical image processing system Download PDFInfo
- Publication number
- CN108922601A CN108922601A CN201810746457.5A CN201810746457A CN108922601A CN 108922601 A CN108922601 A CN 108922601A CN 201810746457 A CN201810746457 A CN 201810746457A CN 108922601 A CN108922601 A CN 108922601A
- Authority
- CN
- China
- Prior art keywords
- image
- algorithm
- processing
- module
- image processing
- 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
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
Abstract
The present invention discloses a kind of medical image processing system, including outside to connection module, algorithm template configuration module, Imaging enhanced processing module and image contrast and post-processing module.The present invention improves signal noise ratio (snr) of image by deep learning algorithm, and image noise reduction is rapidly completed and reduces the sense of image waxen imagen, promotes picture quality, to reduce CT radiological dose, reduces MRI imaging time, improves operation rate and working efficiency;It can be with hospital equipment work station and original PACS system seamless interfacing.
Description
Technical field
The invention belongs to image processing technique fields, more particularly to a kind of medical image processing system.
Background technique
Currently, CT examination has been widely used in medical diagnosis on disease treatment and health examination practice.In order to be formed clearly
Medical image needs longer sweep time and CT exit dose, but CT exit dose can generate radiation injury to human body, but reduce CT
Radiation metering can be such that picture quality reduces.Therefore, picture quality is improved, scanning speed is promoted, reduces radiation quality, optimizes the later period
Processing is the most crucial index of the high-end CT manufacturer main goal of attack in the world and all big enterprises' product technology competition.It is advanced in the world
CT manufacturer be all made of multiple iteration method for reconstructing to handle image, but the iterative processing time is long, and operand is big, and processing generates
Picture waxen imagen sense it is prominent, picture quality is not high.
As medical image most significant end, state-of-the-art equipment, the principle of MRI (Magnetic resonance imaging) is that human body is placed on
In one powerful magnetic field, by Hydrogen Proton in radio-frequency pulse exciting human, nuclear magnetic resonance occurs, then receives proton sending
NMR signal constitutes image in all directions using the operation of computer by the positioning in three directions of gradient fields.By
Damage of the ionising radiation to human body is thoroughly got rid of in it, and has parameter more, is contained much information, multi-faceted can be imaged, and to soft
Group is woven with the prominent feature such as high resolution, causes the attention of various aspects scholar once coming out from it, either the improvement of equipment,
The update and upgrading of software, or the research of the diagnostic effect to each position organ of whole body, develop it is quite fast, at present at
It is ripe, it is widely used in the diagnosis of clinical disease, essential inspection method is become to some lesions.
The main deficiency of MRI is that the time needed for it is scanned is longer, usually 15 to 20 minutes, if reducing scanning
Time can be such that picture quality reduces, thus often feel to the inspection of the critical patient of some state of an illness and part claustrophobia patient tired
Difficulty, to motor organ, such as gastrointestinal tract is due to a lack of suitable contrast medium, usually show it is unclear because MRI machine is expensive,
Operation maintenance cost is high, if under the premise of guaranteeing picture quality can be reduced sweep time, improves equipment efficiency of operation, has
Biggish Social benefit and economic benefit.
Reducing CT influences radiological dose, shortens the MRI scan time, mainstream vendor mainly uses iterative reconstruction currently on the market
The algorithm of (Iterative Reconstruction) carries out image noise reduction.Such as GE in the production of CT with ASiR-V and
VEO algorithm carries out noise reduction to the picture of low dosage, so that the picture noise of the picture of low dosage and high dose is constant.Relatively
It answers, there are SAFIRE and ADMIRE in Siemens, and Philip has iDose and IMR, and there is AIDR 3D etc. in Toshiba.
But traditional iterative reconstruction algorithms have two big defects.First is that the calculating time is very long, if noise is dropped to
60%, one picture of processing needs time-consuming 30min or so;Second is that processed picture has waxen imagen sense, it is exactly mould for popular
The feeling of paste causes doctor's mistaken diagnosis it could even be possible to erasing lesion information.
Summary of the invention
To solve the above-mentioned problems, the invention proposes a kind of medical image processing systems, are mentioned by deep learning algorithm
Hi-vision signal-to-noise ratio is rapidly completed image noise reduction and reduces the sense of image waxen imagen, picture quality promoted, to reduce CT radiological agent
Amount reduces MRI imaging time, improves operation rate and working efficiency;It can be with hospital equipment work station and original PACS system
Seamless interfacing.
In order to achieve the above objectives, the technical solution adopted by the present invention is that:A kind of medical image processing system, including it is external right
Connection module, algorithm template configuration module, Imaging enhanced processing module and image contrast and post-processing module;
The outside carries out data double-way transmission, equipment work to connection module between hospital equipment work station and PACS system
Make station or PACS system and image DICOM file is transferred to Imaging enhanced processing module, Imaging enhanced processing module will pass through calculation
File after method optimization is transmitted back to hospital equipment work station and PACS system;
The algorithm template configuration module, according to the corresponding image processing algorithm of Image Matching, described image Processing Algorithm
For deep learning algorithm;
The Imaging enhanced processing module, the first step are that DICOM image preprocessing is parsed according to user demand using HU
Method carries window position and window width information according to DICOM file, pre-processes to DICOM, prominent image emphasis details;Second
Step is that enhancing is handled, and corresponding image processing algorithm improves signal noise ratio (snr) of image, improving image quality, and ensures image definition;
The image contrast and post-processing module, the image transferred before and after the processing compare in the same window, analysis
Image processing is as a result, as diagnosis or the auxiliary and support of process of scientific research.
Further, the outside includes optical fiber or Double-strand transmission data to connection module;At medical image
Image after reason system optimization will directly be distributed to PACS system, the existing system of seamless interfacing hospital;Simultaneously at medical image
Reason system is based on DICOM3.0 standard progress image data transmission process and is digitally signed encryption, ensures the letter of hospital and patient
Breath safety.
Further, using corresponding sample training collection training algorithm model, each position to partes corporis humani position
Algorithm template is constituted equipped with algorithm model alone;The algorithm template configuration module judges according to the image of acquisition and identifies shadow
As type and scanned position select phase according to the type and scanned position of image to match corresponding algorithm template, or by user
The algorithm template answered.
Further, the method for the DICOM Imaging enhanced processing, including step:
Firstly, carrying out data screening to medical image data collected;
Then, data are pre-processed, preparative algorithm experiment sample, experiment sample includes training set and test set;Needle
A variety of neural network frameworks are constructed to different usage scenarios, each frame is trained using training set, obtains a variety of figures
Image intensifying model;
Finally, being tested using test set each image enhancement model, a variety of images are increased by test result
Strong model is improved, until reaching target.
Further, the processing method of the image contrast, image will be put into unified comparison window before and after the processing, use
Original image and processing result are amplified in family or rotation comparison checks that user evaluates processing result.
Further, the method for the Imaging processing:Including multiplanar reconstruction, curve reestablishing, shaded surface, maximum
Intensity Projection, minimum density projection and measurement;It provides physicians with view mode to refer to for doctor, doctor is helped quickly to make just
Really judgement;
Multiplanar reconstruction:The pixel in image cross section is stacked up and is returned in three-dimensional volume arrangement, is formed as needed
Different direction, reconfigure new faultage image;To show complicated anatomy relationship in histoorgan, makes up cross-section image and see
The deficiency examined is conducive to the accurate positionin of lesion;
Curve reestablishing superposition:After three-dimensional volume arrangement, curve traveling is pressed when choosing section again, rebuilds curved surface;Side
Doctor is helped to observe and study in a short time vascular tissue;
Shaded surface:Volume data, is converted to polygon surface by the voxel information for extracting institutional framework edge in image
Then the contour surface of piece fitting carries out blanking and rendering with sensitive model according to wide recruit;For the big blood vessel of chest and abdomen, hilus pulumonis and intrapulmonary
Blood vessel, mesenteric, Renal vascular and bone and joint Three-dimensional Display;
Maximum intensity projection:Direction is realized along virtual operator, and the highest voxel value of relative density in image is thrown
The new plane projection for being mapped on screen, and being formed;The density variation of true response organization, for showing with relatively high
The institutional framework of density;
Minimum density projection:Each in image is projected into two-dimensional surface along the voxel minimum value that direction of visual lines is encountered
On, to form MinIP reconstruction image;For showing the biggish low-density tissue of density variation;
Measurement:The feature of target in image or region is measured, including:The gray feature of image, textural characteristics and
Geometrical characteristic, the volume of quantitative analysis wall and tumour, the diameter of blood vessel, to obtain distance, area, angle and statistical distribution
Measurement.
Using the beneficial effect of the technical program:
The present invention handles deep learning algorithm by Imaging enhanced and improves signal noise ratio (snr) of image, and image noise reduction is rapidly completed and subtracts
Few image waxen imagen sense promotes picture quality, to reduce CT radiological dose, reduces MRI imaging time, improve operation rate and
Working efficiency;The function that inventive algorithm template freely configures is added according to device type and scanned position independent assortment algorithm
Fast arithmetic speed and make image to reach optimum optimization effect;The processing of DICOM Imaging enhanced mainly drops medical image
It makes an uproar processing, improves signal-to-noise ratio, and ensure the clarity of image;Optimization front and back picture can be checked by image contrast function, intuitively
Experience image optimization effect;Auxiliary and support of the Imaging processing function as diagnostic imaging or process of scientific research can be clinical doctor
It is raw that condition-inference auxiliary is provided;
The present invention can be with hospital equipment work station and original PACS system seamless interfacing;Dock hospital equipment work station and
PACS system can reduce research and development cost, maintain user's use habit, have access in existing equipment, increase universal performance.
Detailed description of the invention
Fig. 1 is a kind of structural schematic diagram of medical image processing system of the invention;
Fig. 2 is the NPS curve graph that conventional iterative restructing algorithm generates;
Fig. 3 is the NPS curve graph generated in the embodiment of the present invention;
Fig. 4 is the image processing result figure generated in the embodiment of the present invention.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention is made into one with reference to the accompanying drawing
Step illustrates.
In the present embodiment, shown in Figure 1, the invention proposes a kind of medical image processing system, including outside are right
Connection module, algorithm template configuration module, Imaging enhanced processing module and image contrast and post-processing module;
The outside carries out data double-way transmission, equipment work to connection module between hospital equipment work station and PACS system
Make station or PACS system and image DICOM file is transferred to Imaging enhanced processing module, Imaging enhanced processing module will pass through calculation
File after method optimization is transmitted back to hospital equipment work station and PACS system;
The algorithm template configuration module, according to the corresponding image processing algorithm of Image Matching, described image Processing Algorithm
For deep learning algorithm;
The Imaging enhanced processing module, the first step are that DICOM image preprocessing is parsed according to user demand using HU
Method carries window position and window width information according to DICOM file, pre-processes to DICOM, prominent image emphasis details;Second
Step is that enhancing is handled, and corresponding image processing algorithm improves signal noise ratio (snr) of image, improving image quality, and ensures image definition;
The image contrast and post-processing module, the image transferred before and after the processing compare in the same window, analysis
Image processing is as a result, as diagnosis or the auxiliary and support of process of scientific research.
Prioritization scheme as above-described embodiment:
Medical imaging device work station divides analog interface and digital interface two types, is for DICOM and non-DICOM
The diagnostic workstation of CT equipment, collection registration, are reported in one at image transmitting or acquisition, diagosis, and provide data management,
The functions professional workstation modules such as backup, statistics.
PACS system is the abbreviation of Picture Archiving and Communication Systems, means image
Archiving and communication system.It is the system applied in hospital image department, and main task is exactly the various doctors daily generation
Image (including nuclear-magnetism, CT, ultrasound, various X-ray machines, the image that the equipment such as various radar stealthy materials, frequency microscope generate) is learned by various
Magnanimity saves interface (simulation, DICOM, network) in a manner of digitized, when needed the energy under certain authorization
It is enough cracking to recall to use, while increasing some auxiliary diagnosis management functions.It transmits data and group between various image documentation equipments
Storing data is knitted to play a significant role.
The outside includes optical fiber or Double-strand transmission data to connection module;After being optimized by medical image processing system
Image will directly be distributed to PACS system, the existing system of seamless interfacing hospital;Medical image processing system is based on simultaneously
DICOM3.0 standard carries out image data transmission process and is digitally signed encryption, ensures the information security of hospital and patient.
Prioritization scheme as above-described embodiment:Corresponding sample training collection training algorithm mould is used to partes corporis humani position
Type, each position are matched with algorithm model alone and constitute algorithm template;The algorithm template configuration module is according to the shadow of acquisition
As judging simultaneously to identify image modality and scanned position to match corresponding algorithm template, or by user according to the type of image and
Scanned position selects corresponding algorithm template.
Prioritization scheme as above-described embodiment:The method of the DICOM Imaging enhanced processing, including step:
Firstly, carrying out data screening to medical image data collected;
Then, data are pre-processed, preparative algorithm experiment sample, experiment sample includes training set and test set;Needle
A variety of neural network frameworks are constructed to different usage scenarios, each frame is trained using training set, obtains a variety of figures
Image intensifying model;
Finally, being tested using test set each image enhancement model, a variety of images are increased by test result
Strong model is improved, until reaching target.
Prioritization scheme as above-described embodiment:The processing method of the image contrast will be put into system by image before and after the processing
In one comparison window, user amplifies original image and processing result or rotation comparison checks that user evaluates processing result.
The method of the Imaging processing:Including multiplanar reconstruction, curve reestablishing, shaded surface, maximum intensity projection, most
Small Intensity Projection and measurement;It provides physicians with view mode to refer to for doctor, doctor is helped quickly to make accurate judgment;
Multiplanar reconstruction:The pixel in image cross section is stacked up and is returned in three-dimensional volume arrangement, is formed as needed
Different direction, reconfigure new faultage image;To show complicated anatomy relationship in histoorgan, makes up cross-section image and see
The deficiency examined is conducive to the accurate positionin of lesion;
Curve reestablishing superposition:After three-dimensional volume arrangement, curve traveling is pressed when choosing section again, rebuilds curved surface;Side
Doctor is helped to observe and study in a short time vascular tissue;
Shaded surface:Volume data, is converted to polygon surface by the voxel information for extracting institutional framework edge in image
Then the contour surface of piece fitting carries out blanking and rendering with sensitive model according to wide recruit;For the big blood vessel of chest and abdomen, hilus pulumonis and intrapulmonary
Blood vessel, mesenteric, Renal vascular and bone and joint Three-dimensional Display;
Maximum intensity projection:Direction is realized along virtual operator, and the highest voxel value of relative density in image is thrown
The new plane projection for being mapped on screen, and being formed;The density variation of true response organization, for showing with relatively high
The institutional framework of density;
Minimum density projection:Each in image is projected into two-dimensional surface along the voxel minimum value that direction of visual lines is encountered
On, to form MinIP reconstruction image;For showing the biggish low-density tissue of density variation;
Measurement:The feature of target in image or region is measured, including:The gray feature of image, textural characteristics and
Geometrical characteristic, the volume of quantitative analysis wall and tumour, the diameter of blood vessel, to obtain distance, area, angle and statistical distribution
Measurement.
As shown in Figures 2 and 3, compared to traditional iterative reconstruction algorithms, this system institute application method not only greatly reduces
Time cost can just complete image noise reduction within 1s, can also reduce the waxen imagen sense of picture.Conventional iterative restructing algorithm
With the comparison of deep learning algorithm, it is illustrated in figure 2 the NPS curve of conventional iterative restructing algorithm generation, is calculated by traditional ASIR
The comparison of method and FBP algorithm, it can be seen that with iterative reconstruction each time, image noise is constantly reduced, the peak value of NPS curve
Position is also in offset of constantly turning left;It is illustrated in figure 3 the NPS curve that this system processing generates, system is used through the invention
The comparison of method (Intelliview in figure) and FBP algorithm, it can be seen that while noise reduction, peak position is remained unchanged.
As shown in figure 4, the left side is the processing image that conventional iterative reconstructing method obtains, and the right is place proposed by the present invention
The obtained processing image of reason method;Compared to the waxen imagen sense that the image more of the invention handled can reduce picture;In Shadows Processing
Aspect also has preferable performance.
The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention.The technology of the industry
Personnel are it should be appreciated that the present invention is not limited to the above embodiments, and the above embodiments and description only describe this
The principle of invention, without departing from the spirit and scope of the present invention, various changes and improvements may be made to the invention, these changes
Change and improvement all fall within the protetion scope of the claimed invention.The claimed scope of the invention by appended claims and its
Equivalent thereof.
Claims (6)
1. a kind of medical image processing system, which is characterized in that including outside to connection module, algorithm template configuration module, image
Enhance processing module and image contrast and post-processing module;
The outside carries out data double-way transmission, equipment work station to connection module between hospital equipment work station and PACS system
Or image DICOM file is transferred to Imaging enhanced processing module by PACS system, Imaging enhanced processing module will be excellent by algorithm
File after change is transmitted back to hospital equipment work station and PACS system;
The algorithm template configuration module, according to the corresponding image processing algorithm of Image Matching, described image Processing Algorithm is deep
Spend learning algorithm;
The Imaging enhanced processing module, the first step is DICOM image preprocessing, according to user demand, using HU analytic method,
Window position and window width information are carried according to DICOM file, DICOM is pre-processed, prominent image emphasis details;Second step is to increase
Strength reason, corresponding image processing algorithm improves signal noise ratio (snr) of image, improving image quality, and ensures image definition;
The image contrast and post-processing module, the image transferred before and after the processing compare in the same window;Analyze image
Processing result, as diagnosis or the auxiliary and support of process of scientific research.
2. a kind of medical image processing system according to claim 1, which is characterized in that the outside includes to connection module
Optical fiber or Double-strand transmission data;Image after being optimized by medical image processing system will directly be distributed to PACS system,
The existing system of seamless interfacing hospital;Medical image processing system is transmitted across based on DICOM3.0 standard progress image data simultaneously
Journey is digitally signed encryption, ensures the information security of hospital and patient.
3. a kind of medical image processing system according to claim 1, which is characterized in that partes corporis humani position using opposite
The sample training collection training algorithm model answered, each position are matched with algorithm model alone and constitute algorithm template;The algorithm
Template configuration module judges and identifies image modality and scanned position according to the image of acquisition to match corresponding algorithm template,
Or corresponding algorithm template is selected according to the type and scanned position of image by user.
4. a kind of medical image processing system according to claim 1, which is characterized in that handle mould in the Imaging enhanced
The method of the processing of DICOM Imaging enhanced described in block, including step:
Firstly, carrying out data screening to medical image data collected;
Then, data are pre-processed, preparative algorithm experiment sample, experiment sample includes training set and test set;For not
A variety of neural network frameworks are constructed with usage scenario, each frame is trained using training set, a variety of images is obtained and increases
Strong model;
Finally, being tested using test set each image enhancement model, by test result to a variety of Image Enhancement Baseds
Type is improved, until reaching target.
5. a kind of medical image processing system according to claim 1, which is characterized in that in the image contrast and rear place
Manage the processing method of image contrast described in module, image will be put into unified comparison window before and after the processing, user to original image and
Processing result amplification or rotation comparison check that user evaluates processing result.
6. a kind of medical image processing system according to claim 5, which is characterized in that in the image contrast and rear place
The method for managing Imaging processing described in module:Including multiplanar reconstruction, curve reestablishing, shaded surface, maximum intensity projection, most
Small Intensity Projection and measurement;It provides physicians with view mode to refer to for doctor, doctor is helped quickly to make accurate judgment;
Multiplanar reconstruction:The pixel in image cross section is stacked up and is returned in three-dimensional volume arrangement, is formed as needed not
Same orientation reconfigures new faultage image;To show complicated anatomy relationship in histoorgan, cross-section image viewing is made up
It is insufficient;
Curve reestablishing superposition:After three-dimensional volume arrangement, curve traveling is pressed when choosing section again, rebuilds curved surface;Help is cured
It is raw to observe and study in a short time vascular tissue;
Shaded surface:Volume data, is converted to polygon surface piece and intended by the voxel information for extracting institutional framework edge in image
Then the contour surface of conjunction carries out blanking and rendering with sensitive model according to wide recruit;For the big blood vessel of chest and abdomen, hilus pulumonis and intrapulmonary blood
Pipe, mesenteric, Renal vascular and bone and joint Three-dimensional Display;
Maximum intensity projection:Direction is realized along virtual operator, and the highest voxel value of relative density in image is projected
On screen, and the new plane projection formed;The density variation of true response organization has relatively high density for showing
Institutional framework;
Minimum density projection:Each in image is projected on two-dimensional surface along the voxel minimum value that direction of visual lines is encountered,
To form MinIP reconstruction image;For showing the biggish low-density tissue of density variation;
Measurement:The feature of target in image or region is measured, including:Gray feature, textural characteristics and the geometry of image
Feature, the volume of quantitative analysis wall and tumour, the diameter of blood vessel are surveyed to obtain distance, area, angle and statistical distribution
Amount.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810746457.5A CN108922601A (en) | 2018-07-09 | 2018-07-09 | A kind of medical image processing system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810746457.5A CN108922601A (en) | 2018-07-09 | 2018-07-09 | A kind of medical image processing system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108922601A true CN108922601A (en) | 2018-11-30 |
Family
ID=64410666
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810746457.5A Pending CN108922601A (en) | 2018-07-09 | 2018-07-09 | A kind of medical image processing system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108922601A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109300534A (en) * | 2018-11-30 | 2019-02-01 | 上海藤核智能科技有限公司 | A kind of construction method of medical image knowledge base and application |
CN110120027A (en) * | 2019-04-16 | 2019-08-13 | 华南理工大学 | The CT sectioning image Enhancement Method and device of Machine oriented learning system data |
CN111161850A (en) * | 2019-12-22 | 2020-05-15 | 武汉儿童医院 | DICOM image uploading matching system and method based on non-real-time additional recording and uploading mode |
CN111243082A (en) * | 2020-01-08 | 2020-06-05 | 浙江大学 | Method, system, device and storage medium for obtaining digital image reconstruction image |
WO2020173054A1 (en) * | 2019-02-28 | 2020-09-03 | 未艾医疗技术(深圳)有限公司 | Vrds 4d medical image processing method and product |
CN112927230A (en) * | 2021-04-26 | 2021-06-08 | 常州市第一人民医院 | Image processing and analyzing system |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101673340A (en) * | 2009-08-13 | 2010-03-17 | 重庆大学 | Method for identifying human ear by colligating multi-direction and multi-dimension and BP neural network |
CN102497805A (en) * | 2009-09-17 | 2012-06-13 | 富士胶片株式会社 | Medical image display device, method, and program |
US20130051516A1 (en) * | 2011-08-31 | 2013-02-28 | Carestream Health, Inc. | Noise suppression for low x-ray dose cone-beam image reconstruction |
US20150201895A1 (en) * | 2012-08-31 | 2015-07-23 | The University Of Chicago | Supervised machine learning technique for reduction of radiation dose in computed tomography imaging |
CN107239666A (en) * | 2017-06-09 | 2017-10-10 | 孟群 | A kind of method and system that medical imaging data are carried out with desensitization process |
CN107767935A (en) * | 2017-09-15 | 2018-03-06 | 深圳市前海安测信息技术有限公司 | Medical image specification processing system and method based on artificial intelligence |
-
2018
- 2018-07-09 CN CN201810746457.5A patent/CN108922601A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101673340A (en) * | 2009-08-13 | 2010-03-17 | 重庆大学 | Method for identifying human ear by colligating multi-direction and multi-dimension and BP neural network |
CN102497805A (en) * | 2009-09-17 | 2012-06-13 | 富士胶片株式会社 | Medical image display device, method, and program |
US20130051516A1 (en) * | 2011-08-31 | 2013-02-28 | Carestream Health, Inc. | Noise suppression for low x-ray dose cone-beam image reconstruction |
US20150201895A1 (en) * | 2012-08-31 | 2015-07-23 | The University Of Chicago | Supervised machine learning technique for reduction of radiation dose in computed tomography imaging |
CN107239666A (en) * | 2017-06-09 | 2017-10-10 | 孟群 | A kind of method and system that medical imaging data are carried out with desensitization process |
CN107767935A (en) * | 2017-09-15 | 2018-03-06 | 深圳市前海安测信息技术有限公司 | Medical image specification processing system and method based on artificial intelligence |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109300534A (en) * | 2018-11-30 | 2019-02-01 | 上海藤核智能科技有限公司 | A kind of construction method of medical image knowledge base and application |
WO2020173054A1 (en) * | 2019-02-28 | 2020-09-03 | 未艾医疗技术(深圳)有限公司 | Vrds 4d medical image processing method and product |
AU2019431568B2 (en) * | 2019-02-28 | 2022-09-22 | Sheng Cao | Method and product for processing of vrds 4d medical images |
CN110120027A (en) * | 2019-04-16 | 2019-08-13 | 华南理工大学 | The CT sectioning image Enhancement Method and device of Machine oriented learning system data |
CN110120027B (en) * | 2019-04-16 | 2021-05-14 | 华南理工大学 | CT slice image enhancement method and device for machine learning system data |
CN111161850A (en) * | 2019-12-22 | 2020-05-15 | 武汉儿童医院 | DICOM image uploading matching system and method based on non-real-time additional recording and uploading mode |
CN111161850B (en) * | 2019-12-22 | 2022-06-14 | 武汉儿童医院 | Image uploading matching system and method based on non-real-time additional recording and later-transmitting mode |
CN111243082A (en) * | 2020-01-08 | 2020-06-05 | 浙江大学 | Method, system, device and storage medium for obtaining digital image reconstruction image |
CN112927230A (en) * | 2021-04-26 | 2021-06-08 | 常州市第一人民医院 | Image processing and analyzing system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108922601A (en) | A kind of medical image processing system | |
US10413253B2 (en) | Method and apparatus for processing medical image | |
US11020077B2 (en) | Simultaneous CT-MRI image reconstruction | |
US7756314B2 (en) | Methods and systems for computer aided targeting | |
CN109961834B (en) | Image diagnosis report generation method and device | |
CN108324310B (en) | Medical image providing apparatus and medical image processing method thereof | |
CN102855618B (en) | Produce for image and the method for graphical analysis | |
CN103607954B (en) | The injectivity optimizing scheme of AC and location on hybrid scanning device | |
US20030187358A1 (en) | Method, system and computer product for cardiac interventional procedure planning | |
EP2245592B1 (en) | Image registration alignment metric | |
CN100361626C (en) | Distributed cardiac image multi-dimensional re-building and interactive visual method | |
CN100559397C (en) | The affine registration method and system of two dimensional image and operation forward three-dimensional viewing in the operation | |
CN110264559B (en) | Bone tomographic image reconstruction method and system | |
JP2010086537A (en) | Active electronic medical record based support system using learning machine | |
US20040030585A1 (en) | Method and system for remote image reconstitution and processing and imaging data collectors communicating with the system | |
CN109961419B (en) | Correction information acquisition method for attenuation correction of PET activity distribution image | |
US20230196573A1 (en) | Image scanning methods and systems for medical devices | |
CN112005314A (en) | System and method for training a deep learning model of an imaging system | |
JP2003107161A (en) | Method for correcting difference between brain function images under abnormal conditions | |
CN112004471A (en) | System and method for imaging system shortcut mode | |
WO2021108398A1 (en) | Automated protocoling in medical imaging systems | |
Mamatov et al. | Analysis of imaging equipments of human internal organs | |
CN111919264A (en) | System and method for synchronizing an imaging system and an edge calculation system | |
Shekhar et al. | Medical image processing | |
CN101868809B (en) | Image processing with computer aided detection and/or diagnosis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181130 |