CN117877689A - Image data processing method and device - Google Patents
Image data processing method and device Download PDFInfo
- Publication number
- CN117877689A CN117877689A CN202410085179.9A CN202410085179A CN117877689A CN 117877689 A CN117877689 A CN 117877689A CN 202410085179 A CN202410085179 A CN 202410085179A CN 117877689 A CN117877689 A CN 117877689A
- Authority
- CN
- China
- Prior art keywords
- image data
- data
- model
- data processing
- interface
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 73
- 238000007639 printing Methods 0.000 claims abstract description 19
- 238000012549 training Methods 0.000 claims description 31
- 238000000034 method Methods 0.000 claims description 25
- 238000003708 edge detection Methods 0.000 claims description 18
- 238000013500 data storage Methods 0.000 claims description 13
- 230000003902 lesion Effects 0.000 claims description 11
- 238000001816 cooling Methods 0.000 claims description 10
- 238000005457 optimization Methods 0.000 claims description 10
- 230000006872 improvement Effects 0.000 claims description 9
- 238000010801 machine learning Methods 0.000 claims description 9
- 230000002093 peripheral effect Effects 0.000 claims description 9
- 241000699666 Mus <mouse, genus> Species 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000013135 deep learning Methods 0.000 claims description 5
- 238000005516 engineering process Methods 0.000 claims description 5
- 241000699670 Mus sp. Species 0.000 claims description 2
- 238000013421 nuclear magnetic resonance imaging Methods 0.000 claims 2
- 238000003745 diagnosis Methods 0.000 abstract description 20
- 210000002615 epidermis Anatomy 0.000 abstract description 3
- 230000000007 visual effect Effects 0.000 abstract description 3
- 239000000853 adhesive Substances 0.000 abstract description 2
- 230000001070 adhesive effect Effects 0.000 abstract description 2
- 238000002595 magnetic resonance imaging Methods 0.000 description 10
- 238000013136 deep learning model Methods 0.000 description 6
- 210000004556 brain Anatomy 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 4
- 238000002604 ultrasonography Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000010191 image analysis Methods 0.000 description 3
- 210000004072 lung Anatomy 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 206010051290 Central nervous system lesion Diseases 0.000 description 2
- 206010056342 Pulmonary mass Diseases 0.000 description 2
- 210000000988 bone and bone Anatomy 0.000 description 2
- 238000002591 computed tomography Methods 0.000 description 2
- 230000005489 elastic deformation Effects 0.000 description 2
- 210000004884 grey matter Anatomy 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000005311 nuclear magnetism Effects 0.000 description 2
- 238000004171 remote diagnosis Methods 0.000 description 2
- 238000011282 treatment Methods 0.000 description 2
- 210000004885 white matter Anatomy 0.000 description 2
- 206010035664 Pneumonia Diseases 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- APTZNLHMIGJTEW-UHFFFAOYSA-N pyraflufen-ethyl Chemical compound C1=C(Cl)C(OCC(=O)OCC)=CC(C=2C(=C(OC(F)F)N(C)N=2)Cl)=C1F APTZNLHMIGJTEW-UHFFFAOYSA-N 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
- 210000002027 skeletal muscle Anatomy 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000007920 subcutaneous administration Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000011269 treatment regimen Methods 0.000 description 1
Landscapes
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention belongs to the technical field of data processing, and discloses a processing method and a processing device of image data. The invention obtains target image data and a 3d structure model after processing medical image data; the real object display output of the internal structure is realized by means of plane picture printing and the like; the internal structure is directly embodied and displayed in a physical mode through a picture adhesion mode, and the surface epidermis can be directly visualized; the structure is visual, so that accurate diagnosis by doctors is facilitated, and the diagnosis efficiency, accuracy and data utilization rate are improved. The entity picture is directly adhered to the target position of the patient and is directly displayed; the adhesive can be adhered in a full range of 360 degrees on the periphery, and the display similar to a three-dimensional structure is formed, so that the internal structure is displayed directly without deviation.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a method and an apparatus for processing image data.
Background
At present, a conventional method for processing medical image data is to acquire image data of a patient target through equipment and then diagnose the patient data through a film printing or a computer display mode. The disadvantages of this technique are: 1. the target can not be directly matched with a patient, and the display is not direct enough; 2. the plane is shown with a certain deviation from the three-dimensional shape of the patient's target.
Various techniques are required by doctors during diagnosis to confirm conditions which are not visually apparent to the naked eye, such as various technical modes of X-ray, CT, ultrasound and nuclear magnetism in operation to confirm conditions of skeletal muscle vessels in subcutaneous interior of a patient. The diagnosis results of the modes are provided for doctors in a mode of image pictures, the doctors need to subjectively combine the image pictures with the body parts of the patients, the risks of position judgment deviation exist, and misjudgment of the doctors is further caused.
The prior art is PACS (Picture Archiving and Communication Systems) system, which is a medical image storage and transmission system. The PACS system includes data acquisition devices (e.g., CT, MRI, etc.), data processing and storage devices, image display devices, user interfaces, and network interfaces. However, PACS systems primarily process and store existing medical images rather than performing complex image processing or model training.
Compared with the scheme of the device, the PACS system has the following technical problems:
1) The ability of complex image processing and model training is inadequate: while PACS systems can store and retrieve medical images, they typically do not have built-in capabilities to perform complex image processing algorithms and machine learning model training. For example, they cannot perform complex image processing tasks such as noise suppression, edge detection, or feature extraction on images, nor training of deep learning models.
2) Inefficient data storage and retrieval: PACS systems typically use Hard Disk Drives (HDDs) for data storage, which are relatively slow in data read and write speeds compared to high speed Solid State Drives (SSDs). This results in data storage and retrieval speeds becoming a bottleneck in processing large amounts of medical image data.
3) User interaction capability is limited: PACS systems typically use a standard mouse and keyboard as the user interface. These devices are inferior in intuitiveness and ease of use compared to touch screens, particularly for complex image operations such as zooming, rotating, and labeling, etc.
4) Lack of high precision printing devices: PACS systems typically do not have a built-in high precision printing device, which limits the physician from printing out processed images for more detailed analysis or discussion with the patient.
Therefore, the device proposal has built-in complex image processing and model training capability, uses high-speed SSD for data storage, and comprises a touch screen user interface and high-precision printing equipment, thereby solving the technical problems existing in the PACS system.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a processing method and a processing device for image data.
The present invention is achieved by a processing apparatus for image data, comprising:
1) Data acquisition device: including CT scanners, ultrasound devices, and Magnetic Resonance Imaging (MRI) devices, are used to acquire raw medical image data.
2) A data processing unit: the system has a high-performance CPU and GPU, and is used for running complex image processing algorithms and model training.
3) A data storage unit: including high-speed SSD or hard disk for storing a large amount of medical image data and processing results.
4) Display device: a high resolution display or a specialized medical image display device for displaying the original image and the processed result.
5) Printing apparatus: the special high-precision printer is used for printing the processed plane pictures.
6) User interface: including mice, keyboards, and touch screens for operating hardware devices and software applications.
7) Network interface: including ethernet ports and/or wireless interfaces for data transmission and remote access.
8) Power supply and cooling system: ensuring stable operation of all hardware devices.
9) Peripheral interface: such as a USB interface, for connecting to other desired external devices or sensors.
Further, connection relation between hardware
1) Data acquisition device and data processing unit
The acquired medical image data is transferred to the data processing unit via a high-speed data interface, such as USB 3.0 or Thunderbolt, or a professional medical image interface.
2) Data processing unit and data storage unit
The data processing unit is connected with the data storage unit (high-speed SSD or hard disk) through the SATA interface or the NVMe interface and is used for quickly reading and writing data.
3) Data processing unit and display device
And displaying the processed image data on a high-resolution display or a professional medical image display device through an HDMI or a DisplayPort video interface.
4) Data processing unit and printing apparatus
The processed plane picture data is transmitted to a special high-precision printer through a USB interface or a network interface.
5) User interface and data processing unit
The mouse, the keyboard and the touch screen are connected with the data processing unit through USB or Bluetooth and are used for operating hardware equipment and software applications.
6) Data processing unit and network interface
Connected to a local area network or the internet through an ethernet port or a wireless network for remote access and data transmission.
7) Power supply and cooling system
The power supply and cooling system is directly connected to all hardware devices to provide a stable power supply and proper cooling.
8) Peripheral interface and data processing unit/data acquisition device
Other required external devices or sensors are connected to the data processing unit or the data acquisition device via a USB interface or other peripheral interface.
In combination with the technical scheme and the technical problems to be solved, the technical scheme to be protected has the following advantages and positive effects:
firstly, acquiring an internal structure and a surface structure of a patient target through a threshold algorithm, establishing a three-dimensional model of the interior and the surface of the patient target through 3d modeling software, projecting the internal structure of the patient target to embody and form the surface structure through a projection principle, expanding the surface structure into a planar structure through 3d modeling software, printing the planar structure into a picture, and adhering the picture of the planar structure to a surface epidermis of the patient target; the invention directly matches the image pictures obtained by various technologies such as CT, ultrasonic, nuclear magnetism and the like with specific parts of the patient, and provides direct and accurate position and state information for diagnosticians, thereby facilitating more accurate diagnosis for the doctors and facilitating subsequent more effective treatment. The invention obtains target image data and a 3d structure model after processing medical image data; the real object display output of the internal structure is realized by means of plane picture printing and the like; the internal structure is directly embodied and displayed in a physical mode through a picture adhesion mode, and the surface epidermis can be directly visualized; the structure is visual, so that accurate diagnosis by doctors is facilitated, and the diagnosis efficiency, accuracy and data utilization rate are improved.
Secondly, the entity picture is directly adhered to the target position of the patient and is directly displayed; the adhesive can be adhered in a full range of 360 degrees on the periphery, and the display similar to a three-dimensional structure is formed, so that the internal structure is displayed directly without deviation.
1) Diagnostic accuracy is improved: by training and optimizing by using the deep learning model, the method can accurately identify lesion information from a large amount of medical image data, thereby improving diagnosis accuracy.
2) The generalization capability of the model is enhanced: by optimizing the model by using an edge detection strategy, the method can further improve the generalization capability of the model, so that the model can maintain a high-precision diagnosis result under a wider condition.
3) The diagnosis efficiency is improved: by using the output result of the model to assist the real image, the method can effectively acquire lesion information, thereby greatly improving the diagnosis efficiency. This is particularly important in situations where medical resources are limited or where rapid diagnosis is required.
4) A new image data processing framework is provided: the method provides a new image data processing framework, which comprises three steps of model training, model optimization and real image assistance. This framework provides new ideas and directions for future research.
5) The value of the medical image data is enhanced: by this means, the value of the medical image data is further enhanced. The preprocessed medical image data can be used for model training, model optimization and real image assistance, so that the medical image data is utilized to the greatest extent.
In general, the image data processing method provided by the embodiment of the invention brings remarkable technical progress and provides important support for medical image analysis and diagnosis.
Thirdly, the image data processing device comprehensively applies various advanced technologies and devices, and the remarkable technical progress also has the following points:
1) The device supports a plurality of medical image acquisition methods such as CT, MRI, ultrasound and the like, and can comprehensively and thoroughly probe the physical condition of a patient.
2) The high-performance CPU and GPU ensure that complex image processing and model training are rapidly and accurately executed, and the efficiency and accuracy of data processing are greatly improved.
3) The high-speed data storage and reading, namely the high-speed SSD or hard disk not only provides a large amount of storage space, but also can rapidly read and write data, thereby further improving the operation efficiency of the system.
4) And the visual user interface is that a user can easily interact with the system through a mouse, a keyboard and a touch screen, so that the operation convenience is greatly improved.
5) The network is powerful in that the network interface supports ethernet and wireless connections so that data can be easily remotely transmitted or shared to other medical institutions and specialists for remote diagnosis or collaboration.
6) Stable operation-the design of the power supply and cooling system ensures that all hardware devices can operate stably and reliably.
7) High resolution display and printing-specialized medical image display devices and high precision printers can present extremely high quality images, which are critical to medical diagnosis.
8) Peripheral interfaces-diversified peripheral interfaces, such as USB, allow for connection of various external devices or sensors, providing flexibility in system expansion.
Through the technical progress in the aspects, the device not only improves the speed and accuracy of medical image processing, but also optimizes the user experience, and simultaneously has high flexibility and expansibility and wide application prospect.
Drawings
FIG. 1 is a schematic diagram of a processing apparatus for image data according to an embodiment of the present invention;
FIG. 2 is a schematic view of a projection principle provided by an embodiment of the present invention;
FIG. 3 is a flowchart of a method for processing image data according to an embodiment of the present invention;
in the figure: 1. a CT scanner; 2. a CPU; 3. a hard disk; 4. a display device; 5. a printer; 6. a keyboard; 7. a wireless interface.
Detailed Description
The present invention will be described in further detail with reference to the following examples in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
As shown in fig. 1, the hardware architecture of the present invention specifically includes:
1) Data acquisition device: including CT scanner 1, ultrasound device, and Magnetic Resonance Imaging (MRI) device, etc., for acquiring raw medical image data.
2) A data processing unit: the system is provided with a high-performance CPU2 and a GPU, and is used for running complex image processing algorithms and model training.
3) A data storage unit: including a high-speed SSD or hard disk 3 for storing a large amount of medical image data and processing results.
4) Display device: a high resolution display or a specialized medical image display device 4 for displaying the original image and the processed result.
5) Printing apparatus: a special high-precision printer 5 for printing the processed planar picture.
6) User interface: including a mouse, a keyboard 6 and a touch screen for operating hardware devices and software applications.
7) Network interface: including an ethernet port and/or a wireless interface 7 for data transmission and remote access.
8) Power supply and cooling system: ensuring stable operation of all hardware devices.
9) Peripheral interface: such as a USB interface, for connecting to other desired external devices or sensors.
In the technical scheme provided by the invention, the processing of the image data involves a plurality of steps and components. The following is a hypothetical detailed working principle:
1) Patient scanning, namely scanning the patient by using CT1, MRI or ultrasonic equipment to acquire original medical image data.
2) And transferring the data, namely transmitting the original image data to a data processing unit through a high-speed data interface.
3) Preprocessing-the high-performance CPU2 and GPU of the data processing unit perform preprocessing tasks such as thresholding, edge detection, etc., to prepare the data for further analysis.
4) Model training and improvement training and optimization of machine learning models using previously processed medical image datasets.
5) Analysis and diagnosis advanced algorithms and machine learning models run on the data processing unit for determining lesion information of the patient.
6) And storing the result, namely storing all processed data and analysis results in a data storage unit.
7) Data visualization-the processed images and 3D structural model are presented to the doctor by means of a professional medical image display device 4.
8) Printing, namely printing out the processed plane picture by a special high-precision printer 5.
9) Operations and feedback the doctor or operator can perform various operations through the user interface (mouse, keyboard 6, touch screen), such as adjusting parameters, starting a new scan, etc.
10 Data sharing and remote diagnosis, namely, the processed medical image data and diagnosis results can be remotely transmitted to other hospitals or specialists through a network interface.
11 Steady operation-the power supply and cooling system ensures that all hardware devices are running steadily in order to reliably perform all computing and data processing tasks.
The system provides a comprehensive, efficient and accurate tool for doctors to acquire, process and analyze medical images by integrating various hardware devices and advanced software algorithms. This helps to improve diagnostic accuracy, optimize treatment regimens, and ultimately improve the quality of patient care.
Fig. 2 is a schematic view of the projection principle provided by the embodiment of the invention.
As shown in fig. 3, the method for processing image data provided by the embodiment of the invention includes the following steps:
s101, performing model training and improvement by utilizing a target image data set obtained after preprocessing medical image data to obtain a new model; determining that the output result of the new model can be output as a first result by determining that the medical image data index is consistent with the original image index;
s102, optimizing a new model by utilizing an edge detection strategy to obtain an optimized generation model and an output result of the optimized generation model as a first result based on the inconsistency of the medical image data index and the original image index;
s103, utilizing the first result to assist the real image, and ensuring that lesion information can be effectively acquired.
The training and improvement of the deep learning or machine learning model are carried out on the preprocessed medical image data, and the working principle of generating a new model can be described as follows:
1) Pretreatment: the raw medical image data is first preprocessed, including steps of scaling, cropping, rotation, noise cancellation, contrast enhancement, etc., to prepare the data for model learning. The purpose of preprocessing is to improve image quality, reduce irrelevant information, and convert the data into a format that the model can handle.
2) Model training: the preprocessed data is used to train a deep learning or machine learning model. During the training process, the model learns how to identify features and patterns in the image through a large amount of training data. These features may be any information that facilitates model prediction, such as in lung CT scan images, model learning identifies features of pneumonia.
3) Model improvement: the performance of the model is tested by validating the data set and then improving based on the test results. Modifications include adjusting parameters of the model, changing the structure of the model, or using more complex models.
4) Generating a new model: after a series of training and improvement, a new deep learning or machine learning model is generated. This new model can identify and predict specific patterns and features in the medical image data.
5) Using the new model: when the medical image data index and the original image index are identical, the newly generated model is used for processing the new medical image data. The output of the model as a first result may be used for diagnosis, disease assessment or treatment planning.
The importance of this process is that it utilizes the powerful capabilities of deep learning or machine learning to automatically identify and analyze medical images, greatly improving the accuracy and efficiency of medical image analysis.
The processing method of the image data provided by the embodiment of the invention mainly comprises the following three steps:
1) Step S101: firstly, the method obtains a new model by performing model training and improvement on the preprocessed medical image data. In this step, it is determined whether the output result of the newly generated model can be output as the first result, depending on whether the medical image data index and the original image index agree. If the two metrics agree, the output of the newly generated model may be directly output as the first result.
2) Step S102: if the medical image data index and the original image index are not consistent in step S101, the method further uses an edge detection strategy to optimize the newly generated model, resulting in an optimally generated model. The output result of the optimized generative model is taken as a first result.
When step S101 finds that the medical image data index and the original image index are inconsistent, step S102 optimizes the newly generated model by using an edge detection strategy. Firstly, processing medical image data by utilizing an edge detection algorithm such as Sobel or Canny and the like to obtain an edge detection result. And then, adjusting model parameters by utilizing an edge detection result or adding a new data enhancement technology in model training to perform model optimization. And finally, processing the medical image data by using the optimized model to obtain an output result of the optimized generation model, wherein the output result is used as a first result for the subsequent real image assistance. Under the condition that the medical image data index is inconsistent with the original image index, the newly generated model is optimized through an edge detection strategy, and the precision and generalization capability of the model are improved.
3) Step S103: after the first result is obtained, the method uses the first result to assist the real image so as to ensure that lesion information can be effectively obtained. This step is to ensure that the output of the model provides useful information in practical applications.
According to the technical scheme, medical image data are effectively processed by combining model training, model optimization and real image assistance, so that lesion information is obtained. The method can provide important support in medical image analysis and diagnosis, and is helpful for improving the accuracy and efficiency of diagnosis.
The following are two specific embodiments listed according to the processing method of image data and implementation schemes thereof:
example 1: chest CT image data processing
In this embodiment, the preprocessed medical image data is derived from a chest CT scan. The original image index includes the lung volume, the lung density distribution, etc.
Step S101: a deep learning model, such as a Convolutional Neural Network (CNN), is trained using the preprocessed chest CT image dataset. If the output of the trained model (such as the detection result of the lung nodule) is consistent with the original image index, the output result of the model can be output as the first result.
Step S102: if the output result of the model does not coincide with the original image index in step S101, the model is optimized using the edge detection strategy. This involves adjusting parameters of the model or introducing new data enhancement techniques, such as rotation, scaling, etc., in model training to improve the generalization ability of the model.
Step S103: after the output result (i.e., the first result) of the optimized model is obtained, this result is used to assist the actual chest CT image to ensure that lesion information such as the location, size and shape of the lung nodule can be obtained efficiently.
Example 2: brain MRI image data processing
In this embodiment, the preprocessed medical image data is derived from brain MRI scans. The original image index includes brain structure such as ventricle size, white matter and gray matter distribution, etc.
Step S101: a deep learning model, such as a Full Convolutional Network (FCN), is trained using the preprocessed brain MRI image dataset. If the output of the trained model (such as the detection result of brain lesions) is consistent with the original image index, the output result of the model can be output as a first result.
Step S102: if the output result of the model does not coincide with the original image index in step S101, the model is optimized using the edge detection strategy. This involves adjusting parameters of the model or introducing new data enhancement techniques, such as noise addition, elastic deformation, etc., into the model training to improve the generalization ability of the model.
Step S103: after obtaining the output result (i.e., the first result) of the optimized model, the result is used to assist the real brain MRI image to ensure that lesion information such as the location, size and shape of the brain lesion can be obtained effectively.
Example 3: leg CT+MRI image data processing
In this embodiment, the preprocessed medical image data is from a leg ct+mri scan. The raw image index includes the bone size, volume, density distribution, white matter and gray matter of the leg, and other information such as blood vessel muscles of the leg can be included.
Step S101: training a deep learning model, such as Convolutional Neural Network (CNN), full Convolutional Network (FCN), uses the preprocessed leg ct+mri image dataset. If the output of the trained model (such as the detection result of bones) is consistent with the original image index, the output result of the model can be output as a first result.
Step S102: if the output result of the model does not coincide with the original image index in step S101, the model is optimized using the edge detection strategy. This involves adjusting parameters of the model or introducing new data enhancement techniques such as rotation, scaling, noise addition, elastic deformation, etc. in the model training to improve the generalization ability of the model.
Step S103: after the output result (i.e., the first result) of the optimized model is obtained, the result is used to assist the real leg ct+mri image to ensure that lesion information such as the position, size and shape of the leg can be obtained effectively.
The above three embodiments all show how to use the processing method of the image data in practical application. By combining training of a deep learning model, optimization of the model and assistance of a real image, the method can effectively process medical image data, thereby obtaining lesion information and improving accuracy and efficiency of diagnosis.
The foregoing is merely illustrative of specific embodiments of the present invention, and the scope of the invention is not limited thereto, but any modifications, equivalents, improvements and alternatives falling within the spirit and principles of the present invention will be apparent to those skilled in the art within the scope of the present invention.
Claims (10)
1. A processing apparatus for image data, comprising:
1) Data acquisition device: the system comprises a CT scanner, ultrasonic equipment and nuclear magnetic resonance imaging equipment, wherein the CT scanner, the ultrasonic equipment and the nuclear magnetic resonance imaging equipment are used for acquiring original medical image data;
2) A data processing unit: for running complex image processing algorithms and model training;
3) A data storage unit: the system comprises a high-speed SSD or a hard disk, and is used for storing a large amount of medical image data and processing results;
4) Display device: the method is used for displaying the original image and the processed result;
5) Printing apparatus: the printing device is used for printing the processed plane picture;
6) User interface: including mice, keyboards, and touch screens for operating hardware devices and software applications.
2. The image data processing apparatus according to claim 1, further comprising:
network interface: including ethernet ports and/or wireless interfaces for data transmission and remote access;
power supply and cooling system: ensuring the stable operation of all hardware devices;
peripheral interface: such as a USB interface, for connecting to other desired external devices or sensors.
3. The image data processing apparatus according to claim 1, wherein the data acquisition device and the data processing unit transmit the acquired medical image data to the data processing unit through a high-speed data interface or a specialized medical image interface.
4. The image data processing device according to claim 1, wherein the data processing unit and the data storage unit are connected to the data storage unit through a SATA interface or an NVMe interface for fast reading and writing data;
the data processing unit and the display device display the processed image data on a high-resolution display or a professional medical image display device through an HDMI or a DisplayPort video interface;
the data processing unit and the printing device transmit the processed plane picture data to a special high-precision printer through a USB interface or a network interface.
5. The image data processing apparatus according to claim 1, wherein the user interface is connected with the data processing unit through a mouse, a keyboard and a touch screen through USB or bluetooth for operating hardware devices and software applications;
the data processing unit is connected with the network interface through an Ethernet port or a wireless network and a local area network or the Internet for remote access and data transmission;
the power supply and cooling system is directly connected to all hardware devices to provide a stable power supply and proper cooling.
6. The apparatus according to claim 1, wherein the peripheral interface and the data processing unit/data acquisition device connect other required external devices or sensors to the data processing unit or data acquisition device via a USB interface or other peripheral interface.
7. A method of processing image data so that medical image data can be more effectively used for training and improvement of a machine learning model, comprising the steps of:
performing deep learning or machine learning model training and improvement on the preprocessed medical image data to generate a new model; when the medical image data index and the original image index are consistent, the output result of the new generation model is used as a first result.
8. The image data processing method according to claim 7, wherein in step S102, when the medical image data index and the original image index are not identical, the newly generated model is optimized by using an edge detection strategy in the image processing, and an optimization generation model is obtained by using the optimization strategy, and at the same time, an output result of the optimization generation model is taken as the first result.
9. The image data processing method according to claim 8, wherein the edge detection strategy comprises processing the medical image data by using an edge detection algorithm such as Sobel or Canny to obtain an edge detection result; and adjusting model parameters by utilizing an edge detection result or adding a new data enhancement technology into model training to perform model optimization.
10. The image data processing method according to claim 7, wherein in step S103, the first result is used to assist the real image to effectively acquire lesion information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410085179.9A CN117877689A (en) | 2024-01-20 | 2024-01-20 | Image data processing method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410085179.9A CN117877689A (en) | 2024-01-20 | 2024-01-20 | Image data processing method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117877689A true CN117877689A (en) | 2024-04-12 |
Family
ID=90584521
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410085179.9A Pending CN117877689A (en) | 2024-01-20 | 2024-01-20 | Image data processing method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117877689A (en) |
-
2024
- 2024-01-20 CN CN202410085179.9A patent/CN117877689A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Scholl et al. | Challenges of medical image processing | |
US10362941B2 (en) | Method and apparatus for performing registration of medical images | |
US20120220875A1 (en) | Mobile Architecture Using Cloud for Hashimoto's Thyroiditis Disease Classification | |
JP6170108B2 (en) | System using a structured library of gestures in a multi-touch clinical system and a storage medium for storing the program | |
US11941812B2 (en) | Diagnosis support apparatus and X-ray CT apparatus | |
EP2996058A1 (en) | Method for automatically generating representations of imaging data and interactive visual imaging reports | |
Noble | Reflections on ultrasound image analysis | |
US11344281B2 (en) | Ultrasound visual protocols | |
Tajbakhsh et al. | Guest editorial annotation-efficient deep learning: the holy grail of medical imaging | |
EP3799662A1 (en) | Anonymisation of medical patient images using an atlas | |
CN115994902A (en) | Medical image analysis method, electronic device and storage medium | |
Kagiyama et al. | Machine learning in cardiovascular imaging | |
US20240104733A1 (en) | Systems and methods to process electronic medical images for diagnostic or interventional use | |
US20120203094A1 (en) | Mobile Architecture Using Cloud for Data Mining Application | |
CN117877689A (en) | Image data processing method and device | |
CA2896496A1 (en) | Measurement and enhancement in a multi-modality medical imaging system | |
Chen et al. | The research and practice of medical image enhancement and 3D reconstruction system | |
CN116664476A (en) | Method and system for determining changes in anatomical abnormalities depicted in medical image data | |
Kumari et al. | Data efficient deep learning for medical image analysis: A survey | |
US8326007B2 (en) | Methods and apparatus for combined 4D presentation of quantitative regional measurements and morphology | |
Orosz et al. | Lung Ultrasound Imaging and Image Processing with Artificial Intelligence Methods for Bedside Diagnostic Examinations | |
CN112270974A (en) | Intelligent auxiliary medical image workstation based on artificial intelligence | |
EP3977409A1 (en) | Deformity-weighted registration of medical images | |
US11334976B1 (en) | Medical imaging with functional architecture tracking | |
WO2023004489A1 (en) | System, method and/or computer-readable medium for mapping and displaying anatomical structures in a user-friendly manner |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication |