Medical image cloud reading method and system

Abstract

The present invention discloses a cloud-based medical image reading method and system; wherein the method, applied to a cloud server, comprises: establishing a virtual display in response to an image reading request from a client; compressing the medical image displayed on the virtual display; and transmitting the compressed image to the client via a network to enable the image to be displayed on the client. Through cloud computing, combined with the high speed and low latency of the 5G network, the reading software runs in the cloud, and the terminal computer displays and operates the cloud-based reading software through a web browser. There are no special requirements for the terminal computer configuration, which solves the problem of hospital departments being unable to use advanced image post-processing and three-dimensional reconstruction. The present invention can also be used on mobile terminals such as mobile phones and tablets.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a medical image cloud reading method and system. Background Art

[0002] The statements in this section merely mention background art related to the present invention and do not necessarily constitute prior art.

[0003] With the development of medical imaging equipment, the number of images generated by imaging examinations such as CT and MRI is increasing, and the resolution of the images is getting higher and higher. Using the traditional method of reading images, doctors need to wait for the image download to be completed before reading the images. Taking a CT examination with 1,000 images as an example, the size of each image is about 1M, and the time used to download the image is usually more than one minute. A radiologist usually has to review more than 50 CT examinations every day. Doctors waste a lot of time every day waiting for image downloads, which seriously restricts the work efficiency of radiologists.

[0004] Computers in various hospital departments are generally old and have low configurations. The computer performance is not enough to support advanced image post-processing, three-dimensional reconstruction and other functions. Especially in clinical departments, they can only use the image review function, and the clinical guidance value of medical images is severely limited. High-level hospital clinicians have an urgent and strong demand for advanced imaging functions, especially in departments such as orthopedics, thoracic surgery, oncology, and surgery that require a large number of clinical operations. Summary of the Invention

[0005] In order to address the shortcomings of the existing technology, the present invention provides a medical image cloud reading method and system; through cloud computing, combined with the high speed and low latency characteristics of 5G networks, the reading software runs in the cloud.

[0006] In a first aspect, the present invention provides a medical image cloud reading method;

[0007] The medical image cloud reading method is applied to the cloud server and includes:

[0008] In response to an image reading request from a client, a virtual display is established;

[0009] compressing medical imaging images displayed on a virtual display;

[0010] The compressed image is transmitted to the client through the network to display the image on the client.

[0011] In a second aspect, the present invention provides a medical imaging cloud reading system;

[0012] Medical imaging cloud reading system, including: client and cloud server;

[0013] The cloud server creates a virtual display in response to the client's image reading request;

[0014] compressing medical imaging images displayed on a virtual display;

[0015] The compressed image is transmitted to the client through the network to display the image on the client.

[0016] Compared with the prior art, the present invention has the following beneficial effects:

[0017] Through cloud computing, combined with the high speed and low latency of 5G networks, the film reading software runs in the cloud, and the terminal computer displays and operates the cloud-based film reading software through a web browser. There are no special requirements for the terminal computer configuration, which solves the problem of hospital departments being unable to use advanced image post-processing and 3D reconstruction. It can also be used on mobile devices such as mobile phones and tablets. At the same time, because the film reading software and image data are all on the cloud server, the film reading software directly loads the images from the cloud server without downloading them to the local department computer. This saves doctors the time of waiting for image downloads and greatly improves their work efficiency in reading films. Moreover, there is no need to install additional software or plug-ins on computers, tablets, mobile phones and other terminals, making it easy and quick to deploy and use. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0019] Figure 1 This is a flow chart of the method of embodiment 1. DETAILED DESCRIPTION

[0020] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0021] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0022] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0023] All data in this embodiment is obtained in compliance with laws and regulations and based on the consent of the user, and is used legally.

[0024] Example 1

[0025] This embodiment provides a medical image cloud reading method;

[0026] like Figure 1 As shown, the medical image cloud reading method is applied to the cloud server and includes:

[0027] S101: In response to an image reading request from a client, a virtual display is established;

[0028] S102: compressing the medical image displayed on the virtual display;

[0029] S103: Transmit the compressed image to the client via the network to display the image on the client.

[0030] Furthermore, the client includes but is not limited to a mobile phone, a laptop computer, a desktop computer, a tablet computer, a wearable electronic device, a virtual reality device, etc., and the embodiments of the present application do not impose any restrictions on this.

[0031] Furthermore, S101: in response to the client's image reading request, a virtual display is established. The specific process includes:

[0032] In response to the client's image reading request, a virtual display is established using a virtual frame buffer (Xvfb, X virtual framebuffer).

[0033] Furthermore, the step S102 of compressing the medical image displayed on the virtual display specifically includes:

[0034] The medical imaging images displayed on the virtual display are compressed by the GPU.

[0035] Furthermore, the compressing of the medical image displayed on the virtual display by relying on the GPU specifically includes:

[0036] JPEG images are compressed using NVIDIA's GPU acceleration library NVJPEG and NVIDIA's GPU hardware encoding engine NVENC.

[0037] It should be understood that the NVJPEG library is NVIDIA's high-performance GPU-accelerated library for decoding, encoding, and transcoding JPEG images. Applications rely on NVJPEG to provide higher throughput and lower encoding latency, using hybrid encoding with the CPU and GPU, the A100 GPU's JPEG decoding hardware-accelerated pipeline, single-phase and multi-phase decoding, and fast color space transcoding. NVIDIA GPUs' hardware encoding engine, NVENC, is used for video compression.

[0038] Furthermore, the step S103 of transmitting the compressed image to the client via the network to display the image on the client specifically includes:

[0039] S1031: Transmit the compressed image to the client via the 5G network;

[0040] S1032: The client decodes the received data and displays the decoding result.

[0041] Furthermore, the client decodes the received data, specifically including:

[0042] Start extracting header information from the compressed JPEG bitstream;

[0043] The Huffman decoding algorithm is used to process the serial process and convert the data flow into 8×8 image blocks;

[0044] Performs a sample conversion on the image block and produces a decoded RGB image.

[0045] Furthermore, the step S101: establishing a virtual display in response to an image reading request from a client may also include:

[0046] S100: Initialize and configure the cloud server.

[0047] Furthermore, the S100: initializing and configuring the cloud server specifically includes:

[0048] S1001: Deployed based on nvidia-docker, an application container engine, using the cudagl library in NVIDIA's cudagl image to obtain CUDA and OpenGL contexts, and using the obtained contexts to establish an OpenGL rendering window;

[0049] At the same time, it uses an open source toolkit based on Xorg and VirtualGL to provide OpenGL acceleration for image encoding and 3D computing. OpenGL commands are redirected to the GPU in the server, and only rendered frames are sent over the network. When the rendering process swaps the OpenGL drawing buffer or refreshes the OpenGL command buffer to indicate that it has finished rendering the frame, VirtualGL reads the rendered frame from the buffer and transmits it, so that the graphics card performance is fully utilized and the purpose of OpenGL acceleration is achieved.

[0050] S1002: Setting the Controller process to provide instructions for starting, closing, and updating images for the medical image reading process, and forwarding data to the medical image reading process;

[0051] The Controller process provides a lightweight HTTP server that accepts HTTP service requests for client calls.

[0052] The medical image reading process and the Controller process communicate with each other through Linux pipes. Linux pipes are a form of redirection. A process can output information to other processes through pipes. The Controller creates a pipe file and writes data to it. The medical image reading process reads the pipe file and obtains the data.

[0053] DOCKER is an open source application container engine that allows developers to package their applications and dependent packages into a portable image and then publish it to any machine with a popular Linux or Windows operating system, and can also achieve virtualization.

[0054] nvidia-docker represents a docker that can use GPU. nvidia-docker is a layer of encapsulation on docker, which is called on docker through nvidia-docker-plugin.

[0055] VirtualGL is an open source toolkit that enables any Linux or Unix remote display software to run OpenGL applications in a fully hardware-accelerated manner.

[0056] It provides browsing functions for medical images, and provides functions such as zooming, moving, rotating, adjusting window values, linear CT value measurement, regional CT value measurement, image post-processing, and three-dimensional reconstruction. The instructions of various operating functions on the computer are sent to the server, and the server returns the rendered and reconstructed image to the computer for display, assisting doctors in formulating surgical plans and improving the accuracy and safety of surgery.

[0057] Furthermore, the method further includes:

[0058] Obtain client image operation requests; image operations include: scaling, rotation, movement, and 3D reconstruction;

[0059] Send image operation requests to the cloud server;

[0060] The cloud server encodes the area of ​​the screen where pixels have changed into an image separately, and sends the encoded image to the client. The client updates and displays the area where pixels have changed.

[0061] The cloud server returns the data to the web browser image reading program for display and operation. Doctors can browse images, adjust image window width and window position, adjust browsing layout, measure the length / area / CT value of the lesion area, perform three-dimensional reconstruction of medical images, observe the spatial anatomical three-dimensional structure of lesions and tissues and organs, evaluate the positional relationship between lesions and organs, arteries, veins and other tissues, design surgical approaches, and assist in the implementation of precise surgery.

[0062] It responds to image reading requests from hospital computers, mobile phones, and tablet clients, and returns them to the clients for display and operation. The hospital does not need to replace computers, and various terminal devices do not need to install any plug-ins. With the high speed and low latency characteristics of the 5G network, image reading can be performed through web browsers such as Chrome and 360, which is convenient for doctors to operate and use. At the same time, it provides a consistent operating experience for different terminals such as computers, mobile phones, and tablets, allowing doctors to perform efficient image reading work in various working environments.

[0063] The image reading module runs on the cloud server, and medical imaging examination data is also stored on the cloud server. A CT examination of 5,000 images can be loaded within 5 seconds. Doctors do not need to wait for the images to download, and can just open the web page to read the images. At the same time, the medical image reading module provides powerful image post-processing and image 3D reconstruction functions: image fusion, image stitching, 3D rendering, 3D cropping, automatic part segmentation, etc., to assist doctors in formulating surgical plans and improve the accuracy and safety of surgery.

[0064] Example 2

[0065] This embodiment provides a medical imaging cloud reading system;

[0066] Medical imaging cloud reading system, including: client and cloud server;

[0067] The cloud server creates a virtual display in response to the client's image reading request;

[0068] compressing medical imaging images displayed on a virtual display;

[0069] The compressed image is transmitted to the client through the network to display the image on the client.

[0070] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A medical image cloud reading method, characterized by: Applicable to cloud servers, including: In response to an image reading request from a client, a virtual display is established; compressing medical imaging images displayed on a virtual display; The compressed image is transmitted to the client through the network to realize the display of the image on the client, which includes: Transmit the compressed image to the client through the communication network; The client decodes the received data and displays the decoding results; The client decodes the received data, specifically including: Start extracting header information from the compressed JPEG bitstream; The Huffman decoding algorithm is used to process the serial process and convert the data flow into n×n image blocks; n is a positive integer; Perform sampling conversion on the image block and generate a decoded RGB image; The method further comprises: Obtain client image operation requests; image operations include: scaling, rotation, movement, and 3D reconstruction; Send image operation requests to the cloud server; The cloud server encodes the area of ​​the screen where pixels have changed into an image separately, and sends the encoded image to the client. The client updates and displays the area where pixels have changed.

2. The medical image cloud reading method according to claim 1, wherein: In response to the client's image reading request, a virtual display is established. The specific process includes: In response to the client's image reading request, a virtual display is established using a virtual frame buffer.

3. The medical image cloud reading method according to claim 1, wherein: Compressing medical images displayed on a virtual display, specifically including: The medical imaging images displayed on the virtual display are compressed by the GPU.

4. The medical image cloud reading method according to claim 1, wherein: In response to the client's image reading request, a virtual display is established, and the step also includes: Initialize the configuration of the cloud server.

5. The medical image cloud reading method according to claim 4, wherein: Initialize the configuration of the cloud server, including: It is deployed based on nvidia-docker, an application container engine DOCKER. It uses the cudagl library in NVIDIA's cudagl image to obtain CUDA and OpenGL contexts, and uses the obtained contexts to establish an OpenGL rendering window. At the same time, it uses an open source toolkit based on Xorg and virtualgl to provide OpenGL acceleration for image encoding and 3D computing. OpenGL commands are redirected to the GPU in the server, and only rendered frames are sent over the network. When the rendering process swaps the OpenGL drawing buffer or refreshes the OpenGL command buffer to indicate that it has finished rendering a frame, VirtualGL reads the rendered frame from the buffer and transmits it, fully utilizing the graphics card performance and achieving the purpose of OpenGL acceleration. The Controller process is set up to provide instructions for starting, shutting down, and updating images for the medical image reading process, and forward data to the medical image reading process. The Controller process provides a lightweight HTTP server that accepts HTTP service requests for client calls. The medical image reading process and the Controller process communicate with each other through Linux pipes. The Controller creates a pipe file and writes data to it. The medical image reading process reads the pipe file and obtains the data.

6. The medical image cloud reading method according to claim 1, wherein: The cloud server returns the data to the Web browser image reading program for display and operation. The client browses the image, adjusts the image window width and window position, adjusts the browsing layout, measures the length, area or CT value of the lesion area, and performs three-dimensional reconstruction of medical images according to the input instructions.

7. The medical imaging cloud reading system is characterized by: The medical image cloud reading method according to any one of claims 1 to 6 comprises: a client and a cloud server; The cloud server creates a virtual display in response to the client's image reading request; compressing medical imaging images displayed on a virtual display; The compressed image is transmitted to the client through the network to display the image on the client.

Images