CN112652382B - Gallbladder-pancreas disease multidisciplinary combined consultation and consultation system based on mobile terminal - Google Patents
Gallbladder-pancreas disease multidisciplinary combined consultation and consultation system based on mobile terminal Download PDFInfo
- Publication number
- CN112652382B CN112652382B CN202011639119.5A CN202011639119A CN112652382B CN 112652382 B CN112652382 B CN 112652382B CN 202011639119 A CN202011639119 A CN 202011639119A CN 112652382 B CN112652382 B CN 112652382B
- Authority
- CN
- China
- Prior art keywords
- resolution
- image data
- image
- low
- consultation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4053—Super resolution, i.e. output image resolution higher than sensor resolution
-
- 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
- G16H80/00—ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring
Abstract
The invention discloses a multi-disciplinary counseling consultation system for biliary pancreatic diseases based on a mobile terminal, which comprises: high in the clouds memory cell and server image processing unit, server image processing unit includes: the data calling module is configured to call the low-resolution image data and obtain gradient maps of the low-resolution image data in an X direction and a Y direction; an image splitting module configured to divide the blurred image data, the X-direction gradient map, and the Y-direction gradient map into a number of image blocks; the image reconstruction module is configured to determine the category to which each image block belongs, and further determine a filter corresponding to a reconstruction pixel of the image block; and performing convolution operation on each image block and the corresponding filter to obtain high-resolution image pixels of the image block, and finally obtaining restored high-resolution medical image data.
Description
Technical Field
The invention relates to the technical field of medical information processing, in particular to a multi-disciplinary counseling and consultation system for a biliary-pancreatic disease based on a mobile terminal.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The incidence of malignant tumors of bile ducts, pancreas and ampulla duodenalis is increasing year by year, the diagnosis and treatment difficulty is high, the interdisciplinary nature is strong, and for the diseases, multidisciplinary consultation and consultation are the development directions of standardized and individualized treatment, and a better diagnosis and treatment scheme can be provided for patients. However, the current biliary-pancreatic disease consultation mostly adopts a face-to-face communication mode of traditional consultation, a specific consultation system aiming at the biliary-pancreatic disease based on a mobile client is lacked, and great inconvenience is caused to patients and doctors.
The existing remote consultation system supporting the mobile terminal of the mobile phone has a single diagnosis and treatment mode, on one hand, case data is single, most of the case data are data in the forms of characters, documents, pictures and the like, and complete image data and pathological data cannot be transmitted. The complete medical image comprises a large data volume image of pancreas enhancement scanning, the CT and MR images of the enhanced gallbladder pancreas are generally 1000-3000 (images/examination), the original image occupies a disk space and has a memory size of 1-3(GB), the network flow of 1-3(GB) is also needed for complete transmission of one examination, and the traditional technology is difficult to realize stable and rapid transmission and browsing at the mobile end of the mobile phone. However, when a doctor uses a mobile phone to develop consultation services, the flow of the mobile phone is limited, and the 4G network speed is not ideal, and in order to reduce the size of data transmitted to the mobile phone end, it is necessary to compress images. However, for a consulting scene of difficult and complicated diseases, the image requires high definition during display and cannot be compressed with loss, and the traditional JPG image compression method has an unsatisfactory application effect on the scene, and is difficult to meet the requirements of high definition and undistorted image while achieving high compression ratio.
Most of remote mobile combined consultation systems disclosed in the prior art adopt a real-time communication mode, a plurality of invited persons discuss and finish consultation opinions on line at the same time, the invited persons can hardly freely and repeatedly check case data according to own time, and a plurality of persons are required to discuss on line at the same time, a special time period needs to be coordinated and arranged in multiple ways, and a large amount of time is comprehensively wasted. Moreover, the conventional consultation time is quite limited, the diagnosis and treatment opinions of the invited people for consultation are not displayed to other invited people in a message mode in a non-real-time manner and are fully interacted, communication of different invited people for consultation is not facilitated, inconvenience in multidisciplinary consultation is increased, and the advantage of full communication of multidisciplinary consultation is difficult to embody.
Disclosure of Invention
In order to solve the problems, the invention provides a gallbladder-pancreas disease multidisciplinary combined consultation and consultation system based on a mobile terminal, which compresses medical images and medical record images of gallbladder-pancreas diseases through an image compression technology, so that high-fidelity and high-definition medical images can be retrieved at the mobile terminal with small data volume, and a doctor can smoothly develop remote consultation and consultation services at a mobile terminal of a mobile phone.
According to a first aspect of the embodiments of the present invention, a multidisciplinary consultation system for biliary-pancreatic diseases based on a mobile terminal is disclosed, which includes:
the cloud storage unit is configured to acquire original high-resolution medical image data of a patient with the cholepancreatic disease, and convert the original high-resolution medical image data into low-resolution image data for storage;
a server image processing unit, the server image processing unit comprising:
the data calling module is configured to call the low-resolution image data and obtain gradient maps of the low-resolution image data in an X direction and a Y direction;
an image splitting module configured to divide the blurred image data, the X-direction gradient map, and the Y-direction gradient map into a number of image blocks;
the image reconstruction module is configured to determine the category to which each image block belongs, and further determine a filter corresponding to a reconstruction pixel of the image block; and performing convolution operation on each image block and the corresponding filter to obtain high-resolution image pixels of the image block, and finally obtaining restored high-resolution medical image data.
As a further scheme, acquiring raw high-resolution medical image data of a patient with a biliary-pancreatic disease, and converting the raw high-resolution medical image data into low-resolution image data specifically comprises:
reducing the original high-resolution medical image data to a set size by adopting bicubic interpolation; and then, the reduced image is amplified by a set multiple by adopting double cubic interpolation to obtain low-resolution image data.
As a further scheme, constructing an original high-resolution medical image data set of the biliary pancreatic disease;
downsampling original high-resolution medical image data into a low-resolution image to form a corresponding low-resolution-high-resolution image pair;
by the machine learning method, a mapping filter of each medical image data from a low-resolution image to a high-resolution image is learned.
As a further scheme, a mapping filter of each medical image data from a low resolution image to a high resolution image is learned through a machine learning method, and the specific process includes:
acquiring gradient maps of each low-resolution image in the X direction and the Y direction;
cutting the low-resolution image, the X-direction gradient map and the Y-direction gradient map set into a plurality of rectangular blocks and original high-resolution pixels corresponding to the centers of the rectangular blocks in the same coordinate region;
constructing a rectangular Block matrix GkTo find
The eigenvalues and eigenvectors of (a); wherein, WkThe matrix is a matrix with diagonal elements of one-dimensional Gaussian kernels and other positions of 0;
solving a gradient Angle, a gradient Strength and a gradient Coherence based on the eigenvalue and the eigenvector;
and dividing each rectangular block and the corresponding original high-resolution pixel into different categories, and calculating a filter corresponding to each category.
As a further scheme, determining a category to which each image block belongs, and further determining a filter corresponding to a reconstructed pixel of the image block specifically includes:
finding out the pixel type corresponding to the central pixel of each image block, and solving the gradient (Angle, Strength, Coherence) of the image block;
and finding the class of each image block according to the pixel type and the gradient (Angle, Strength, Coherence), and obtaining a filter corresponding to the reconstructed pixel of each image block.
As a further scheme, the method further comprises the following steps:
the patient mobile terminal is configured to initiate a multidisciplinary joint consultation application and upload medical image data and case data of the biliary pancreatic disease to the cloud storage unit.
As a further scheme, the method also comprises the following steps:
the doctor mobile terminal is configured to receive a patient consultation application, call medical image data and case data of a patient from the server image processing unit, and initiate consultation invitation and consultation suggestion to different expert mobile terminals according to needs;
and the expert mobile terminal is configured to receive the consultation invitation, call medical image data and case data of the patient from the server image processing unit, and leave a message according to the consultation suggestion or reply to the existing message.
As a further scheme, a doctor receiving the consultation application and all invited experts participating in the discussion can view and participate in editing the consultation suggestion, carry out the statement retention and reply discussion on other messages, and support multi-level reply, voice reply and self-deletion reply.
According to a second aspect of the embodiments of the present invention, a terminal device is disclosed, which comprises a processor and a memory, wherein the processor is used for implementing instructions; the memory is adapted to store a plurality of instructions, wherein the instructions are adapted to be loaded by the processor and to perform the following process:
acquiring original high-resolution medical image data of a patient with the biliary pancreatic disease, converting the original high-resolution medical image data into low-resolution image data and storing the low-resolution image data to a cloud end;
when the data of the cloud end needs to be checked, the low-resolution image data is called, and gradient graphs of the low-resolution image data in the X direction and the Y direction are obtained;
dividing the blurred image data, the X-direction gradient map and the Y-direction gradient map into a plurality of image blocks;
determining the category of each image block, and further determining a filter corresponding to a reconstruction pixel of the image block; and performing convolution operation on each image block and the corresponding filter to obtain high-resolution image pixels of the image block, and finally obtaining restored high-resolution medical image data.
According to a third aspect of the embodiments of the present invention, a computer-readable storage medium is disclosed, in which a plurality of instructions are stored, the instructions being adapted to be loaded by a processor and to perform the following processes:
acquiring original high-resolution medical image data of a patient with the biliary pancreatic disease, converting the original high-resolution medical image data into low-resolution image data and storing the low-resolution image data to a cloud end;
when the data of the cloud end needs to be checked, the low-resolution image data is called, and gradient graphs of the low-resolution image data in the X direction and the Y direction are obtained;
dividing the blurred image data, the X-direction gradient map and the Y-direction gradient map into a plurality of image blocks;
determining the category of each image block, and further determining a filter corresponding to a reconstruction pixel of the image block; and performing convolution operation on each image block and the corresponding filter to obtain high-resolution image pixels of the image block, and finally obtaining restored high-resolution medical image data.
Compared with the prior art, the invention has the beneficial effects that:
(1) compress the medical image and the case history image of the biliary pancreas disease, and when data is retrieved and checked, the mobile terminal of the mobile phone can read the high-fidelity and high-definition medical image with small data volume, so that the consultation specialist can comprehensively know the illness state information of the patient.
(2) The consultation application is initiated through the mobile phone terminal, the patient does not need to go to the hospital specially, the consultation application can be initiated at any time and any place in a remote mode, the expert can also check the information of the patient repeatedly through the remote terminal, the limitation of time and place is avoided, multiple people are not needed to be on line at the same time, the consultation expert can analyze the state of an illness in sufficient time, and the consultation quality and efficiency are improved.
(3) The multidisciplinary consultation is carried out in the form of forum message leaving, so that the discussion time is sufficient, the time can be prolonged to 24 hours, the multidisciplinary consultation doctors can communicate sufficiently, and diagnosis and treatment suggestions are given. The advantage of full communication and cooperation of multidisciplinary consultation is highlighted.
Additional features and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
FIG. 1 is a schematic structural diagram of a multidisciplinary counseling consultation system for biliary-pancreatic diseases based on a mobile terminal in an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a process of training a filter by machine learning according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a process for uploading data by a patient according to an embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating a process of data retrieval by a consultation specialist according to an embodiment of the present invention.
Detailed Description
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Example one
The multidisciplinary diagnosis and treatment model (Multi disciplinary team MDT) is a work group consisting of experts from two or more departments that, for a certain disease, propose an optimal treatment plan suitable for a patient through a regular consultation form, and then perform the treatment plan by related disciplines or a combination of the disciplines.
In one or more embodiments, disclosed is a mobile terminal-based gallbladder-pancreas disease multidisciplinary joint consultation system, referring to fig. 1, including:
the cloud storage unit is configured to acquire original high-resolution medical image data of a patient with the cholepancreatic disease, and convert the original high-resolution medical image data into low-resolution image data for storage;
a server image processing unit, the server image processing unit comprising:
the data calling module is configured to call the low-resolution image data and obtain gradient maps of the low-resolution image data in an X direction and a Y direction;
an image splitting module configured to divide the blurred image data, the X-direction gradient map, and the Y-direction gradient map into a number of image blocks;
the image reconstruction module is configured to determine the category to which each image block belongs, and further determine a filter corresponding to a reconstruction pixel of the image block; and performing convolution operation on each image block and the corresponding filter to obtain high-resolution image pixels of the image block, and finally obtaining restored high-resolution medical image data.
The patient mobile terminal is configured to initiate a multidisciplinary joint consultation application and upload medical image data and case data of the biliary pancreatic disease to the cloud storage unit.
The doctor mobile terminal is configured to receive a patient consultation application, call medical image data and case data of a patient from the server image processing unit, and initiate consultation invitation and consultation suggestion to different expert mobile terminals according to needs;
and the expert mobile terminal is configured to receive the consultation invitation, call medical image data and case data of the patient from the server image processing unit, and leave a message according to the consultation suggestion or reply to the existing message.
Specifically, the mobile terminal of the patient transmits medical images and/or medical record image data of the biliary pancreatic disease to the cloud storage unit; and the doctor mobile terminal and the expert mobile terminal call data stored in the cloud storage unit through the server image processing unit.
In the embodiment, a machine learning method is adopted, a high-resolution image is downsampled into a low-resolution image, a low-resolution-high-resolution image combination is generated in pairs, a mapping filter from the low resolution to the high-resolution image is obtained through learning a large amount of sample data, then the filter obtained through machine learning can be used for processing images except training data, and the high-resolution image is obtained from the low resolution. The method has the advantages that the low-resolution images can be quickly restored to the high-resolution images on the mobile terminal device, and image details are not lost.
It should be noted that the high-resolution image described in this embodiment refers to original medical image data, and the image resolution is high; and the low resolution image refers to image data after image compression.
The specific image compression process and decompression process are as follows:
1. the data preparation and data preprocessing process is as follows:
(1) an original augmented biliary pancreas CT (X-ray computed tomography), MR (magnetic resonance) original image dataset HR to be trained is prepared.
(2) And (4) scaling each original image to a quarter size by adopting double cubic interpolation to obtain a clear small image data set.
(3) And amplifying each clear small image data by twice through double cubic interpolation to obtain a low-resolution image data set LR with the same size as the corresponding original image.
2. The training steps refer to fig. 2, which specifically includes:
(1) and solving gradients in the X direction and the Y direction of the blurred image to obtain a gradient map set gradX and gradY.
And cutting the blurred image set LR, gradX and gradY gradient image sets into a plurality of rectangular blocks sqrt (n) multiplied by sqrt (n) and original high-resolution pixels corresponding to the centers of the rectangles in the same coordinate area.
(2) Fill the blocks of gradX and gradY rectangles into the following matrices:
where √ n × √ n is calculated with each pixel, for the kth pixel, we consider all pixels located at k 1. The basic method starts with computing a 2 × n matrix consisting of a horizontal gradient gxkAnd a vertical gradient gykConstituting k pixels.
Find out
Eigenvalues and eigenvectors of, WkCan be simply understood as: diagonal elements are one-dimensional Gaussian kernels, and other positions are matrixes of 0.
Two eigenvalues (r1, r2) and eigenvectors should be found here, and the eigenvector corresponding to the largest eigenvalue is taken to calculate the gradient Angle:
(3) sqrt (r1) is expressed as gradient Strength (Strength).
(4) The following formula is utilized: the gradient Coherence value (Coherence) was determined:
wherein the content of the first and second substances,
is that
The feature vector of (2);
the square root of (c) is similar to the "strength" of the gradient,
the square root of (b) can be seen as the "spread" of the local gradients, or rather their change in direction.
Intensity and coherence are useful for detecting various local image properties. Low intensity low coherence typically indicates a lack of image structure, typically corresponding to noise or compression. High intensity but low coherence generally indicates corners or other multidirectional structures. There is a high coherence, usually an edge or a series of fringes in the same direction, and the intensity measures the relative intensity of the fringes. The strength and coherence enable us to detect semantically different local image properties, so by using them as part of the hash, the filter learning process can be adapted to these conditions. Thus combining angle, intensity and coherence into a hash function can produce a series of learning-type filters that can handle a variety of situations.
(5) Setting the Angle range of gradient as [0,180], dividing the range into 24 sections; the gradient Strength and gradient Coherence ranges are [0,1.0], dividing the ranges into 3 parts each. Thus, each low-resolution tile and the original high-resolution pixels corresponding to the center of the tile can be classified into 24 × 3 — 216 categories according to their respective gradients (Angle, Strength, Coherence).
(6) A2-fold interpolation is understood to mean that each pixel of the original image corresponds to four types of interpolated pixels
In (1). According to the center pixel of the divided matrix window, the rectangular window is divided into four types according to different pixel types, and finally, each divided rectangular window is classified into: 4 × 24 × 3 ═ 864 classes.
(7) In each class, low-resolution rectangular block data is stored in Q, original high-resolution pixels are stored in V, and least squares are solved according to the following formula to obtain a filter corresponding to each class.
Where h is the calculated filter for the 2X2 matrix,
a is the low resolution tile data,
b is the original high resolution pixel and B is the low resolution rectangular block center coordinate.
3. The data decompression process is as follows:
(1) when downloading the zoomed image from the server image processing unit, firstly acquiring a low-resolution image;
(2) and acquiring gradient maps gradX and gradY in the X direction and the { Y direction of the low-resolution image.
(3) The low resolution image, gradX, gradY image are divided into image blocks of the same size as during training.
(4) The same method as training is used to find the pixel type corresponding to the central pixel of each image block, and the gradient (Angle, Strength, Coherence) of the image block is found.
(5) According to the pixel type and the gradient (Angle, Strength, Coherence), in 864 classes, the class to which the block belongs is found, and the filter corresponding to the reconstructed pixel of the block is obtained.
(6) And performing convolution operation on each image block and the corresponding filter to obtain the HR image pixel of the pixel at the center of the rectangle, and traversing the whole image to generate a corresponding high-resolution image.
In this embodiment, referring to fig. 3, the process of uploading data by the patient includes:
(1) the patient initiates a consultation application through a two-dimensional code generated by an MDT discussion module under AnyMed APP application of a WeChat scanning doctor.
(2) The patient inputs information such as disease description and consultation purpose;
(3) the patient uploads medical history data, examination data pictures and medical image data to the cloud;
(4) sequentially carrying out preprocessing processes such as patient information desensitization, format conversion (from DICOM to bitmap) and image size conversion on the medical image data, converting the medical image data into low-resolution image data and storing the low-resolution image data to a cloud; and completing the consultation application.
Referring to fig. 4, the process of the consultation specialist calling the data is as follows:
(1) the doctor receiving the consultation receives the consultation application and determines whether to receive the consultation application according to the patient data. Acceptance may host the initiation of a case discussion for the patient and send a visit invitation to each of the disciplines.
(2) After the invited experts of each subject accept the consultation application, the subject can be discussed in multiple subjects.
(3) The doctor who accepts the consultation initiates the consultation suggestion, all invited experts who participate in the discussion can check and participate in editing the consultation suggestion, leave words and reply the discussion to other messages, and support multi-level reply, voice reply and self-deletion reply.
(4) The consultation advice was finally reviewed by the doctor receiving the consultation.
(5) The doctor who receives the consultation clicks the button of 'consultation ending' to finish the multidisciplinary remote consultation of the gallbladder-pancreas diseases.
Example two
In one or more embodiments, a terminal device is disclosed that includes a processor and a computer-readable storage medium, the processor to implement instructions; the computer readable storage medium is used for storing a plurality of instructions adapted to be loaded by a processor and for performing the following process:
acquiring original high-resolution medical image data of a patient with the biliary pancreatic disease, converting the original high-resolution medical image data into low-resolution image data and storing the low-resolution image data to a cloud end;
when the data of the cloud end needs to be checked, the low-resolution image data is called, and gradient graphs of the low-resolution image data in the X direction and the Y direction are obtained;
dividing the blurred image data, the X-direction gradient map and the Y-direction gradient map into a plurality of image blocks;
determining the category of each image block, and further determining a filter corresponding to a reconstruction pixel of the image block; and performing convolution operation on each image block and the corresponding filter to obtain high-resolution image pixels of the image block, and finally obtaining restored high-resolution medical image data.
In other embodiments, a computer-readable storage medium is disclosed having stored therein a plurality of instructions adapted to be loaded by a processor and to perform the following process:
acquiring original high-resolution medical image data of a patient with the biliary pancreatic disease, converting the original high-resolution medical image data into low-resolution image data and storing the low-resolution image data to a cloud end;
when the data of the cloud end needs to be checked, the low-resolution image data is called, and gradient graphs of the low-resolution image data in the X direction and the Y direction are obtained;
dividing the blurred image data, the X-direction gradient map and the Y-direction gradient map into a plurality of image blocks;
determining the category of each image block, and further determining a filter corresponding to a reconstruction pixel of the image block; and performing convolution operation on each image block and the corresponding filter to obtain high-resolution image pixels of the image block, and finally obtaining restored high-resolution medical image data.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (4)
1. A multi-disciplinary gallbladder-pancreas disease combined consultation system based on a mobile terminal is characterized by comprising:
the cloud storage unit is configured to acquire original high-resolution medical image data of a patient with the cholepancreatic disease, and convert the original high-resolution medical image data into low-resolution image data for storage; the method specifically comprises the following steps:
reducing the original high-resolution medical image data to a set size by adopting bicubic interpolation; then, amplifying the reduced image by a set multiple by adopting double cubic interpolation to obtain low-resolution image data;
a server image processing unit, the server image processing unit comprising:
the data calling module is configured to call the low-resolution image data and obtain gradient maps of the low-resolution image data in an X direction and a Y direction;
an image splitting module configured to divide the low resolution image data, the X-direction gradient map, and the Y-direction gradient map into a number of image blocks;
the image reconstruction module is configured to determine the category to which each image block belongs, and further determine a filter corresponding to a reconstruction pixel of the image block; performing convolution operation on each image block and a corresponding filter to obtain high-resolution image pixels of the image block, and finally obtaining restored high-resolution medical image data;
wherein, the determining the category to which each image block belongs specifically comprises the following steps:
according to the image blocks in the image splitting module, a rectangular block matrix is constructed
To find
The eigenvalues and eigenvectors of (a); wherein, the first and the second end of the pipe are connected with each other,
a matrix with diagonal elements of one-dimensional Gaussian kernel and other positions of 0;
finding out the pixel type corresponding to the central pixel of each image block, and solving a gradient Angle, a gradient Strength and a gradient Coherence based on the characteristic value and the characteristic vector;
finding the class of each image block according to the pixel type and the gradient Angle, the gradient Strength and the gradient Coherence corresponding to the central pixel of each image block;
constructing an original high-resolution medical image data set of the biliary pancreas disease;
downsampling original high-resolution medical image data into a low-resolution image to form a corresponding low-resolution-high-resolution image pair;
learning a mapping filter of each medical image data from a low-resolution image to a high-resolution image by a machine learning method; the specific process comprises the following steps:
acquiring gradient maps of each low-resolution image in the X direction and the Y direction;
cutting the low-resolution image, the X-direction gradient map and the Y-direction gradient map set into a plurality of rectangular blocks in the same coordinate area, and obtaining original high-resolution pixels corresponding to the centers of the rectangular blocks;
dividing each rectangular block and the corresponding original high-resolution pixels into different categories, and calculating a filter corresponding to each category;
the patient mobile terminal is configured to initiate a multidisciplinary joint consultation application and upload medical image data and case data of the biliary pancreatic disease to the cloud storage unit;
the doctor mobile terminal is configured to receive a patient consultation application, call medical image data and case data of a patient from the server image processing unit, and initiate consultation invitation and consultation suggestion to different expert mobile terminals according to needs;
and the expert mobile terminal is configured to receive the consultation invitation, call medical image data and case data of the patient from the server image processing unit, and leave a message according to the consultation suggestion or reply to the existing message.
2. The system as claimed in claim 1, wherein the doctor receiving the consultation request and all invited experts participating in the discussion can view and participate in editing the consultation advice, make a leave word and perform a reply discussion on other messages, and support multi-level reply, voice reply and delete the reply thereof.
3. A terminal device comprising a processor and a memory, the processor being arranged to implement instructions; the memory is adapted to store a plurality of instructions, wherein the instructions are adapted to be loaded by the processor and to perform the following process:
acquiring original high-resolution medical image data of a patient with the biliary pancreatic disease, converting the original high-resolution medical image data into low-resolution image data and storing the low-resolution image data to a cloud end; the method specifically comprises the following steps:
reducing the original high-resolution medical image data to a set size by adopting bicubic interpolation; secondly, amplifying the reduced image by a set multiple by adopting bicubic interpolation to obtain low-resolution image data;
when the data of the cloud end needs to be checked, the low-resolution image data is called, and gradient graphs of the low-resolution image data in the X direction and the Y direction are obtained;
dividing the low-resolution image data, the X-direction gradient map and the Y-direction gradient map into a plurality of image blocks;
determining the category of each image block, and further determining a filter corresponding to a reconstruction pixel of the image block; performing convolution operation on each image block and a corresponding filter to obtain high-resolution image pixels of the image block, and finally obtaining restored high-resolution medical image data;
wherein, the determining the category to which each image block belongs specifically comprises the following steps:
according to the image blocks in the image splitting module, a rectangular block matrix is constructed
To find
The eigenvalues and eigenvectors of (a); wherein the content of the first and second substances,
the matrix is a matrix with diagonal elements of one-dimensional Gaussian kernels and other positions of 0;
finding out the pixel type corresponding to the central pixel of each image block, and solving a gradient Angle, a gradient Strength and a gradient Coherence based on the characteristic value and the characteristic vector;
finding the class of each image block according to the pixel type and the gradient Angle, the gradient Strength and the gradient Coherence corresponding to the central pixel of each image block;
constructing an original high-resolution medical image data set of the biliary pancreas disease;
downsampling original high-resolution medical image data into a low-resolution image to form a corresponding low-resolution-high-resolution image pair;
learning a mapping filter of each medical image data from a low-resolution image to a high-resolution image by a machine learning method; the specific process comprises the following steps:
acquiring gradient maps of each low-resolution image in the X direction and the Y direction;
cutting the low-resolution image, the X-direction gradient map and the Y-direction gradient map set into a plurality of rectangular blocks in the same coordinate area, and obtaining original high-resolution pixels corresponding to the centers of the rectangular blocks;
dividing each rectangular block and the corresponding original high-resolution pixels into different categories, and calculating a filter corresponding to each category;
the patient mobile terminal initiates a multidisciplinary joint consultation application and uploads medical image data and case data of the biliary pancreatic disease to the cloud storage unit;
the doctor mobile terminal receives the patient consultation application, calls medical image data and case data of the patient from the server image processing unit, and initiates consultation invitation and consultation suggestion to different expert mobile terminals according to needs;
the expert mobile terminal receives the consultation invitation, calls medical image data and case data of the patient from the server image processing unit, and leaves a corresponding message according to the consultation suggestion or replies the existing message.
4. A computer readable storage medium having stored therein a plurality of instructions adapted to be loaded by a processor and to perform the following process:
acquiring original high-resolution medical image data of a patient with the biliary pancreatic disease, converting the original high-resolution medical image data into low-resolution image data and storing the low-resolution image data to a cloud end; the method specifically comprises the following steps:
reducing the original high-resolution medical image data to a set size by adopting bicubic interpolation; then, amplifying the reduced image by a set multiple by adopting double cubic interpolation to obtain low-resolution image data;
when the data of the cloud end needs to be checked, the low-resolution image data is called, and gradient graphs of the low-resolution image data in the X direction and the Y direction are obtained;
dividing the low-resolution image data, the X-direction gradient map and the Y-direction gradient map into a plurality of image blocks;
determining the category of each image block, and further determining a filter corresponding to a reconstruction pixel of the image block; performing convolution operation on each image block and a corresponding filter to obtain high-resolution image pixels of the image block, and finally obtaining restored high-resolution medical image data;
wherein, the determining the category to which each image block belongs specifically comprises the following steps:
according to the image blocks in the image splitting module, a rectangular block matrix is constructed
To find
The eigenvalues and eigenvectors of (a); wherein the content of the first and second substances,
is one-dimensional as diagonal elementA Gaussian kernel, a matrix with 0 in other positions;
finding out the pixel type corresponding to the central pixel of each image block, and solving a gradient Angle, a gradient Strength and a gradient Coherence based on the characteristic value and the characteristic vector;
finding the class of each image block according to the pixel type and the gradient Angle, the gradient Strength and the gradient Coherence corresponding to the central pixel of each image block;
constructing an original high-resolution medical image data set of the biliary pancreas disease;
downsampling original high-resolution medical image data into a low-resolution image to form a corresponding low-resolution-high-resolution image pair;
learning a mapping filter of each medical image data from a low-resolution image to a high-resolution image by a machine learning method; the specific process comprises the following steps:
acquiring gradient maps of each low-resolution image in the X direction and the Y direction;
cutting the low-resolution image, the X-direction gradient map and the Y-direction gradient map set into a plurality of rectangular blocks in the same coordinate area, and obtaining original high-resolution pixels corresponding to the centers of the rectangular blocks;
dividing each rectangular block and the corresponding original high-resolution pixels into different categories, and calculating a filter corresponding to each category;
the patient mobile terminal initiates a multidisciplinary joint consultation application and uploads medical image data and case data of the biliary pancreatic disease to the cloud storage unit;
the doctor mobile terminal receives the patient consultation application, calls medical image data and case data of the patient from the server image processing unit, and initiates consultation invitation and consultation suggestion to different expert mobile terminals according to needs;
the expert mobile terminal receives the consultation invitation, calls medical image data and case data of the patient from the server image processing unit, and leaves a corresponding message according to the consultation suggestion or replies the existing message.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011639119.5A CN112652382B (en) | 2020-12-31 | 2020-12-31 | Gallbladder-pancreas disease multidisciplinary combined consultation and consultation system based on mobile terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011639119.5A CN112652382B (en) | 2020-12-31 | 2020-12-31 | Gallbladder-pancreas disease multidisciplinary combined consultation and consultation system based on mobile terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112652382A CN112652382A (en) | 2021-04-13 |
| CN112652382B true CN112652382B (en) | 2022-05-06 |
Family
ID=75366916
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011639119.5A Active CN112652382B (en) | 2020-12-31 | 2020-12-31 | Gallbladder-pancreas disease multidisciplinary combined consultation and consultation system based on mobile terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112652382B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106326675A (en) * | 2016-09-05 | 2017-01-11 | 深圳市六联科技有限公司 | Remote consultation system based on three-dimensional model |
| CN107818828A (en) * | 2017-10-30 | 2018-03-20 | 广州医加多医疗股份有限公司 | The method that Multidisciplinary Cooperation diagnosis and treatment are realized using mobile remote medical |
| CN108538402A (en) * | 2018-04-16 | 2018-09-14 | 深圳零壹云医科技有限公司 | A kind of MDT consultation of doctors method, system |
| CN108765343A (en) * | 2018-05-29 | 2018-11-06 | Oppo（重庆）智能科技有限公司 | Method, apparatus, terminal and the computer readable storage medium of image procossing |
| CN110070486A (en) * | 2018-01-24 | 2019-07-30 | 杭州海康威视数字技术股份有限公司 | A kind of image processing method, device and electronic equipment |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100453503B1 (en) * | 2002-01-08 | 2004-10-20 | 주식회사 케이티프리텔 | Remote medical treating method and system with local wireless interface |
| US8384755B2 (en) * | 2009-08-26 | 2013-02-26 | Intouch Technologies, Inc. | Portable remote presence robot |
| CN103390100A (en) * | 2013-07-10 | 2013-11-13 | 山东大学齐鲁医院 | 3G wireless communication-based colostomy complication remote diagnosis system |
| CN104574316B (en) * | 2015-01-13 | 2018-09-14 | 南京佰能唯医疗科技有限公司 | A kind of digitized X-ray machine system |
| CN110211143A (en) * | 2019-06-10 | 2019-09-06 | 张家港江苏科技大学产业技术研究院 | A kind of medical image analysis method based on computer vision |
| CN111179252B (en) * | 2019-12-30 | 2021-02-05 | 山东大学齐鲁医院 | Cloud platform-based digestive tract disease focus auxiliary identification and positive feedback system |
-
2020
- 2020-12-31 CN CN202011639119.5A patent/CN112652382B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106326675A (en) * | 2016-09-05 | 2017-01-11 | 深圳市六联科技有限公司 | Remote consultation system based on three-dimensional model |
| CN107818828A (en) * | 2017-10-30 | 2018-03-20 | 广州医加多医疗股份有限公司 | The method that Multidisciplinary Cooperation diagnosis and treatment are realized using mobile remote medical |
| CN110070486A (en) * | 2018-01-24 | 2019-07-30 | 杭州海康威视数字技术股份有限公司 | A kind of image processing method, device and electronic equipment |
| CN108538402A (en) * | 2018-04-16 | 2018-09-14 | 深圳零壹云医科技有限公司 | A kind of MDT consultation of doctors method, system |
| CN108765343A (en) * | 2018-05-29 | 2018-11-06 | Oppo（重庆）智能科技有限公司 | Method, apparatus, terminal and the computer readable storage medium of image procossing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112652382A (en) | 2021-04-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109886871B (en) | Image super-resolution method based on channel attention mechanism and multi-layer feature fusion | |
| CN106657713A (en) | Video motion amplification method | |
| CN112801904B (en) | Hybrid degraded image enhancement method based on convolutional neural network | |
| US11348291B2 (en) | System and method for reconstructing magnetic resonance images | |
| CN105513033B (en) | A kind of super resolution ratio reconstruction method that non local joint sparse indicates | |
| CN110211205B (en) | Image processing method, device, equipment and storage medium | |
| Zhu et al. | Residual dense network for medical magnetic resonance images super-resolution | |
| CN104574316A (en) | Digital X-ray imaging system | |
| CN110288524A (en) | Deep learning super-resolution method based on enhanced up-sampling and discrimination syncretizing mechanism | |
| Seeram et al. | Image postprocessing in digital radiology—a primer for technologists | |
| CN115375548A (en) | Super-resolution remote sensing image generation method, system, equipment and medium | |
| CN108460723A (en) | Bilateral full variation image super-resolution rebuilding method based on neighborhood similarity | |
| CN112652382B (en) | Gallbladder-pancreas disease multidisciplinary combined consultation and consultation system based on mobile terminal | |
| Wu et al. | Deep learning-based image super-resolution restoration for mobile infrared imaging system | |
| CN103903239B (en) | A kind of video super-resolution method for reconstructing and its system | |
| Wang et al. | Pixel attention convolutional network for image super-resolution | |
| Gulati et al. | Application of an enhanced deep super-resolution network in retinal image analysis | |
| Duanmu et al. | The image super-resolution algorithm based on the dense space attention network | |
| US20080298700A1 (en) | Method and device for selecting transform matrices for down-sampling dct image using learning with forgetting algorithm | |
| Zhao et al. | Chest X-ray images super-resolution reconstruction via recursive neural network | |
| CN111325756A (en) | Three-dimensional image artery and vein segmentation method and system based on deep learning network | |
| CN116071239A (en) | CT image super-resolution method and device based on mixed attention model | |
| CN106558021A (en) | Video enhancement method based on super-resolution technique | |
| CN107169925B (en) | The method for reconstructing of stepless zooming super-resolution image | |
| CN113592973A (en) | Magnetic resonance image reconstruction method and device based on multi-frequency complex convolution |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |