JP2005278991A - Remote image diagnostic reading service system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a remote image diagnostic reading system for efficiently transmitting, receiving or storing a remote image by selecting a characteristic part and its neighborhood in a system using imaged data for remote diagnostics. <P>SOLUTION: In the system used for remote diagnostics is provided with: an extraction means for extracting a noticing part where a numerical value expressing the feature of an image is over a threshold given in advance and the neighborhood of the noticing part about a newly acquired medical image; a transmission means for transmitting image data extracted by the extraction means or digital data obtained by extracting the feature; a reception means for receiving the image data or the digital data; and an image processing means for burying the image data and restored image data in the prepared medical image of a model or the past medical image of the same patient about a part containing the noticing part. The image data outputted from the image processing means is displayed, printed or transmitted. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a remote image interpretation service system that uses imaged data for remote diagnosis and efficiently transmits, receives, or saves by selecting a characteristic part and its vicinity.

  Transmits image information of organs and lesions obtained by high-definition imaging using XCT (X-ray computed tomography) equipment and MRI (magnetic resonance imaging) equipment installed in medical institutions to remote locations via a high-speed communication network In addition, it is desired to develop a remote image diagnosis system that performs 3D computer graphics (3D CG) display, generation of a real model, and pre-operative simulation by 3D reconstruction processing.

  In addition, research on imaging techniques for medical images (XCT, MRI, etc.) has been advanced, and research on medical image processing for the purpose of feature extraction and quantitative analysis of the extracted feature amounts has been advanced. In addition, the development of IT (information technology) in recent years and the development of social infrastructure for high-speed networks between domestic and international countries are also progressing. Under such circumstances, the development and practical application of the remote image diagnosis system described above have become possible.

  The following documents are disclosed as a system for diagnosis using images from a remote place. First, Patent Document 1 discloses a pathological remote image diagnostic system. This is a conversation between the hospital and the pathologist, and the tissue image necessary for pathological examination / diagnosis is transmitted and displayed on the monitor as a moving image, a medium-quality still image, and a high-quality still image. The result judged by the pathologist while looking at the image is quickly transmitted to the hospital. In this system, a medium image quality analog output obtained by converting a tissue image photographed by a camera using a down converter is sent to a system controller, and a high image quality digital output is sent to each host machine. The system control device processes moving images and medium-quality still images from the medium-quality analog output, and also transfers high-quality still images sent from the host machine to the receiving side, and simultaneously performs dialogue processing of audio and outputs them. On the receiving side, from the image data received by the system control device, a moving image and a middle still image are displayed on the TV monitor, and a high-quality still image is displayed on the monitor via the host machine, and at the same time, an audio interactive process is performed and output.

  Also, in Patent Document 2, the display conditions necessary for stereoscopic display in the display condition setting step of the terminal including the viewpoint and the light source are displayed on the display of the three-dimensional image displayed by the doctors at the remote places on the terminals of both doctors. After the setting, stereoscopic display is performed according to the set display condition in the three-dimensional display step of the terminal, and the set display condition is transmitted to the partner terminal using the stereoscopic display command in the display condition transmission step. A telemedicine support method for performing stereoscopic display according to the set display conditions in the three-dimensional display step is disclosed.

Further, Patent Document 3 discloses a system that performs feature image recognition and feature image identification on a two-dimensional spatially defined intensity image. In this system, the intensity image can be divided into a plurality of image blocks, each block comprising a plurality of image pixels comprising a memory storage unit, an image processor, and a discrete cosine transformed divided intensity image filter output. Signal feature identifier. An image processor is coupled to the memory storage unit and adapted to provide a discrete cosine transformed divided intensity image filter output signal in response to the spatially defined intensity image. It includes a transform (DCT) divided intensity image filter.
JP-A-9-117417 JP 2001-1118015 A JP 2001-250117 A

  When performing image interpretation of a disease or pathological diagnosis of a tissue from an image, it is necessary to promptly present a captured medical image, and it is desirable that transmitted image data be as small as possible while maintaining information necessary for diagnosis. . In addition, conferences (discussion about image interpretation, creation of comments and interpretation reports, etc.) at multiple medical institutions (hospitals, clinics, etc.) that exist in remote locations, using original images, feature extraction images, video, and audio When this is done, a large amount of image data is used, and therefore it is necessary to devise such that individual image data becomes smaller.

  The present invention is a system that uses imaged data for remote diagnosis, and realizes a remote image interpretation service system that efficiently transmits, receives, or stores by selecting a characteristic part and its vicinity. Objective.

  By using the present invention, it becomes possible to quickly display features included in a medical image when diagnosing a patient at a remote place. In addition, since a site likely to have a disease is narrowed down, a pathological image extracted during treatment can be quickly analyzed and used for treatment. In addition, since the patient's past images can be used, changes in the patient over time can be captured. In addition, even when medical institutions are different, it is possible to search and present medical images of patients. In addition, medical conferences between remote locations (such as discussions on diagnosis such as interpretation) can be performed using video and audio.

The present invention relates to a remote image interpretation service system used for remote diagnosis, and is a remote image interpretation service system that connects a doctor who specializes in diagnosis and a doctor or engineer who is remote from the doctor. First, for a newly acquired medical image of a patient, an attention part in which a numerical value representing the feature of the image exceeds a predetermined threshold value and an extraction means for extracting the vicinity of the attention part are provided. The medical image referred to here is an X-ray image, an ultrasonic image, a CT scan image, an MRI image, an image obtained by imaging, or the like. Such images are narrowed down at remote locations as candidates for diseased areas. This narrowing down is performed, for example, by comparison with standard models, patient's own past image data, or data extracted from image data (size, arrangement, contrast, etc.). In this comparison, the characteristic of the imaged part obtained from each image is converted into a numerical value, and the numerical value is compared to determine whether or not a predetermined threshold value is exceeded. If a part exceeding this threshold is found, the vicinity including this part is extracted by the extracting means. The extraction means is, for example, a personal computer (PC), and the method of extracting as described above using a PC is already well known.
Further, a transmission means for transmitting the image data extracted by the extraction means or the digital data extracted from the characteristics thereof is provided. The transmission means here means, for example, a modem, a communication card or a transmission line connected to a PC.
In addition, receiving means for receiving the image data or the digital data is provided. The receiving means here means, for example, a modem or a communication card connected to a PC, and a receiving PC.
Further, image processing means for embedding the above image data or the image data restored from the above digital data into a medical image of a model prepared in advance or a past medical image of the same patient with respect to a portion including the above-described attention portion is provided. ing. This image processing means is, for example, a PC.
The image data output from the image processing means is displayed or printed on a display device connected to the PC, but may be transmitted to another PC or the like.

  Further, the extracting means includes a database of medical images of models prepared in advance for a predetermined part, a database of past medical images of the same patient, or a database in which features relating to the above-described medical images are extracted, and a newly acquired medical Comparing means for comparing an image with data from the database, an attention portion selected by the comparison means, and an extraction means for extracting the vicinity of the attention portion may be used.

  Further, the transmission means may compress the image data or the digital data extracted from the characteristics and transmit it, and restore it on the receiving side.

  Further, in the middle of the transmission means, a database for storing a plurality of image data extracted by the extraction means or digital data obtained by extracting characteristics of the image data can be stored, so that a large number of patient image data can be stored. .

  Further, a database that stores digital data obtained by extracting a plurality of image data or features thereof may transmit the stored data to the above or other receiving means in response to a request from the receiving means side. desirable.

  In addition, a database of medical images of models prepared in advance for a predetermined part, a database of past medical images of the same patient, or a database from which features relating to the medical images are extracted are commonly used on the transmission side and the reception side. Thus, even when there are a plurality of receiving means, a diagnosis can be made using a common medical image.

  Further, the image processing means includes the image data extracted by the extracting means or the image data restored from the digital data obtained by extracting the characteristics, and a standard model medical image prepared in advance or a past medical image of the same patient. By performing processing that can reproduce the actual tissue as much as possible, such as by adjusting it continuously by deformation operations including expansion and contraction or rotation, or by adjusting the brightness, shade, or color of the image, etc. Easy-to-view medical images can be obtained.

  Embodiments of the present invention will be described below in detail with reference to the drawings. In the following description, an example is used. However, there is no reason that the present invention should be limited to the example, and it is obvious that the same form can be widely applied in the same manner.

(1) Overall Configuration of System The overall configuration of the remote image interpretation service system is shown in FIGS. 1A and 1B, and the data flow in the system is shown in FIG. Medical images such as XCT and MRI that have been imaged at medical institutions such as hospitals and clinics are registered and stored in a DICOM image management machine in DICOM (Digital Imaging and Communications in Medicine) format. On the other hand, the distribution machine performs a scheduled automatic search for the DICOM image management machine, obtains newly registered image data, and automatically generates a compressed image. This machine has two-dimensional and three-dimensional image processing functions and presents the characteristics of the extracted image (details of the image processing function will be described later). The distribution machine has an original image (DICOM format), a compressed image (the compression method of the image is arbitrary, but here, a result using a JPEG [Joint Photographic Experts Group] format is illustrated), and an image processed image The feature is distributed to the client (user of this system), and various images are displayed on the terminal of each user. In addition, when interpreting images between clients, a teleconference (a conference using video and audio at a remote location) is managed as necessary. The teleconference can be arbitrarily performed between clients, and the system configuration enables independent N-point to N-point communication.

  The system input screen is shown in FIG. The use of the system requires user registration in advance, and is classified into two types, “diagnostic doctor” (physician specializing in image diagnosis) and “general user” depending on available functions. The former can use all the functions of this system. In addition, it has the authority to extract image features by image processing, create rendered images, and add these results to the system. On the other hand, the latter can use only the registration ID of the user and an image captured at the facility to which the user belongs. Although messages can be exchanged with specialists and teleconference functions can be used, image processing, rendering image creation, and addition to the system are not allowed. A registrant is issued with a registration ID and password, and the system can be used by inputting them on the screen shown in the figure. The registration ID for “diagnosis doctor” is “1001 to 1999”, and the “general user” is “0001 to 0999”. These registrants can be increased.

(2) Storage and retrieval of medical image data This system is an image processing for the purpose of periodic automatic retrieval, feature extraction and quantitative analysis for medical images captured in medical institutions such as hospitals and registered in the DICOM format image DB. It has functions such as two-dimensional or three-dimensional image display, teleconference system for interpretation using video and audio, input of interpretation reports, comments, messages, etc. of specialists and creation of a database. Here, the processing contents in each function will be described.

(A) Scheduled automatic search This system searches the DICOM image management machine as an image server registered on the network at a certain time and automatically transfers the additionally registered image to the distribution machine. Then, a compressed image is automatically generated when it is registered in the DB of the distribution machine. The search time can be set to any time such as “every Monday at 9:00 am”.

(B) Image Search Method The patient ID, imaging conditions, and the like are appended to the header portion assigned to the image data. A search is performed based on “patient ID”, “acquisition date [search period]”, and “hospital name” taken from among these pieces of information. FIG. 4 shows the search screen, and FIG. 5 shows an example of the search result. If all are not input, all images related to the registration ID are searched. FIG. 5 shows the search results of three images registered in the DICOM image management machine.

(3) Image data and database In this system, image data of different formats (original image, compressed image, feature amount, rendered image), characters and figures are handled and individually converted to a DB. Here, the structure of these data and DB will be described. The DB is constructed in the DICOM image management machine and distribution machine in FIG.

(A) Original image (DICOM format image)
Medical images captured by medical institutions such as hospitals are registered and managed in the DB of the DICOM image management machine based on the DICOM format. As shown in FIG. 6, the structure of the DB is, for example,
"/ Year / month / day / file_directory / file_name"
The layer structure follows. The layer structure is based on the date and file name when the image was taken. Each image has a data structure conforming to the DICOM format, and the patient ID, patient name, hospital name, imaging conditions, etc. are recorded in the header portion.

(B) Compressed image The compressed image is registered and managed in the distribution machine, and the structure of the DB is a layered structure. here,
"JPG / year.month.day / file_directory / file_name"
(FIG. 7). The image data is standardized to 8 bits in black and white density based on the JPEG format. The image search is performed based on the structure of the DB and the header information attached to each image. For the purpose of improving the image search function, header information of DICOM format images was also given to compressed images. Therefore, the compressed image is different from the conventional JPEG format image. This is automatically added when a compressed image is generated by the distribution machine.

(C) Generation of Rendered Image and Stack Image When the original image is given as a three-dimensional image, a rendered image indicating the surface structure of the object and a stack image indicating the layer structure in the depth direction are generated by rendering processing and stack processing. These images are individually registered in the DB (FIG. 8). Each process removes the header portion from the original image, generates a rendered image or a stack image from the image data, adds the header information of the original image to each image, and registers and stores it in the DB (FIG. 9).

(D) Figures and characters (image features and interpretation messages)
The feature quantity of the image extracted from the original image or the compressed image by the two-dimensional or three-dimensional image processing is converted into a DB on the distribution machine, and the structure is a layer structure.
"PIX / year.month.day / file_directory / file_name"
(FIG. 10). The data structure is searched and displayed as an image or a figure in cooperation with the compressed image. On the other hand, a message, a comment, an interpretation report, and the like input in a displayed image (image display described later) can be written in a character format in the message column. The characters input at this time are separately registered and stored in the DB in a character format on the distribution machine. In the message history column, the character information input in cooperation with the image is automatically searched and displayed.

(4) Image display When a client searches for and displays a necessary image, the DICOM format original image or compressed image format can be displayed, and the image format can be freely selected. The original image is provided from the DICOM image management machine, and the compressed image is provided from the distribution machine. Here, an image display method based on each format will be described.

(A) Display of Original Image (DICOM Format) FIG. 11 shows an example of display for an image in DICOM format. The searched image is reduced and displayed in the lower part of the display screen together with the related image. When an image is read, it is enlarged and displayed by dragging and dropping it on one of the upper two enlarged screens, and a high-definition image is displayed. A display menu is provided at the bottom of the display screen to enable processing of “moving image”, “distance”, “conformity”, “PIX”, and “WindowLevel”. Each stack display of a 3D image is displayed as a moving image, the distance between two points designated on the image is displayed, the absolute sizes of the two left and right images to be enlarged are adapted, and the average density of the region designated on the image And the density gradation (window level) for displaying the image is set. “Image file name, patient ID and imaging date” is shown at the top left of the display screen, and “Details” at the top right is “Patient ID, patient name, date of birth, sex, age, imaging conditions” ”Etc. header information of each image is presented.

(B) Display of compressed image and recording of message FIG. 12 shows a display example of a compressed image in JPEG format. The medical image registered and stored in the DICOM image management machine automatically generates an image in a compressed format (such as JPEG format) after being searched. These compressed images are searched and all the images are reduced and displayed. Each image can be displayed by dragging and dropping it in the window at the center of the screen. In the image processing, “moving image”, “distance”, “conformity”, “PIX”, and “WindowLevel” are given in the same manner as the display screen of the DICOM image. In addition, “input”, “confirmation”, and “room registration” are provided to input a message in a character format and display the input message over time. “Image file name, patient ID and imaging date” is shown at the top left of the display screen, and “Details” at the top right is “Patient ID, patient name, date of birth, sex, age, imaging conditions” ”Etc. header information of each image is presented.

(C) Three-dimensional display Three-dimensional medical images are displayed in a surface rendering method for three-dimensionally drawing the surface of the target organ or tissue and a volume rendering method for displaying the internal structure of the target organ or tissue. Separated. FIG. 16 shows the result of three-dimensional image reconstruction (surface rendering method) for the three-dimensional XCT image given in FIG. The figure shows a display example when the aorta part, spinal cord and rib part extracted from the input image (512 × 512 × 580 pixels) by image processing are observed from the left side of the abdomen (a) and the front of the abdomen (b). . As a display example based on the volume rendering method, FIG. 17 shows a three-dimensional MRI image (256 × 256 × 64 pixels) of the head. A longitudinal fault plane (a) and a transverse fault plane (b) are shown respectively. In the three-dimensional display, three-dimensional data of a medical image to be captured is analyzed, and the characteristics are analyzed and displayed. In addition to the three-dimensional display of tissue, the result of quantitative analysis of features by image processing can be displayed in a superimposed manner.

(5) Video and audio conferences The “teleconferencing” function that enables conferences and video conferencing for image interpretation is used by medical institutions that utilize this system to exchange opinions about image interpretation and create image interpretation reports. Take advantage of. FIG. 13 shows an example of the display. Use video and audio to conduct conferences on displayed images.

  14 to 20 show an embodiment of the present system. A demonstration test was conducted between Sapporo City and Tsukuba City using B FLET'S (maximum transfer rate 100 Mbps; NTT) and JGN (maximum 100 Mbps; communication / broadcasting organization, OC3) lines (FIG. 14). Using these networks, this system was operated, the operation of the functions of this system was confirmed, and technical evaluation was performed.

(1) Verification Test System The verification test of this system was performed using the network shown in FIG. FIG. 15 shows the equipment configuration and location at that time. Medical images captured in hospitals and registered in the DB of the DICOM image management machine sv3sn02 in accordance with the DICOM format are automatically transferred to other machines (distribution machines) on the network by the scheduled automatic search system. A compressed image is automatically generated and registered and managed in the DB of the image distribution machine pctkb01. This system can be used equally by all users registered with a registration ID. This time, it was distributed to Sapporo City: 5 locations, Tsukuba City: 1 location, Tokyo: 1 location, and the operation was confirmed.

(2) Example of Image Processing FIG. 16 shows an example of a search result of an image registered in the DICOM image management machine. The patient ID, name, hospital name, imaging conditions, etc. are registered in the header portion of each image. FIG. 17 shows an example of the result of quantitative analysis of the shape feature. In the figure, the calculated distance between tissues (blood vessel diameter of the aortic site) was given. In addition to this, the features in the set region (such as “cross-sectional area”, “average density”, “density dispersion”, “perimeter”) are quantitatively obtained and displayed together with the image. An example of three-dimensional display is shown in FIG. The figure is a rendering display of the chest XCT image given in FIG. 16, and shows the results of (a) observation from the left side of the abdomen and (b) observation from the right side of the abdomen.

The present invention can be applied to the following fields.
(1) Medical device industry: Diagnosis of diseases in remote areas: Remote imaging diagnosis Possible to be installed in multiple medical institutions and medical institutions located in remote areas, at home / school / workplace, and mobile objects (cars, ships, aircraft, etc.) The medical field where it is necessary to communicate images, video, and audio between portable communication devices, etc., and present these information to each other. 1) Interpretation of images for detecting clinical diseases, 2) pathological diagnosis during surgery such as surgery, and 3) surgery simulation.
(2) Health care equipment industry: Remote outpatient care, remote checkup Health management is possible by conducting interviews and counseling with doctors at remote locations such as homes, public institutions, workplaces, and hospitals.
(3) Industrial measurement equipment industry: remote monitoring, remote measurement If the application of the present invention is not limited to remote diagnosis of patients, but also remote diagnosis of devices and facilities, etc., it can be accessed from a remote location via a network. Equipment monitoring, status (transportation, facility security, etc.), health management at home, etc. can be implemented. This is when XCT, ultrasound, and MRI are used for industrial measurements such as non-destructive inspection inside materials and measurement of fluids (flow rate, flow velocity). These image information is distributed to remote locations and the images are analyzed. The analysis of changes in the internal structure and fluid state can be performed in a similar procedure in patients, devices, and facilities.

FIG. (A) Overall configuration, (b) Diagram showing the physical configuration of the system. Diagram showing system processing functions and data flow System input screen. The figure which shows the example of the search screen of a medical image. “ID” indicates the patient ID, “Acquisition Date (from)” and “Acquisition Date (to)” indicate the search period of the imaging date (year / month / day), and Hospital indicates the name of the hospital where the imaging was performed. The figure shows a list of searched images. The figure shows an example in which three types of images are searched. The searched images (compressed format or DICOM format) are displayed in the “Conference System (JPG)” or “ Select “Image Server Access (DICOM)”. The figure shows the DB structure and registration / retrieval of the original image. In the case of a three-dimensional image, each file name is assigned and stored in the same area. A diagram showing the DB structure of compressed images and registration / retrieval. In the case of a three-dimensional image, file names are individually assigned and stored in the same area. A diagram showing the DB structure of a processed image and compression / registration / retrieval. A rendering image that automatically retrieves the original image registered in the DICOM image management machine and displays the features and tissue surface extracted from the image. Are registered as a DB with the structure shown in the figure, and are registered as a stack image in the rotation (rendering image) or depth direction for a three-dimensional image. In the figure showing the registration of the image processing result, when registering the result of image processing, the header information of the original image is given to the processed image in the same format with the same contents, but similarly, the rendered image and the stack The same header is also applied to the image (the figure shows an example of rendering processing). Figure showing character / graphic DB and registration / retrieval. Character information such as reports describing image features and the like, and graphics such as markers superimposed on the image are registered in a DB different from the above-mentioned DB. These pieces of information are searched in conjunction when searching for a compressed image, and are displayed in cooperation when displaying an image (see FIG. 12). It is a figure which shows the example of a display of an original image (DICOM format), and all the searched images are reduced and displayed on the lower stage of a display screen. Each stack display of a 3D image is displayed as a moving image, the distance between two points designated on the image is displayed, the absolute sizes of the two left and right images to be enlarged are adapted, and the average density of the region designated on the image And the density gradation (window level) for displaying the image is set. This figure shows a display example of a compressed image (JPEG format, etc.). The compressed image is also displayed in the same format as the original image. In addition to this, character and graphic input functions are provided so that teleconference can be performed. However, the input characters and graphics are searched in conjunction with the search for the compressed image, and in the case of graphics, the character information is displayed on the message history screen with the input date and time. . This figure shows a teleconference. When a conference is held at a plurality of remote locations, text information such as medical images (original images, compressed images, feature values extracted from images), reports, and other image features are displayed. Use audio and video while browsing the graphic data. The figure which shows an Example (demonstration test between Sapporo city and Tsukuba city). FIG. 14 is a diagram showing a demonstration system, which is a case where the system shown in FIGS. 1 and 2 is constructed based on FIG. 14. The original image is taken by a medical institution, and the image is registered in a DICOM image management machine. The original image is automatically searched and transferred to the distribution machine, and the functions shown in FIGS. 4 to 13 are utilized. The DICOM format original image, the compressed image, and the feature amount of the image are distributed from the WEB server to the client. Enables teleconference between clients. This figure shows the search result of the original image, and shows an example of the original image searched from the DICOM image management machine, with the patient ID, patient name, date of birth, and imaging conditions included in the header of the image added. . It is a figure which shows an example of the result of having image-analyzed a geometric feature, and shows geometric features, such as a blood vessel diameter obtained by processing the original image shown in FIG. It is a figure which shows an example of a three-dimensional display, shows the result of rendering display of the image given in FIG. 16, and shows the observed blood vessels and abdominal bone tissue from the left side of the abdomen (a) and the right side of the abdomen (b), respectively. . The figure which shows the actual display screen in the client side.

Claims (7)

A system used for remote diagnosis,
Extraction means for extracting a portion of interest in which a numerical value representing the feature of the image exceeds a predetermined threshold for a newly acquired medical image, and the vicinity of the portion of interest
Transmission means for transmitting image data extracted by the extraction means or digital data extracted from the characteristics thereof;
Receiving means for receiving the image data or the digital data;
The above-described image data or the image data restored from the above-mentioned digital data, the medical image of the model prepared in advance or the past medical image of the same patient for the part including the above-mentioned attention part,
And image processing means embedded in
A remote image interpretation service system which displays, prints, or transmits image data output from the image processing means. The extraction means is
A database of medical images of a model prepared in advance for a predetermined part, a database of past medical images of the same patient, or a database from which features relating to the above medical images are extracted;
A comparison means for comparing the newly acquired medical image with the data from the database;
The remote image interpretation service system according to claim 1, wherein the remote image interpretation service system is an extraction unit that extracts a portion of interest selected by the comparison unit and a neighborhood of the portion of interest. The transmission means is
3. The remote image interpretation service system according to claim 1, wherein the image data or digital data obtained by extracting the image data is compressed and transmitted.   4. The remote image interpretation service according to claim 1, further comprising a database for storing a plurality of image data extracted by the extraction means or digital data obtained by extracting characteristics of the image data in the middle of the transmission means. system.   5. The database storing digital data obtained by extracting a plurality of image data or characteristics thereof transmits the stored data to the above or other receiving means in response to a request from the receiving means side. The remote image interpretation service system described in 1.   A database of medical images of a model prepared in advance for a predetermined part, a database of past medical images of the same patient, or a database from which features relating to the medical images are extracted use a common database on the transmission side and the reception side. The remote image interpretation service system according to any one of claims 2 to 5, The image processing means
Image data extracted by the extraction means or image data restored from the digital data extracted from the features;
A standard model medical image prepared in advance or a past medical image of the same patient,
The remote image according to any one of claims 1 to 6, wherein a process for adjusting the images so as to be continuously connected is performed by a deformation operation including expansion and contraction or rotation, or by adjustment of light and darkness, lightness and shade, or color of the image. Interpretation service system.

Images