JP3145342B2 - Processing device for medical image data - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for processing medical image data necessary for medical treatment of a doctor or the like.

[0002]

2. Description of the Related Art At medical sites such as hospitals, digital image processing, especially radiographic images, has been dramatically advanced. However, the development of processing technology for analog image data obtained in consultation rooms and physiological function test rooms has been delayed compared to the digital image processing, and there is a possibility that collection and integration of medical information in a hospital information system may become a major obstacle. is there. That is, in the medical field, character information such as a visit history, a medication history, and a test history and X-ray image information are conventionally stored and used as a digital database. On the other hand, for example, moving images and still images such as the facial condition of a patient that is important for diagnosis in neurosurgery, the movement and paralysis of limbs in neurology, and the change of arm and indirect dislocation in orthopedics, In the past, there were few suitable means for taking pictures at medical sites, and even if they existed, they could only be captured by a video camera and recorded on videotape, and could be searched and used. Not. Therefore, conventionally, analog image data is not used and is discarded. At medical sites, still images are attached to medical records, filed and stored on shelves, and data such as moving images and sounds are simply recorded on magnetic tape or the like and stored on shelves.

[0003]

SUMMARY OF THE INVENTION It is an object of the present invention to efficiently obtain analog images and audio data existing at medical sites such as hospitals and to obtain image and audio data necessary for medical treatment. An object of the present invention is to provide a processing apparatus for medical image data which is recorded and stored in a recording medium, and the stored data can be searched and used effectively for medical treatment.

[0004]

According to the present invention, basically, a collection device for collecting and acquiring medical image data and a storage device for storing the extracted image data are respectively provided by a client and a server system. Configured as
Analogue Picture Archiving and Communicating
It can be called a System. According to the present invention, (1) an acquisition unit serving as a collection device for acquiring and acquiring medical image data by which a doctor can capture an image to be taken by a simple operation as much as possible;
An image data extraction unit, which is a medical expert system that extracts only image data necessary for medical diagnosis from captured image data, and (3) a database created by assigning diagnostic attributes to the extracted medical image data and vague It includes a database unit of image data as a storage device having a medical image data search agent function that also enables a search, thereby facilitating the use of medical image data. It is the most desired in medical practice to give doctors time to think and think for new diagnosis, so the image input has been matured so that peripheral technology is developed and a large amount of images can be handled. After extracting medical image data using analog image data technology, it is handled as digital image data.

According to the present invention, a television camera 10 for imaging a patient's human body 12 for a predetermined period of time to obtain moving image and still image data, and a driving means 11 for changing the optical axis of the television camera 10 three-dimensionally. Operating means 9 operated by the operator 8, control means 13 for driving the driving means 11 in response to the output of the operating means 9, and visual output so that the operator 8 can view the output of the television camera 10. A monitor 14 to be displayed; a recording device 15 for storing image data of the television camera 10; setting means for setting medical subjects and regions; and reproduction of image data stored in the recording device 15 in response to an output of the setting means. Then, for the elbows and arms of the patient's body, and for the parts of the head from the neck to the upper part of the patient's body for each medical treatment subject and region, a video of the rolling condition and a video of the initial movement from the instruction to start walking to the start are taken. Obtaining the coordinate position and area of each region of the skin exposed from the patient's clothing, detecting the coordinates of the centroid of each region, and responding to the output of the obtaining unit. Time change rate calculating means for calculating the time change rate L1 of the centroid, moving image template setting means 43 for storing a predetermined moving image template value L01 for each medical treatment subject and region, time change rate calculating means, and moving image In response to the output from the template setting means
L1 (= L1−L01) moving image matching degree calculating means, and the moving image matching degree Δ among the image data stored in the recording device 15 in response to the output of the moving image matching degree calculating means.
Extracting means for extracting image data for a predetermined time T0 from the start of extraction when image data in which L1 is less than a predetermined value ΔL01 for a moving image is obtained, and an extracted image in response to an output of the extracting means. Extraction memory 41 for storing data in association with patient identification numbers
And a display means 42 for visually displaying the extracted image stored in the extraction memory. The present invention also provides a recording device 15
Area detecting means for extracting one screen of still image data of the image data stored in the storage area, and detecting the discolored areas 31 and 34 which are the changed parts where the color has changed greatly from the normal skin color, and the discolored area detecting means Discolored area 3 detected by
Discoloration area calculating means for calculating each of the areas Sj of 1, 34;
Discolored area 3 in the entire area S0 of the skin in one screen
The ratio L2 of the total area ΣSj of 1,34 (= ΣSj / S
0), a still image template setting means for storing a predetermined reference still image template value L02, and a still image template response means responsive to the output of the ratio calculation means and the still image template setting means. Matching degree ΔL2 (= L
2-L02), and in response to the output of the still image matching degree calculating means, the still image matching degree of the extracted still image is a predetermined value ΔL0 for the still image.
Means for storing image data of a still image which is less than 2 in the extraction memory 41. Further, the invention is characterized in that the extracting means extracts the image data for a predetermined time retroactively from the start of the extraction.

According to the present invention, image data of a moving image or a still image obtained by two-dimensionally imaging a human body with an analog video camera or the like is subjected to characterization image processing and reference data obtained by performing similar processing in advance. Is compared with the template value, and the degree of matching ΔLi is calculated. When the degree of matching is high, that is, when the image data is required by a doctor for diagnosis, the image data is extracted from a memory for extraction, for example, a magnetic tape or a magnetic tape. The data is stored in a recording medium such as a disk, an optical disk, and a semiconductor memory. As a result, image data that is not useful for a doctor's diagnosis can be removed, and only useful image data can be extracted. Therefore, the capacity of the extraction memory can be reduced, and the diagnosis can be performed quickly.

The characterization image processing refers to, for example, a process of calculating and calculating a value Li relating to image data in an embodiment of the present invention described later.

[0008]

According to the present invention, in order to obtain moving image data, when it is determined that the matching degree ΔLi is high, that is, in the embodiment described later, the calculated matching degree ΔLi is less than a predetermined value. Is a predetermined time starting from the image data, for example, only one minute, or
The image data is stored in the extraction memory continuously or intermittently for a predetermined time retroactively.

[0010]

According to the present invention, in order to extract image data of a still image, image data determined to have a high degree of matching is stored in the extraction memory as it is.

The present invention also provides a first transmission means for transmitting, via a communication line, image data stored in the extraction memory 41 and an identification signal corresponding to the image data,
A database memory for storing the image data from the first transmitting means via the communication line in accordance with the identification signal; and reading out the image data corresponding to the identification signal input to the database memory via the communication line. A second transmitting means for transmitting, and a means for receiving and displaying image data from the second transmitting means via the communication line are provided.

According to the present invention, the image data of a moving image or a still image stored in the extraction memory is transmitted via a communication line, for example, a LAN (Local Area Network) or
The data is transmitted by the first transmission means via a telephone line such as SDN or PHS, and stored in a database memory corresponding to an identification number used for a search or the like. For example, when a doctor for each medical department or a plurality of doctors in a hospital needs image data stored in the database memory, the image data in the database memory is transmitted by the identification number in the second transmission. Means for transmitting to a receiving and displaying means via a communication line, such as the LAN or a telephone line, for visual display;
Alternatively, it can be printed out on a recording paper, and further displayed and output by voice or the like.

Thus, the diagnosis support system according to the present invention can be used for ambulances, helicopters, clinics, clinics, home medical treatments, etc. by using a communication line such as a mobile phone. In addition, as part of a medical diagnosis,
Software for medical image processing can be developed and spread, and the present invention can be easily realized by a combination of matured and reliable existing device devices. As a communication line, for example, a terminal device such as an ISDN and a PHS can be used so that image data can be closely linked between hospitals and clinics of various scales. In addition, medical treatment using image data for a plurality of subjects is further possible.

The image data may be not only image data of the external shape of the human body, but also fundus photographs, gastrointestinal endoscopy images, etc., as well as electroencephalograms, electrocardiograms, cystoscope photographs, surgical photographs, skin symptom photographs, and others. Image data related to medical treatments.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a simplified block diagram showing the overall configuration of an embodiment of the present invention. LAN
Is connected to an acquisition unit 2 constituting image data output means for outputting image data of a moving image or a still image obtained by imaging a human body. Further, an extraction unit 3 is connected to the communication line 1. The extraction unit 3 includes image processing means for processing the image data from the acquisition unit 2 for characterization image processing, and performs the characterization image processing for the moving image or still image data stored in the template memory. A template memory for storing reference data prepared in advance as a template is provided. The image processing data of the image data obtained from the acquisition unit 2 is compared with the template to calculate the matching degree ΔLi. When the matching degree is high, that is, when the matching degree ΔLi is less than a predetermined value, the image data from the acquisition unit 2 is stored in the extraction memory provided in the extraction unit 3.

The extracted image data stored in the extraction memory of the extraction unit 3 is transmitted from the first communication means of the extraction unit 3 to the database unit 4 via the communication line 1 and stored in the database memory. You. Each image data is provided with an identification signal, thereby enabling a search. Thus, the image data stored in the database memory provided in the database unit 4 can be searched based on the identification signal via the communication line 1 and transmitted by the second transmission unit.
Further, a terminal device 6 is connected to the second transmitting means of the database unit 4 via a communication line 5, where the terminal device 6 is stored in a database memory, for example, in a room or a clinic for each medical subject in a hospital. Image data can be received and output such as visual observation, printing, or sound can be performed. Further, processing means such as an analog television camera and a personal computer are provided on the bus of the reception display means 6, and the image data obtained by imaging here is transmitted to the extraction unit 3 via the communication lines 5 and 1. can do.

FIG. 2 is a simplified block diagram showing the configuration and functions of the acquisition unit 2, and FIG. 3 is a block diagram showing a specific configuration of the acquisition unit 2. Room 7
Inside, an operator 8, for example, a doctor, operates the operating means 9 to move the optical axis of the analog color television camera 10 to 3.
The dimension is changed by the driving means 11. TV camera 1
Numeral 0 is placed behind, beside, or in some cases in front of the operator 8 and images the patient's human body 12 to be imaged. The output of the operation means 9 is given to the control means 13, whereby the control means 13 drives the drive means 11 to change the optical axis of the camera 10, performs imaging from a desired angle of the patient 12, Image data of a still image can be obtained as an analog signal. The output of the camera 10 is displayed on a monitor 14 serving as a visual display means, and the display contents can be viewed by the operator 8. Further, image data obtained by imaging with the television camera 10 is recorded on a magnetic tape 17 in a magnetic tape recording device 15 and stored. The control means 13 and the recording device 15 are a processing circuit 1 realized by a personal computer or the like.
6, the processing circuit 16 is connected to the communication line 1 and can transmit and receive. Operation means 9, control means 1
3. The driving means 11, the monitor 14 and the like constitute a man-machine interface 18, which makes it possible to obtain accurate image data without requiring a high level of skill.
Thereby, the camera 10 is driven as described above, and the image data is stored on the magnetic tape 17.

Still images may have high definition (HDPT), and moving images may have NTSC image quality. Further, the image may be compressed or expanded. Further, the acquisition unit 2 including the television camera 10 may be of a portable type, for example, a structure capable of being dispatched for disaster response. The acquisition unit 2 obtains an analog image of a medical site, and uses a digital imaging device to collect the analog image.
Utilizing digital bus such as EE1394, efficiently
High quality, international standard compliant image collection. In the image data, an identification signal such as a medical care subject of a doctor, a part of a human body, an identification number for identifying a patient, and a patient name is input by the operation means 9, and can be searched according to the image data of the magnetic tape 17. Store to become.
Further, the date and time of imaging and the name of the associated doctor may be stored as identification signals.

FIG. 4 is a block diagram showing the structure of the extraction unit 3. As shown in FIG. The extraction unit 3 performs characterization image processing in order to extract a primary image required for medical treatment of a doctor from medical image data having a particularly high degree of redundancy. This characterization image processing is, for example, a pre-processing technique, and includes speech recognition, pattern recognition, difference processing, differentiation processing, and the like, and has an expert function of learning recognition by a doctor. Further, an assignment for identification signal search such as automatic assignment of a keyword required for search can be performed to support on-site operations. The present invention can be implemented as an auxiliary system of a comprehensive medical information system by sharing this identification signal, particularly a patient identification number, with an existing hospital information system.

In FIG. 4, the manager unit 19 has a function of managing the state transition of the entire expert function.
For example, if an unexpected operation occurs, for example, the same operation is repeated three times, and if the operation does not return to the normal state, the manager unit 19 outputs an alarm display and an alarm sound, and stands by in the same state. Only the manual operation by the operator is accepted, and the operation can be returned to the normal operation.

The input / output control unit 20 includes a target section designation function,
That is, it includes a reservation function, a cue function, and a video device control function. The target category designation function means that the capture, that is, the header information of the target recording tape to be selected is a single or multiple (eg, 10) headers,
This is a function to specify reservation in chronological order. Regarding the reserved header, even after the target image is detected and acquired,
Pass through the expert function. The cueing function has a function of cueing a measurement start portion along a reservation order by making full use of the video device control function according to a definition that defines the acquisition unit 2 and connection conditions. The video device control function is a function for performing control in accordance with an instruction from the expert manager unit 19, for example, power on / off, magnetic tape 17
This includes fast forward / rewind / playback, speed control, that is, selection of normal, slow or double speed, and jog / shuttle, that is, a function of performing forward or backward during display of the magnetic tape 17.

The capture unit 21 has an image compression function and an image transfer function. The output unit 22 functions to transmit the image extracted by the expert function to the communication line 1. The output unit 22 includes an image compression transmission circuit 39,
The image data extracted via the communication line 1 that is a LAN is transmitted to the database unit 4. Further, an attribute table maintenance unit 23 is provided, and performs a function of adding an attribute, for example, the above-described identification signal to the extracted image and storing the extracted image.

FIG. 5 is a block diagram showing the structure and operation of the extraction unit 3. FIG. 6 is a flowchart for explaining the operation of the manager unit 19 and the capture unit 21 in the extraction unit 3 shown in FIGS. It is. Referring to these drawings, the input / output control unit 20 first moves from step a1 to step a2 in FIG.
The magnetic tape 17 is reproduced. The image data to be extracted is
It differs depending on the medical subject and each part of the human body 12 input by the input / output control unit 20 or the like. The outline of the extracted image data of the medical treatment subject and the detection site of the human body is as shown in Table 1 below.

[0025]

[Table 1]

Thus, it is determined whether the image data to be extracted is a moving image or a still image. The input / output control unit 20 can also manually input whether the image data to be extracted is a moving image or a still image. If it is a moving image, it is an extraction start image, and if it is a still image, it is one extracted image. When it is instructed in step a4 to extract moving image data, the next step a5
In steps a to a8, characterization image processing is performed. In the present embodiment, a so-called template matching method is performed. First, in step a5, characterization image processing is performed.

In step a5, an image for each specific medical subject to be an expert is used as a model, and a characterization image processing is carried out on the model to create a reference image, which is stored as a reference template. Keep it. When the same part of the compared image is captured, after performing the same processing as the above-mentioned reference template creation, the comparison image is compared with the reference image, centering on both features, and the degree of matching is measured. To decide. Extraction is
Since it is extremely difficult to determine the extraction using the raw image as it is, a method is used in which the image processing is performed as simple as possible, and the image is searched for in the comparison image based on the image processing.
In this way, the comparative image is also processed according to the procedure for creating the reference image, and if the comparison between the two matches the shape and color extraction amount at a certain ratio (threshold value, that is, ΔL0 described later), it is determined as an extraction point. I do.

The characterization image processing of the image data in step a5 in FIG. 6 will be further described. FIG. 7 shows image data obtained by imaging the human body 12 as a patient with the television camera 10 shown in FIG. 3 and recorded on the magnetic tape 17 for a relatively long time, for example, 10 minutes to 1 hour. As shown in FIG. 7 (1), a state in which the human body 12 starts moving in accordance with the voice instruction of the operator 8 such as a doctor or in accordance with display instruction means such as a lamp while standing up is shown in FIG. 7 (2). It becomes as follows. Further, the human body 12 moves as shown in FIG. The television camera 10 images the entire human body 12 as shown in FIG.

In another embodiment of the present invention, instead of imaging the entire human body 12, as shown in FIG. 8, a state in which the lower limbs 27 of the human body 12 are moving starting from a stopped state is a television camera. 10 may be taken. At this time, the human body 12, which is a patient, swings the lower limb 27 while sitting on a chair, for example.

Further, as shown in FIG. 9, the upper limb 28 of the human body 12 may be swung, and the state may be imaged by the television camera 10. As described above, the television camera 10 may image each part constituting a part of the human body 12.

FIG. 10 is a flowchart for explaining a more specific operation of the characterization image processing.
For example, when image data is obtained as shown in FIGS. 7 to 9, the process proceeds from step b1 to step b2 to detect a skin area exposed from clothes of the human body 12.
The television camera 10 can obtain color image data, and detects, in the obtained image data, an area in one screen having a skin color. In the next step b3,
The coordinate position and area of the pixel for each region obtained in step b2 described above are obtained, and the coordinates of the centroid of each region are calculated and obtained. In this manner, in step b4, the coordinates of the centroid of each region at each predetermined time are detected, and the time change rate L1 of each region is calculated and obtained. For example, when the horizontal direction in FIG. 7 is the X axis and the vertical direction is the Y axis, L1 = dx
/ Dt or L1 = dy / dt. Where x, y
Are the coordinates (x, y) of the centroid of each area of the skin exposed from clothing such as a face or hands. Thus, the time change rate L1 may be, for example, the moving speed in the X-axis direction and / or the Y-axis direction. FIG. 7 to FIG.
The characterization image processing operation indicated in relation to 0 is performed for extracting a moving image or a still image from the image data.

As shown in FIG. 11 for a still image,
In a visual field 30 of a part of the skin of the human body 12 which is a photograph of the skin symptom, the vicinity of a discolored area 31 greatly different from the normal skin color is extracted as a high-definition still image data in a visual field 30 of the camera 10. You can also.

Further, as shown in FIG. 12, a field of view 33 of a part of the back 32 of the human body 12 is picked up by the television camera 10, and the discolored area 34 is set as a still image in the same manner as in FIG. Data may be extracted.

An example of the characterization image processing for still image data described in connection with FIGS. 7 to 12 is shown in FIG. Step c1 to step c
2 and the data of one screen of the image data is taken out. In the next step c3, for example, FIGS.
As described in relation to the above, areas 31 and 34, which are discolored portions where the color has changed greatly from the normal skin color, are detected, and each area Sj is calculated. In step c4, the area 3
The number j in the fields of view 30, 33 of 1, 34 is counted.
Thus, in step c5, the ratio L2 (= Σ) of the total area ΣSj of the discolored regions 31 and 34 occupying the visual fields 30 and 33 in the entire area S0 of the back 32 which is the skin in one screen.
Sj / S0), and a series of operations is ended in step c6.

When the ratio L2 of the areas occupied by the discolored regions 31 and 33 is large, the image data is extracted as still image data as described later. In order to accurately obtain the ratio L <b> 2, the visual field regions 30 and 33 are set such that a part of the human body 12 to be imaged, such as the hand or the back of the hand, is set to a predetermined position at the time of imaging by the television camera 10. The optical axis to be imaged by the camera 10, the focal length of the optical lens, the magnification of the optical lens, and the like are set to predetermined values.

At this time, referring to FIG. 6, as described in connection with FIGS. 7 to 13, the characteristic values L1 or L2 obtained by performing the characterization image processing (sum of Li and i
Is 1 or 2 in each of the above-described embodiments), and is compared with the template in step a6. This template is a predetermined value L0i (i is 1 or 2 as described above) individually corresponding to Li of the image data.

The degree of matching ΔLi (= Li−
L0i) is calculated and measured. In the next step a8,
It is determined whether the matching degree ΔLi is smaller than a predetermined value ΔL0i. If so, the matching degree between the image data recorded on the magnetic tape 17 and the template is high and good.

In step a9, extraction of image data of the magnetic tape 17 having a good matching degree for a moving image is started. From the start of the extraction, a time T is measured by a timer in step a10, and the time T is equal to a predetermined time T
0, for example, after 1 minute, the extraction is terminated,
In step a12, extraction of the moving image data is ended.

In extracting image data of a still image,
The process moves from step a13 to step a14. In step a14, the above-described steps a5 to a8 described in connection with the extraction of the moving image data are executed. Thus, one image data is extracted as a still image in step a15.

The extraction of moving image or still image data is as follows.
An input is instructed in the input / output control unit 20. The engine 36 of the extraction unit 3 in FIG. 5 has a configuration for executing each operation step in FIG. 6, and particularly performs a characterization image processing process for each subject to be treated as an expert. Thus, the image extraction operation indicated by reference numeral 37 in FIG. 5 is performed as shown in FIG.

The extraction unit 3 also includes a magnetic tape 17
A memory 40 is provided for storing the image data taken from the memory 4, and further stores the image data calculated by each engine 36 and / or the data related thereto.
1 and can be visually displayed by the display means 42. With the learning function 38,
The template is corrected, and the optimal template for each medical department is set. The data obtained by the learning function 38 can be stored in the memory 43, and the stored contents can be read out to be useful for extracting the next image data.

In another embodiment of the present invention, FIG.
Instead of steps a5 to a8 described in relation to
As shown in FIG. 7, the human body 12 as a patient generates sound, which is detected by the microphone 29 and stored on the magnetic tape 15 together with the image data. When this audio data is detected in step d1 of FIG. 14, steps a9 and a9 for extracting a moving image or a still image
15 is executed.

In still another embodiment of the present invention, FIG.
When the patient 12 operates the operation switch 57 as shown in FIG. 15 instead of the steps a5 to a8, the output of the operation switch 57 is recorded on the magnetic tape 17 together with the image data by the magnetic tape recording device 15. Is done. When the switch 57 is operated by the human body 12, the operation steps a9 and a15 for extracting the image data of the moving image and the still image in FIG. 6 are performed.

FIG. 16 is a block diagram for explaining a specific configuration and operation of the database unit 4. As shown in FIG. The image data extracted via the LAN communication line 1 is received by the communication control circuit 45 and stored in the memory 46.
It is stored together with the identification signal for searching and is made into a database. The identification signal is also output by the man-machine center connection circuit 47, and the type of image data to be stored can be selected. The contents stored in the memory 46 can be processed by the processing circuit 49 and displayed by the visual display means 50. Further, the contents stored in the memory 46 can be stored in the auxiliary memory 51. Further, the content stored in the memory 46 is image-compressed by the image compression processing circuit 52, and can be transmitted to the terminal device 6 by the transmission unit 53 via the communication line 5 such as a LAN. Further, data such as image data from the terminal device 6 can be received by the receiving means via the communication line 5.

Further, the man-machine center connection circuit 47
Is transmitted by the server 54 to an image server 56 (see FIG. 1 described above and FIG. 18 described later) via a communication line 55 such as a LAN.

FIG. 17 is a block diagram showing a specific configuration of the terminal device 6. The terminal device 6 includes some components of the acquisition unit 2 and the extraction unit 3 described above, and corresponding portions are denoted by the same reference numerals.

FIG. 18 is a block diagram showing a specific configuration of the image server 56. The image server 56 shown in FIG. 18 is a device for storing and retrieving image data. The image server 56 stores extracted image data of moving images or still images obtained by an expert system for medical image data processing, compresses or expands the image as necessary, and stores it in a searchable state as a database. store.

For the communication line, for example, multiplexing of a transmission system such as ISDN or PHS can be performed, and the communication line can be made portable.

[0049]

According to the present invention, medical image data is
For example, image data of a moving image or a still image obtained by imaging using an analog camera or the like is obtained by characterization image processing, compared with a prepared template, and when the matching degree is high, the image data is extracted. Since the image data is stored in the extraction memory, only the image data necessary for medical diagnosis can be stored in the extraction memory.
As a result, the capacity of the extraction memory can be reduced, and diagnosis by a doctor or the like can be performed quickly.

The moving image data is stored in the extraction memory for a predetermined period of time from the image data having a high degree of matching. For example, a series or intermittent image data after a part of the human body starts moving is stored. Can be kept. Thus, extraction of moving image data is easy.

The image data determined to have a high degree of matching is stored in an extraction memory, so that still image data can be easily extracted. In particular, according to the present invention, the television camera 10
In the recording device 15, for example, image data of a moving image for 10 minutes to 1 hour is stored, and a rolling state moving image and an initial state moving image for each medical treatment subject and each part are obtained from the obtaining unit, and the moving image of the patient is obtained. The coordinates of the centroid of each area exposed from the clothes are detected, and the time rate of change L1 is calculated.
By comparing the moving image matching value ΔL1 (= L1−L01) by comparing with the moving image template value L01, if the moving image matching value ΔL1 is less than the predetermined moving image value ΔL01, it is determined that the moving image matching degree is high. Image data is extracted for a predetermined time T0, for example, for one minute, stored in the extraction memory 41, and displayed by the display unit. As a result, the capacity of the extraction memory 41 can be reduced, and diagnosis by a doctor or the like can be performed quickly. The recording device 15 may be a magnetic tape, for example, as described above, and the extraction memory 41 stores the extracted image data together with the patient's identification number, and can be realized by, for example, a semiconductor memory.

The image data stored in the extraction memory is transmitted by the first transmission means via the communication line and stored in the database, and the image data stored in the database memory is searched by the identification number. Is transmitted by the second transmitting means via the communication line, and the receiving device can be used in another place, so that the extracted image data can be effectively used in various places. In addition, appropriate medical treatment can be performed even in an emergency.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an overall configuration of an embodiment of the present invention.

FIG. 2 is a block diagram for explaining the configuration and operation of an acquisition unit 2 shown in FIG.

FIG. 3 is a block diagram illustrating a configuration of an acquisition unit 2.

FIG. 4 is a block diagram showing a configuration of an extraction unit 3 of FIG.

FIG. 5 is a block diagram for explaining the configuration and operation of the extraction unit 3.

6 is a flowchart for mainly explaining operations of the input / output control unit 20 and the capture unit 21 in FIG. 4 of the extraction unit 3 and the engine 36 and the image extraction unit 37 in FIG.

FIG. 7 is a diagram illustrating a moving state of a human body 12 for extracting image data of a moving image or a still image.

FIG. 8 is a diagram for explaining a moving state of the lower leg 27 of the human body 12.

FIG. 9 is a diagram for explaining a moving state of the upper limb 28 of the human body 12;

FIG. 10 is a flowchart showing a specific operation of characterization image processing.

FIG. 11 is a diagram showing a field of view 30 of a part of a hand for extracting image data of a still image.

FIG. 12 is a diagram for explaining image data of a visual field 33 on the back 32 of the human body 12;

FIG. 13 is a flowchart for explaining an operation of characterization image processing for extracting the image data of the still image shown in FIGS. 11 and 12;

FIG. 14 is a flowchart illustrating a part of an operation for extracting image data of a moving image or a still image according to another embodiment of the present invention.

FIG. 15 is a flowchart showing a part of an operation for extracting image data of a moving image or a still image according to still another embodiment of the present invention.

FIG. 16 is a block diagram for explaining the configuration and operation of the database unit 4;

FIG. 17 is a block diagram for explaining the configuration and operation of the terminal device 6.

FIG. 18 is a block diagram for explaining the configuration and operation of an image server 56.

[Explanation of symbols]

 1 Communication line such as LAN 2 Acquisition unit 3 Extraction unit 4 Database unit 5 Communication line such as ISDN or PHS 6 Terminal device 8 Operator 9 Operating means 10 Television camera 12 Human body 15 Video tape recording means 16 Processing circuit 17 Magnetic tape 19 Manager Unit 20 input / output control unit 21 capture unit 22 output unit 23 attribute table maintenance unit 30, 33 field of view 31, 34 discoloration area

Claims (4)

(57) [Claims] 1. A television camera 10 for imaging a patient's human body 12 for a predetermined time to obtain moving image and still image data, and a driving unit 1 for changing the optical axis of the television camera 10 three-dimensionally.
1, operating means 9 operated by the operator 8, control means 13 for driving the driving means 11 in response to the output of the operating means 9, and the output of the television camera 10 so that the operator 8 can see the output. Monitor 14 for visual display, and recording device 1 for storing image data of TV camera 10
5, setting means for setting a medical subject and a region; and reproducing the image data stored in the recording device 15 in response to the output of the setting device, and the general elbows and arms of the patient's body for each medical subject and each region. , Regarding the part of the head above the neck,
An acquisition means for acquiring a video of a rolling state, an initial motion state video from a walking start instruction to the start, and in response to an output of the acquisition means, determine a coordinate position and an area of each area of the skin exposed from the patient's clothing. , The coordinates of the centroid of each region are detected, and the time change rate L1 of the centroid of each region is detected.
And a moving image template setting means 4 for storing a predetermined moving image template value L01 for each medical treatment subject and each site.
3 and the response from the time change rate calculating means and the moving image template setting means, and the moving image matching degree ΔL1 (= L1-L01)
And a moving image matching degree ΔL1 among the image data stored in the recording device 15 in response to the output of the moving image matching degree calculating means.
Extracting means for extracting image data for a predetermined time T0 from the start of extraction when image data having a value less than the predetermined value ΔL01 for a moving image is obtained; and extracting the image data in response to the output of the extracting means. And a display means 42 for visually displaying the extracted image stored in the extraction memory. . 2. A still image data of one screen of the image data stored in the recording device 15 is taken out, and a discolored area 31, which is a changed part where the color has changed greatly from the normal skin color, is obtained.
A color-change area detecting means for detecting the color-change area 34;
A color changing area calculating means for calculating each area Sj of the color changing area 3 in the entire area S0 of the skin in the one screen
The ratio L2 of the total area ΣSj of 1,34 (= ΣSj / S
0), a still image template setting means for storing a predetermined reference still image template value L02, and a still image template responding to the output of the ratio calculating means and the still image template setting means. The still image matching degree calculating means for calculating the matching degree ΔL2 (= L2-L02) and the still image matching degree of the extracted still image in response to the output of the still image matching degree calculating means Predetermined value ΔL
2. The medical image data processing apparatus according to claim 1, further comprising means for storing image data of a still image which is less than 02 in the extraction memory 41. 3. The medical image data processing apparatus according to claim 1, wherein the extraction means extracts the image data for a predetermined time retroactively from the start of the extraction. 4. A first transmission means for transmitting image data stored in the extraction memory 41 and an identification signal corresponding to the image data via a communication line, and a first transmission means via the communication line. A database memory for storing the image data corresponding to the identification signal, and reading the image data corresponding to the identification signal input to the database memory and transmitting the read image data via a communication line.
The medical image data processing apparatus according to claim 1, further comprising: a transmitting unit configured to receive and display image data from the second transmitting unit via the communication line.