JP2000350701A - Image processing apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize improvement of time efficiency of image processing operation by extracting characteristic points corresponding to each characteristic point set by a characteristic point setting means on another medical image, and allotting all the regions of interest in taken medical images so as to be visibly recognizable based on the extracted results. SOLUTION: A taken medical image is inputted as image information, the image which becomes a standard for making region setting of region of interest(ROI) is displayed (a), and ROI which becomes a standard for region setting ROI of a medical image displayed on a TV monitor is set (b). Then, optical flow is prepared (c) which shows time series movement of each characteristic point by extracting and setting characteristic points of ROI of next time series for the displayed standard image, and by adding a move parameter to the characteristic point and making operational treatment, and the standard ROI is transferred and deformed based on it. Then, it is judged whether a medical image which has not been treated with (a) to (d) in plural medical images taken in time series, and, if so, returning to (c) and a similar treatment is repeated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus suitable for being applied to a medical image diagnostic apparatus which captures a medical image used in medical examination, diagnosis, treatment and the like of a subject.

[0002]

2. Description of the Related Art Conventionally, various medical image diagnostic apparatuses for obtaining a medical image by noninvasively imaging a lesion site of a subject have been developed and applied. As such a medical image diagnostic apparatus, an ultrasonic image diagnostic apparatus, a CT image diagnostic apparatus, an X-ray image diagnostic apparatus, an MRI image diagnostic apparatus, and the like are generally used. According to these imaging means, a high-definition medical image can be obtained in a short time, especially with the recent development of image processing technology.

[0003] Based on these medical images, methods of image diagnosis utilizing characteristics unique to each modality have been established, and research presentations have been actively made at medical societies in various specialized fields. . One of the common features of today's medical image diagnostic techniques is
For example, interpretation of a lesion site such as a tumor has occupied an important position.

[0004] Compared with the past medical image diagnostic apparatuses, medical images provided by today's medical image diagnostic apparatuses have high resolution, and in particular, the region of interest R
OI (Regin of Interest) is remarkable in delineating a lesion site such as a tumor. Regarding the interpretation of parenchyma, such as a tumor, it is important to determine whether the boundary between the tumor and surrounding normal tissue is finely defined. The judgment of benign and malignant is given. By providing a high-resolution medical image, it is possible to accurately capture a boundary tissue such as a tumor, thereby providing more accurate diagnostic information.

A medical image picked up by the medical image diagnostic apparatus is stored in an image processing apparatus built in the medical image diagnostic apparatus or provided externally and connected thereto. The image processing apparatus can store and classify medical images and perform marking on ROIs by processing the captured medical images as image information. The configuration for performing such image processing is a general schematic configuration as shown in FIG.

A medical image picked up by the image diagnostic apparatus is input as image information. The input image information is temporarily received by the data buffer 8. By providing the data buffer 8, for example, when a large amount of image information is generated at a time by performing imaging on the image diagnostic apparatus side, a large amount of image information is temporarily stored regardless of the operation state of the image processing apparatus side. Can be kept. As a result, it does not affect the imaging on the image diagnostic apparatus side and does not affect the operation on the image processing apparatus side.

The image information from the data buffer 8 is sent to the CPU 1 in this way. The CPU 1 performs a predetermined calculation process on the input image information. In order to perform the predetermined arithmetic processing, the external input device 3, the trackball 2, the RO
An M4, a storage device 5, and an external storage device 6 are provided.
The external input device 3 is provided, for example, in a main body of the image processing apparatus, and receives an instruction regarding image processing of an operator. Similarly, the trackball 2 receives the instruction of the operator from the CPU 1.
This is a means for inputting to.

An algorithm for a predetermined operation, a predetermined coefficient, an initial set value, self-diagnosis information at the time of initial startup, an emergency operating condition at the time of an apparatus error, and the like, which are input in advance in the ROM 4 for the purpose of design. Are stored as data. Further, the storage device 5 is configured by a storage element such as a RAM, for example, which can freely write and read information. The storage device 5 temporarily stores intermediate progress information that the user wants to temporarily store during the arithmetic processing of the CPU 1, an operator's instruction from the external storage device 3, or image information output from the data buffer 8. Used for

The external storage device 6 is composed of, for example, a hard disk drive (HDD). The image information obtained by the arithmetic processing in the CPU 1 or the image information of a large amount of information output from the data buffer 8 is stored. If a large amount of image information is output from the data buffer 8 while the CPU 1 is performing arithmetic processing on one piece of image information, the CPU 1 temporarily stores the image information in the external storage device 6 and re-executes the arithmetic processing later. It can be performed.

The configuration of the external storage device 6 is, for example, M
According to a storage / reproducing device such as O, CD-ROM, DVD, etc., image information obtained by another image diagnostic device or image information obtained by another modality can be image-processed. Further, the image information after the image processing is completed can be stored and stored.

A D / A converter 7 is connected to the CPU 1 and converts image information into an image signal that can be output to an external output device 10 such as a TV monitor 9 or a printer. Although not shown, it is also possible to output image information as a digital signal without passing through the D / A converter 7. For example, digital transfer can be made to a remote place without deterioration of image information.

In the image processing apparatus having the configuration shown in FIG. 5, an image is displayed on the TV monitor 9 when image information is input. The operator operates the external input device 3 and the trackball 2 while watching the displayed image to execute image processing on the CPU 1.
Input instructions.

Here, when the trackball 2 is operated, T
The image displayed on the V monitor 9, that is, the medical image obtained by capturing the ROI that is the subject of the subject,
Marking (area designation) is performed to clearly distinguish the OI from surrounding tissues. More specifically, the trackball 2 is operated, for example, by tracing a boundary line between a parenchymal tissue portion and a surrounding tissue portion of a tumor or a valve of a heart. By tracing the boundary line, a trajectory of the trackball 2 is drawn on the boundary line. The ROI is designated by drawing this locus. Of course, the drawn trajectory can be hidden, or the portion surrounded by the trajectory is displayed as a ROI so as to be visually recognized by color, brightness, or other expressions.

When a medical image diagnostic apparatus continuously images a subject to be imaged in a time-series manner, a large number of medical images are generated as image information at one time, depending on the imaging performance. This generated image is the ROI displacement /
The deformation is continuously recorded, and the displacement / deformation of the ROI can be grasped by observing the images in chronological order.

For example, when the ROI is a heart valve or the like,
When there is a respiratory movement, it becomes possible to examine the tissues in the vicinity of the valves and the diaphragm which are the respective target areas. Other RO
For example, when examining a tumor in the liver, a pustular tissue in the thyroid gland, a strong reflection echo tissue in the prostate, or the like where the I does not move, the imaging cross section itself may be moved in time series. The operation of this movement is performed by changing the imaging section,
This is because it is necessary to grasp the three-dimensional shape of the OI.
The three-dimensional shape can be easily grasped by moving the imaging cross section continuously little by little and performing imaging in time series.

[0016]

However, in the conventional image processing apparatus, an operator, for example, a doctor or a clinical laboratory technician selects an arbitrary one of many time-series medical images taken, R imaged in the selected medical image
The area was specified only for the OI.

[0017] For this reason, other than the selected arbitrary one,
RO captured in other medical images captured in chronological order
In order to specify an area in I, it was necessary to call an image and perform an area specification operation each time. When there are a large number of captured images, it is necessary to repeatedly specify the region for the ROI many times, resulting in a heavy burden on the operator.

In addition, it is necessary to image and examine a large number of subjects within a limited time. For example, in outpatient clinics in relatively large hospitals, the decrease in patient throughput is ignored. This is an issue that cannot be obtained. In order to follow the displacement / fluctuation of the ROI of a medical image taken for examination / diagnosis in chronological order, it is necessary to manually specify an area for each required image, thereby reducing work efficiency. Had led to a decline.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and only sets an ROI to an arbitrary one of a plurality of consecutive medical images in a time series, and the other remaining medical images are used. It is an object of the present invention to provide an image processing apparatus in which a region of an ROI corresponding to an image is set and the time efficiency of an image processing operation can be improved.

[0020]

According to a first aspect of the present invention, there is provided an image processing apparatus capable of displaying a medical image including a set region of interest in a time-series manner. An area designating unit capable of designating a region of interest in an image so that the region of interest can be visually recognized;
A feature point extracting unit that extracts feature points corresponding to each feature point set by the feature point setting unit on another medical image, and all the medical images captured based on the extraction result of the feature point extracting unit. An image processing apparatus comprising: an area resetting unit that specifies an area of interest in an image so that it can be visually recognized.

According to a second aspect of the present invention, in the image processing apparatus according to the second aspect, the feature point extracting unit detects a point in time when the medical image has no feature point corresponding to the feature point set by the feature point setting unit. An image processing apparatus according to claim 1, further comprising an extraction processing ending means for ending the characteristic point extraction processing.

According to the image processing apparatus according to the first aspect of the present invention, by setting a region of interest for an arbitrary one of a plurality of medical images, it is possible to set a region of interest in another medical image. The region of interest to be set is set by the region resetting means.

According to the image processing apparatus of the present invention, when there is no corresponding region of interest in another medical image, the region setting by the region resetting means is stopped.

An object of the image processing apparatus of the present invention is to provide an image processing apparatus in which an area is set in any one medical image, so that an area is set in a substantial portion of another medical image. .

[0025]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An image processing apparatus according to the present invention can be applied to ROI setting for medical images in a time series taken by a medical image diagnostic apparatus.

<First Embodiment> The first embodiment of the present invention can be applied to the configuration of the image processing apparatus already shown in FIG. FIG. 1 shows a flowchart of ROI setting applied to the configuration of the image processing apparatus of FIG.
In the following description, the configuration of the image processing apparatus in FIG. 5 will be described.

According to the flowchart of FIG. 1, a medical image picked up by a medical image diagnostic apparatus (not shown) is input as image information. As the input image information, an image serving as a reference for setting an ROI area in step (a) is displayed. This image display is performed by the TV monitor 9 shown in FIG. In step (b), the operator sets a reference ROI in order to set an ROI in the medical image displayed on the TV monitor 9. This ROI
In the area setting, the operator visually observes the medical image displayed on the TV monitor 9 and sets the area using the trackball 2 and the external input device 3 for the read ROI. Also, for the set ROI, the ROI
The points that characterize the shape or range of the area are set as feature points at a plurality of locations. The setting of the feature points will be described later in detail.

In step (c), the ROI of the next time-series image is extracted and set for the reference image displayed in step (a). This R
The OI and the feature points correspond to the ones set in step (a). The newly extracted feature point is compared with the feature point extracted in step (b) to determine a motion parameter between the reference image and the next image.

In order to obtain the motion parameters, calculation is performed according to the image processing flow shown in FIG. (A) indicates the reference ROIt, and t is a value indicating the time t at which the medical image was captured in time series among the plurality of medical images. The reference ROIt is a portion specified by the operator on the medical image using the trackball 2 or the external input device 3. (B) is a medical image whose time value advances by +1 from the time at which the medical image of (a) was captured, and is captured again, and is indicated by an image t + 1. (C) is an image on which the reference ROIt shown in (a) is designated when an area is designated, and is represented by a reference image t. (D) is a medical image captured in time series with respect to the reference image t of (c) at time t-1, and is shown as an image t-1.

The calculation of the motion parameters is performed by calculation using the medical image information (a) to (d). This calculation is performed in (e), and the outline of the calculation procedure is as follows.

The reference image displayed on the TV monitor 9 is observed, and the ROI displayed on the image is designated using the trackball 2 or the external input device 3. As a result, the reference ROI is determined, and the reference image is represented by t indicating which elapsed time of all the medical images captured in time series is located on the time axis. (A) Reference ROIt
And (c) the reference image t are set in this way.

(C) A medical image taken after +1 unit time has elapsed based on the reference image t. (B) Image t + 1
Similarly, a medical image captured in -1 unit time (d)
A total of two medical images including the image t-1 are selected. The two medical images are compared with (c) the reference image t, and the difference is obtained by calculation and used as the motion amount. For example,
By comparing (b) the image t + 1 with (c) the reference image t, +
The amount of motion after one unit time can be obtained. Also,
(D) Comparing the image t-1 with the (c) reference image t, -1
The amount of movement from the unit time before to the unit time t can be obtained.

As one method for obtaining the amount of motion,
(C) Divide the reference image t into a plurality of blocks. Since each block is used as a reference for grasping the amount of motion, the observation accuracy of motion improves as the number of blocks increases. For example, luminance, contrast, position coordinates, and the like, which are image information constituting the image of the block portion divided into the plurality, are set as evaluation criteria.
Similarly, a plurality of blocks are divided for the (b) image t + 1 and the (d) image t-1, and similar evaluation criteria are set for each block.

Each block set in this way is specified as individual information based on the evaluation criterion, and when one block is determined based on (c) the reference image t, (b) the image t
+1 or (d) In the image t-1, what kind of movement is performed is extracted. The extracted value is obtained as a motion parameter of each block.

On the other hand, (a) a feature point for specifying the ROI is set in the designated area of the reference ROIt. By setting the feature points at a plurality of locations, the accuracy of specifying the ROI is improved. This feature point is selected from each block set for extracting the motion parameter. For example, ROI
All the blocks located on the boundary line surrounding the set area may be set as feature points, and the ROI can be specified with higher accuracy by taking into account the distribution of the blocks in the part within the set area.

By adding the motion parameters to the ROI feature points set in this way and performing arithmetic processing, an optical flow indicating the temporal movement of each feature point can be created. This optical flow can extract two motions of (e () optical flow and (e (() optical flow.) The former is the motion from the image t-1 to the image t, and the latter is the motion from the image t to the image t + 1. Represents movement.

Here, as shown in FIG. 4, (a)
The optical flow of (c) is added to the reference ROI by the calculation. In the figure, the length, direction, and starting point are shown as an aggregate of different arrows, but this shows the concept of optical flow. The starting point of each arrow corresponds to the coordinate position of the set block. Also,
The position of the starting point indicates the position in the image t serving as a reference. The arrow represents a mathematical vector parameter, the length is the movement amount itself, and the direction is also the movement direction itself.

The deformed ROI of (b) is displaced / deformed by the optical flow of (c). The displacement / deformation of the ROI is a result of adding the optical flow to the feature point of the reference ROI of (a), and based on only the optical flow of the block set as the feature point of the ROI in the optical flow of (c). Have been.

Based on these optical flows,
(A) Displace / deform the reference ROI in step (f). The movement parameters for displacement / deformation are thus extracted from the (a) designated area of the reference ROI on the medical image, and (g) ROIt + 1 and (h) ROIt-1.
Is displayed after the displacement / deformation of.

Here, referring again to FIG. 1, the determination shown in step (e) is performed. For a plurality of medical images taken in time series, it is determined whether or not there is a medical image for which processing from step (a) to step (d) has not been performed. If there is an unprocessed image, the process returns to step (c) and step (d), and the same processing is repeated. By the way, regarding the processing to be repeated here, the designated area of the reference ROI as a reference is based on the reset ROI after the original displacement / deformation processing. Of course, when the ROI displacement / deformation processing has been completed for all medical images, the processing flow ends.

According to the image processing apparatus by such a series of processing, the operator designates an ROI area for an arbitrary image, and based on image information of the designated ROI, another medical device is used. In the image, the region of the ROI including the temporal displacement / deformation is specified without depending on the operation of the operator.

As a result, R can be used for many medical images.
Even when setting the OI area, it can be easily performed in a short time, and the time efficiency can be improved.

<Second Embodiment> The processing for incorporating and displaying temporal displacement / deformation in a plurality of medical images is the same as that described in the first embodiment. Therefore, in the description of the second embodiment according to the present invention, the description of the technical overlapping parts will be omitted for easy understanding.

In the processing flow of FIG. 2, step (c) is a new component. For the processed medical image,
It is determined whether or not the ROI exists in the unprocessed image based on the feature points, the eigenvalue of each block, and the like. When there is an ROI, the displacement / deformation process is repeated, but when there is no ROI, a series of processes is stopped.

With such a configuration, even when, for example, a large number of medical images are targeted, the displacement / deformation processing is performed only on the image on which the ROI is imprinted, so that the time required for image processing is reduced. It is possible and efficient.

The embodiments described above are described for facilitating the understanding of the present invention, and are not described for limiting the present invention. Therefore, each element disclosed in the above embodiment is intended to include all design changes and equivalents belonging to the technical scope of the present invention.

[0047]

According to the image processing apparatus of the present invention configured as described above, by setting an area on any one medical image, an area can be set on a substantial portion of another medical image.
It is possible to provide an image processing apparatus that stops the area setting by the area resetting means when there is no corresponding area of interest in another medical image.

[Brief description of the drawings]

FIG. 1 shows a flowchart in an image processing apparatus according to a first embodiment of the present invention.

FIG. 2 shows a flowchart in an image processing apparatus according to a second embodiment of the present invention.

FIG. 3 is a diagram for explaining an outline of image processing in the image processing apparatuses according to the first and second embodiments of the present invention.

FIG. 4 is a diagram illustrating an outline of image processing in the image processing apparatuses according to the first and second embodiments of the present invention.

FIG. 5 is a schematic diagram showing the overall configuration of a conventional image processing apparatus.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... CPU 2 ... Trackball 3 ... External input device 4 ... ROM 5 ... Storage device 6 ... External storage device 7 ... D / A converter 8 ... Data buffer 9 ... TV monitor 10 ... External output device

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G06F 15/66 470A F-term (Reference) 4C093 AA22 CA18 FF17 FF18 FF28 FF31 4C096 AB39 AD14 DC20 DC21 DC28 DC31 5B057 AA07 AA08 AA09 CA02 CA08 CA12 CB02 CB08 CB12 CC03 CE09 DA08 DB02 DB05 DB09 DC05 5C076 AA01 AA02 AA40 CA02

Claims (2)

[Claims] 1. An image processing apparatus capable of displaying a medical image including a set region of interest in chronological order, comprising: a region specifying means capable of specifying a region of interest in an arbitrarily selected medical image so that the region of interest can be visually recognized; A feature point setting means for setting a plurality of feature points of an area designated by the area designation means; and a feature point for extracting feature points corresponding to each feature point set by the feature point setting means on another medical image. An image processing apparatus comprising: an extracting unit; and an area resetting unit that specifies an area of interest in all of the captured medical images so as to be visually recognizable based on an extraction result of the feature point extracting unit. . 2. The method according to claim 1, wherein the feature point extracting unit terminates the extraction process of the feature point when the medical image has no feature point corresponding to the feature point set by the feature point setting unit. The image processing apparatus according to claim 1, further comprising: