JPH07246207A - Ultrasonic image processor - Google Patents

Abstract

PURPOSE:To suitably and speedily extract the contour line of a tissue inside an ultrasonic image. CONSTITUTION:Plural sample points S1 and S2 are set on a border line 102 of tissues by an operator. A reference straight line Ml connecting those two sample points is set, and reference points P1 and P2 are set at positions away from both the ends of that straight line just for a fixed distance K. Retrieval straight lines L1 and L2 are perpendicular through the reference points P1 and P2 to the reference straight line M1, and the border line 102 is extracted along those retrieval straight lines L1 and L2. This processing is repeatedly executed to an interpolating block with calculated interpolating points A and B as both the ends.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to an ultrasonic image processing device,
Particularly, the present invention relates to an ultrasonic image processing apparatus capable of extracting a tissue contour for cross-sectional area calculation and the like.

[0002]

2. Description of the Related Art In an ultrasonic diagnostic apparatus, an ultrasonic image (a tomographic image or the like) is formed by transmitting and receiving ultrasonic waves to and from a living body. Conventionally, the following method has been used to calculate the cross-sectional area of a living tissue (for example, liver) that appears in a two-dimensional tomographic image.

The first conventional method is an artificial tracing method in which an observer moves a cursor, for example, while watching a two-dimensional tomographic image to input and specify the contour of a living tissue. The second conventional method is an automatic tracing method, in which an image processing apparatus for processing an ultrasonic image extracts a contour of a living tissue by an automatic calculation by utilizing a variance value of brightness levels and the like. . There is also a method of setting a major axis point and a minor axis point and performing a cross-sectional area calculation or a volume calculation described below, on the assumption that the tissue shape is an ellipse.

In order to calculate the volume of a tissue, it is necessary to extract the surface of the tissue three-dimensionally, and even in that case, surface extraction is performed based on the above-mentioned method. By the way, the present applicant has proposed an ultrasonic image processing apparatus capable of automatically extracting a living tissue three-dimensionally in Japanese Patent Application No. 5-152272.

[0005]

However, the above-mentioned first conventional method has a problem that it is inefficient and burdens the observer too much, especially when the contour is complicated. Further, in the above-mentioned second conventional method, if the contour extraction is automatically performed on the ultrasonic image containing a lot of noise, there is a problem that the original contour is likely to be lost on the way. there were.

[0006] In the conventional automatic contour extraction processing, extraction is performed sequentially along a coordinate axis fixedly set, for example, in the horizontal direction or the vertical direction of the ultrasonic image without considering the directionality of the contour. It is being appreciated.

The present invention has been made in view of the above-mentioned conventional problems, and an object thereof is to suppress complication below a certain level by combining input operation and automatic extraction based on the peculiarity of an ultrasonic image. Another object of the present invention is to provide an ultrasonic image processing apparatus that can accurately extract a contour (boundary).

It is another object of the present invention to provide an ultrasonic image processing apparatus capable of setting an appropriate search path (and section) when extracting contour points in an interpolation section.

[0009]

In order to achieve the above object, the invention according to claim 1 automatically interpolates between a plurality of sample points input and designated on a boundary line of a tissue in an ultrasonic image. An ultrasonic image processing apparatus for setting the reference straight line calculating means for obtaining a reference straight line connecting both ends of the input designated sample points or the already obtained interpolation points as an interpolation section, and the reference straight line. A reference point calculating means for setting a reference point at a position separated by a predetermined distance from both ends thereof, a search straight line calculating means for obtaining a search straight line perpendicular to the reference straight line and passing through the reference points, and along the search straight line An interpolating point calculating means for obtaining the interpolating point by searching the boundary line, and repeatedly setting the interpolating point until a straight line distance connecting both ends of the interpolating section becomes a predetermined value or less. And

The invention according to claim 2 is characterized in that the search range on the search line is a fixed range centered on the reference point.

[0011]

According to the structure of the first aspect, the operator inputs and designates a plurality of sample points in the ultrasonic image captured by transmitting and receiving the ultrasonic waves. That is, a plurality of sample points are set at arbitrary intervals on the boundary line of the tissue by using, for example, a cursor. Then, the next automatic interpolation process is executed. First, a reference straight line connecting both ends of the interpolation section is obtained. Here, the interpolation section refers to between the sampling points designated by input or between the interpolation points already obtained. That is, this reference straight line means a vector extending from one sample point to the other sample point. Next, a reference point is set on the reference straight line. Here, the reference point is set at a position separated by a predetermined distance from both ends of the reference straight line. Then, a search straight line extending perpendicularly to the reference straight line from the reference point is obtained. A boundary line is searched for on the search line, and the found boundary position is used as an interpolation point.

Such processing is repeated until the straight line distances at both ends of the interpolation section become equal to or less than a predetermined value, and interpolation points are sequentially set between two sample points while the interpolation section is reduced. .

In the present invention, since the search line is set as a perpendicular line of the reference line connecting both ends of the interpolation section, the search line, that is, the search range can be appropriately set according to the positions of both ends of the interpolation section. It will be possible. That is, for example, when the interpolation section as a whole is inclined downward to the right, the reference line is also inclined downward to the right, and accordingly the search line is also inclined to the right. Therefore, the search can be performed from the direction as orthogonal as possible to the boundary line to be obtained, so that the boundary points, that is, the interpolation points can be extracted efficiently and quickly.

According to the second aspect of the invention, the search range is set on the search straight line centering on the reference point, and the boundary line is searched and extracted in the range.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of the present invention will be described below with reference to the drawings.

First, the principle of the present invention will be described with reference to FIGS. FIG. 2 shows a part of an ultrasonic tomographic image captured by transmitting and receiving ultrasonic waves in the ultrasonic diagnostic apparatus. That is, FIG. 2 shows a cross section 100 of an organ as a living tissue, and 102 in the figure is a boundary line with another tissue.

The present invention uses, for example, an organ 1 by a method combining the conventional artificial tracing method and the automatic tracing method.
The cross-sectional area of 00 is calculated. That is, in the present invention, first, on the boundary line (contour) 102 of the organ 100,
The operator sets a plurality of sample points S. That is,
When the automatic tracing method is completely applied, there is a possibility that the boundary line to be extracted may be lost due to the influence of noise or the like. Therefore, in the present invention, a plurality of sample points S are input and designated by the operator as representative values. Specifically, for example, the input is designated by clicking a specific coordinate in the display with a cursor linked to a mouse. In addition,
The number of sample points may be determined according to the complexity of the shape of the contour line to be extracted, and the sample points may be concentrated on the part having a complicated shape. In the example shown in FIG. 2, for example, the sample points S are preferably about 5 to 10 points.

After the sample points are designated and input as shown in FIG. 2, an interpolation process between the sample points is executed as shown in FIG. In FIG. 1, two adjacent lines are on the boundary line 102.
Two sample points S1 and S2 are shown. Then, in order to provide a plurality of interpolation points on the boundary line 102 for the curved interpolation section between the sample points S1 and S2, the following processing is performed.

First, a reference straight line M1 connecting between S1 and S2, which are both ends of the interpolation section, is set. Specifically, the inclination of the straight line and the start point / end point are calculated by coordinate calculation. Then, next, reference points P1 and P2 are set on the reference straight line M1 which is apart from each end by a constant distance K.
Here, the predetermined distance K may be determined by interpolation accuracy,
If K is increased, the interpolation processing time can be shortened, and if K is decreased, interpolation accuracy can be improved.

After the reference points P1 and P2 are set in this way, the reference points P1 are perpendicular to the reference straight line M1 and are respectively.
And search lines L1 and L2 passing through P2 and P2 are set. That is, the search straight lines L1 and L2 are lines perpendicular to the reference straight line M1, and the boundary line 102 is extracted along the search straight lines L1 and L2.

That is, the search lines L1 and L2 have a predetermined length centered on the reference points P1 and P2,
That is the search range. Therefore, the search extraction is performed from one end or the other end of the search line or from the center along the boundary line 102. Specifically, the boundary position of the living tissue is specified by using a conventionally known difference method or dispersion method.
In FIG. 1, the specified boundary positions, that is, the interpolation points are indicated by A and B, respectively.

After the two interpolation points are obtained in this way, the above-described interpolation processing of the interpolation points is repeatedly executed for the interpolation section between the two interpolation points A and B.

That is, the reference straight line M2 having both ends of the interpolation sections A and B is set, and the reference points P3 and P are set at a position separated from the both ends by a constant distance K as described above.
4 is set. Then, after the search lines L3 and L4 passing through P3 and P4 are set, the boundary line 10 described above is set.
2 extractions are performed. The boundary points specified by this, that is, the interpolation points are indicated by C and D in FIG.
It is the interpolation points E and F shown in FIG. 1 that the above is repeated once more and specified.

According to the interpolation processing according to the present invention as described above, the inclination of the reference straight line, and thus the inclination of the search straight line, can be set in accordance with the positional relationship between the coordinates of the both ends of the interpolation section. The boundary line 102 can be specified with. That is, in the present invention, the coordinates for the search can be dynamically changed based on the coordinate positional relationship of the section to be interpolated.

Next, the overall configuration of the ultrasonic image processing apparatus according to the present invention will be described with reference to FIG.

In FIG. 3, the ultrasonic diagnostic apparatus 10 forms an ultrasonic image such as a tomographic image by transmitting ultrasonic waves to a living body and receiving reflected waves from the inside of the living body. is there. Then, the image signal is supplied to the ultrasonic image processing apparatus 12. That is, the image signal is
First, after being converted into a digital signal in the A / D converter 14, it is sequentially written in the data memory 16. If the ultrasonic diagnostic apparatus 10 forms a two-dimensional tomographic image, the two-dimensional echo information is stored in the data memory 16
On the other hand, when the ultrasonic diagnostic apparatus 10 captures three-dimensional data, the three-dimensional echo information is stored in the data memory 16.

The input unit 20 connected to the bus 18 is composed of, for example, a plurality of dials, and the input of the sample points S and the designation of the cut surface are performed using the input unit 20.

As will be described later, the calculation unit 22 calculates the cross-sectional area and volume of the organ based on the ultrasonic image after the interpolation processing.

The frame memory 24 temporarily stores the ultrasonic image after the image processing, and the ultrasonic image stored therein is read out and displayed on the CRT 26.

On the other hand, the interpolation processing unit 2 connected to the bus 18
8 executes the interpolation processing shown in FIG. 1. As shown in FIG. 1, the interpolation points A to F are sequentially stored in the interpolation point memory 30. Further, the plurality of sample points S initially input and designated by the input unit 20 are stored in the sample point memory 32.

Next, the operation of this ultrasonic image processing apparatus will be described with reference to FIG.

FIG. 4 shows the interpolation process as a flowchart. That is, ultrasonic waves are transmitted and received by the ultrasonic diagnostic apparatus 10, and after the ultrasonic image is stored in the data memory 16, the steps shown in FIG. 4 are executed.

In S101, the operator uses the input unit 20 to input and designate a plurality of sample points S as shown in FIG. In this case, the larger the number of sample points S, the more the subsequent cross-sectional area calculation accuracy can be improved. However, according to the interpolation processing of the present embodiment, even with a small number of sampling points,
If the number is a certain number or more, it is possible to obtain an extremely accurate calculation result. The coordinates of the input sample points S are stored in the sample point memory 32, respectively.

In S102, the interpolation processing section 28 is activated to read the first two sample points S. That is,
As shown in FIG. 1, two adjacent sample points S1,
The coordinates of S2 will be read. And S10
In 3, it is determined whether or not the distance between the two sample points is K or more. Here, if the distance between the sample points is smaller than K, S115 is executed and it is determined whether or not the interpolation processing has been completed between all the sample points. on the other hand,
When it is determined that the distance between the two sample points is K or more, it is further determined in S104 whether the distance is 2K or more. If it is determined that the result is 2K or more, the steps S105 to S109 are performed, while if it is determined that the value is K or more but less than 2K, the steps S111 to S114 are performed. The process will be performed. The steps S111 to S114 basically perform substantially the same processes as the steps S105 to S108.

In S105, as shown in FIG. 1, the reference straight line M connecting the two sample points with the two sample points at both ends is calculated. Specifically, the inclination of the reference line, the start point, and the end point are calculated based on the coordinates of the two sample points.

Then, in S106, two reference points P are calculated on the reference straight line. Specifically, the reference point P is set at a position separated by a constant distance K from both ends.

Then, in S107, a search line L passing through the reference point P and perpendicular to the reference line M is calculated.
That is, two search straight lines L are calculated.

In S108, for example, the dispersion value is sequentially calculated from the one end based on the image data of the ultrasonic image, and when the dispersion value becomes large, the boundary point, that is, the interpolation point is specified. Various conventional methods can be applied to the extraction of the boundary line, and for example, a boundary value can be extracted by obtaining a differential value, or a certain kind of filter or the like can be used.

As described above, after the interpolation point is set for the first interpolation section, the above steps are repeatedly executed for the interpolation section having the two new interpolation points at both ends.

Finally, if the straight line distances at both ends of the interpolation section are smaller than K, S115 is executed. On the other hand, if it is more than K but smaller than 2K, S11 is executed.
Each process from 1 to S114 is executed. That is,
In S111, one reference straight line M is calculated, and then S1
In 12, the constant distance K from one end of the reference straight line M
One reference point P is set at the separated positions. Then, in S113, the reference straight line L passing through the one reference point P is calculated, and in S114, the boundary line is searched and extracted, and finally one interpolation point is specified.

In S115, it is determined whether or not the above-mentioned processing has been completed for all the sample points. If not, the processing is moved to the next adjacent section and S102.
The steps from are executed. On the other hand, if it is determined in S115 that the processing has been completed in all the sections, this interpolation processing ends and the next arithmetic processing is executed.

That is, as shown in FIG.
The cross-sectional area of the organ is calculated by. Alternatively, the volume of the organ is calculated by executing such interpolation processing in each cross section and executing, for example, triangulation processing between the cross sections.

According to the above-described interpolation processing, the boundary line can be extracted while dynamically changing the reference coordinates for the search according to the coordinate positional relationship of the interpolation section. Interpolation points can be set.
Note that the sample points and the interpolation points are connected to each other by a straight line, and the cross-sectional area and volume of the organ are calculated based on them.

[0044]

As described above, according to the present invention,
The slope of the entire interpolation section is indicated by a reference line, and the search line is set based on that reference line, so the boundary line should be detected from the appropriate direction and the boundary line should be extracted quickly and accurately. There is an effect that can be. Therefore,
According to the present invention, it is necessary to input a plurality of sample points,
As a whole, there is an effect that the complexity can be avoided as compared with the conventional case and a certain interpolation accuracy can be obtained.

The sample points and the interpolation points are connected to each other by a straight line, and the cross-sectional area and the volume of the organ are calculated based on them.

[Brief description of drawings]

FIG. 1 is an explanatory diagram showing an interpolation process between sample points according to the present invention.

FIG. 2 is an explanatory diagram showing input designation of sample points.

FIG. 3 is a block diagram showing the overall configuration of an ultrasonic image processing apparatus according to the present invention.

FIG. 4 is a flowchart showing an interpolation process according to the present invention.

[Explanation of symbols]

 102 boundary line S sample point M reference straight line P reference point L search straight line A to F interpolation point (boundary point)

Claims (2)

[Claims] 1. An ultrasonic image processing apparatus for automatically setting an interpolation point between a plurality of sample points input and designated on a boundary line of a tissue in an ultrasonic image, wherein the sample point between the sample points input is designated. Alternatively, a reference straight line calculating means for obtaining a reference straight line connecting both ends of the already obtained interpolation points as an interpolation section, and a reference point calculation for setting a reference point at a position on the reference straight line separated from the both ends by a predetermined distance. Means, a search straight line calculation means for obtaining a search straight line perpendicular to the reference straight line and passing through the reference point, and an interpolation point calculation means for searching the boundary line along the search straight line to obtain the interpolation point The ultrasonic image processing apparatus, wherein the interpolation point is repeatedly set until a straight line distance connecting both ends of the interpolation section becomes equal to or less than a predetermined value. 2. The ultrasonic image processing apparatus according to claim 1, wherein the search range on the search line is a fixed range centered on the reference point.