JP3217025B2 - Ultrasonic diagnostic apparatus and ultrasonic image processing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an ultrasonic diagnostic apparatus and an image processing method, and more particularly to an image processing for extracting a contour of a predetermined organ and displaying the contour.

[0002]

2. Description of the Related Art In order to diagnose the function of the heart, a contour of a heart chamber (for example, a left ventricle) is automatically extracted on an ultrasonic image such as a B-mode image, and the contour is alone or combined with a tomographic image. Ultrasound diagnostic apparatuses that display the images are being put to practical use. In such a device, when extracting the contour,
For example, a binarization process using a threshold (judgment level) for each pixel of the ultrasonic image is performed. In this case, a user sets a region of interest (ROI) having, for example, an elliptical shape so as to include the outline as a processing range of the outline extraction.

For example, when a tomographic image of a heart chamber is formed by using electronic sector scanning or the like, an echo is weak at a heart wall portion where an ultrasonic beam and a muscle fiber run in parallel, and the portion is displayed with low brightness. Further, there is a similar problem in a part located on the back side of the high reflector when viewed from the ultrasonic probe. That is, the same heart wall may have low brightness or high brightness depending on its position.

If contour extraction processing for extracting the contour of the heart chamber (contour of the heart wall) is performed on such an ultrasonic image, it may not be possible to accurately extract the contour of a low-luminance part. In particular, a tomographic image of a person with a lot of body fat tends to be unclear, and the above problem becomes remarkable.

Conventionally, there has been known an ultrasonic diagnostic apparatus having a function of performing gain adjustment for each ultrasonic beam. According to such an apparatus, for example, an upper right part of a left ventricle in a sector tomographic image in which the left ventricle is displayed at the center. It is also possible to compensate for the luminance by increasing the sensitivity for each of the low-luminance parts at the upper left and the lower part. But,
Even if there is a portion on the ultrasonic beam where the sensitivity is not desired to be raised, there is a problem that the sensitivity of the portion is also increased uniformly.

Therefore, there is a demand for a method for accurately extracting the contour of a low-brightness part in a heart wall part.

The present invention has been made in view of the above-mentioned conventional problems, and an object of the present invention is to accurately extract a contour of a predetermined organ on an ultrasonic image.

[0008]

To achieve the above object, the present invention relates to an ultrasonic diagnostic apparatus including a contour extracting means for extracting a contour of a predetermined tissue from a two-dimensional ultrasonic image. A region of interest setting means for setting a region of interest on an ultrasound image, and a weighting pattern setting for setting a two-dimensional weighting pattern for emphasizing the contour with reference to the shape of the region of interest. Means, and weighting means for weighting the ultrasound image according to the weighting pattern, wherein the contour extraction means performs contour extraction based on the weighted ultrasound image. .

According to the above configuration, a weighting pattern is set based on a region of interest (ROI) set by a user or automatically set, and weighting of an ultrasonic image is performed using the weighting pattern. . In this case, for example, addition of an offset value to each pixel value according to the weighting pattern, gain control of a received signal according to the weighting pattern, or variable control of a contour extraction determination criterion according to the weighting pattern is performed. . According to such weighting, a large weight is given to a portion having a high contour existence probability based on the ROI, and the contour determination accuracy can be improved.

In a preferred aspect of the present invention, weighting is performed by adding an offset value based on the weighting pattern to a pixel value of each pixel forming an ultrasonic image. Alternatively, the weighting unit controls the amplification degree of a reception signal for forming the ultrasonic image, and thereby weights the pixel value of each pixel constituting the ultrasonic image based on the weighting pattern. I do. Alternatively, the weighting means performs weighting by variably setting a determination level to be compared with a pixel value of each pixel of the ultrasonic image in the contour extraction according to the weighting pattern.

According to the above-described addition of the offset values, weighting can be performed by a simple method called image addition.
The above-described control of the amplification degree of the received signal has an influence on the displayed tomographic image, but can more accurately weight the received signal. According to the above variable control of the determination level,
There is an advantage that processing such as image addition is not required, and weighting can be realized with a simpler configuration. In any case, according to the present invention, since the weighting pattern is set based on the region of interest, it is possible to determine with high accuracy even a contour portion where the echo intensity is slightly reduced, and the contour portion is hardly affected by noise. Extraction can be realized.

In a preferred aspect of the present invention, the weighting pattern setting means sets, as the weighting pattern, a pattern in which the weighting value decreases continuously or stepwise from the outer edge of the region of interest or the vicinity thereof to the inside thereof. I do.
According to the continuous setting, weighting can be performed more finely, and according to the stepwise setting, the load related to weighting can be reduced and the processing time can be shortened.

In a preferred aspect of the present invention, the weighting pattern setting means is for generating a similar shape of an outer edge of the region of interest, and generates a plurality of similar shapes stepwise inside the region of interest. A similarity generating unit; and a unit for assigning a weight value to each of the band-shaped regions divided by the outer edge of the region of interest and the plurality of similarities. The weighting pattern depends on the shape of the region of interest. Therefore, the weighting pattern is quickly set using a similar shape of the region of interest.

In the weighting, an azimuth from a reference point in the region of interest may be further considered. That is, if the azimuth at which the echo intensity becomes weaker is known in advance, the azimuth (part) is weighted more.

In order to achieve the above object, a method according to the present invention is directed to an image processing method for extracting a contour of a predetermined tissue from a two-dimensional ultrasonic image. A step of setting a region of interest by a user, a step of setting a two-dimensional weighting pattern in which a weight value is gradually reduced from at least the outer edge to the inner portion based on the outer edge of the region of interest, and using the weighting pattern Performing a weighting process for the pixel value, a variable control of the gain for the received signal, or a variable control of the contour discrimination criterion in the contour extraction process.

The weighting based on the outer edge of the region of interest is different from the weighting based on the gain adjustment according to the depth and the beam direction as in the conventional apparatus, and the method according to the present invention performs the contour discrimination more than the conventional method. It can be performed accurately, and as a result, for example, the calculation of the left ventricular area and the stroke volume can be performed accurately, and thus more useful information can be provided in diagnosing a disease.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows a preferred embodiment of an ultrasonic diagnostic apparatus according to the present invention, and FIG. 1 is a block diagram showing the overall configuration.

The probe 10 is, for example, an ultrasonic probe which is used in contact with a body surface to transmit and receive ultrasonic waves. The probe 10 has a built-in array transducer composed of a plurality of transducers, and an ultrasonic beam is scanned by electronically scanning the array transducer. As a result, a two-dimensional data acquisition area is formed. You. Display 1 described later
8, a tomographic image of the subject corresponding to the data capturing area is displayed.

The transmission / reception unit 12 is connected to the probe 10,
A transmission signal is supplied from the transmission / reception unit 12 to the probe 10. Further, a reception signal from the probe 10 is input to the transmission / reception unit 12, and the reception signal is subjected to processing such as amplification and detection.

The tomographic image forming section 14 is a means for forming a two-dimensional tomographic image (B-mode image) based on a received signal, that is, echo data. Of course, the present invention can be applied to a case where a two-dimensional Doppler image is formed.
The luminance (pixel value) of the image in that case corresponds to the size of Doppler information.

The display control unit 16 includes a tomographic image forming unit 14
Is input, and the display control unit 1
6 synthesizes another image with the tomographic image and outputs the synthesized image to the display 18. The display 18 displays the composite image.

The ultrasonic diagnostic apparatus shown in FIG. 1 has a calculation function for, for example, the area of the left ventricle and the amount of stroke, and extracts the contour of the heart chamber prior to such calculation. The outline extraction of such a tissue is executed in the outline extraction unit 26 described later, and various operations based on the extracted outline are executed in the operation unit 28 described later.

In the present embodiment, an ROI (region of interest) setting unit 20 is provided to accurately perform contour extraction. The ROI setting unit 20 is configured by, for example, a keyboard, a trackball, or the like, and is a device that allows a user to set an ROI on a tomographic image. The ROI is set outside the outline of the left ventricle, for example. In this case, the region of interest may be set to a free shape, or may be set as a basic figure such as a circle or an ellipse. Of course, the present invention
The present invention is also applicable to a case where an ROI is automatically set according to various methods.

In this embodiment, the ROI setting unit 20
A weighting pattern is generated in the weighting pattern generation unit 22 in accordance with the ROI set by (1). The weighting pattern is a pattern for weighting each pixel (echo data) of the ultrasonic image in order to perform contour extraction more accurately. In the embodiment shown in FIG. 1, an offset pattern is generated as the weighting pattern, and the offset pattern is added to the tomographic image by the offset adding unit 24. Although the specific contents thereof will be described later, the ultrasonic image to which the offset pattern has been added is sent to the contour extracting unit 26, and the contour extracting unit 26 outputs each of the pixels of the ultrasonic image to which the offset pattern is added. Is compared with a predetermined judgment level, and the contour of a pixel exceeding the judgment level is judged. In this case, various methods can be applied as the outline determination method. For example, a reference line is set radially from a reference point corresponding to the center of the ROI, and the pixel value of each pixel is set along the reference line from the reference point. By comparing with the determination level, a portion where the pixel value exceeds the determination level can be determined as the contour. The contour image extracted by the contour extraction unit 26 is output to the calculation unit 28,
The calculation unit 28 calculates, for example, the area of the left ventricle based on the contour image, calculates the stroke volume based on the area, and the like.
The calculation result is output to the display control unit 16, and finally,
For example, it is displayed on the display 18 as a numerical value. The image of the ROI set by the ROI setting unit 20 is also output to the display control unit 16, and the RI image is displayed on the tomographic image.
A line representing the OI is synthesized and displayed. Incidentally, the outline image output from the outline extraction unit 26 is also output to the display control unit 16, and the outline image can be synthesized and displayed on the tomographic image.

FIGS. 2 and 3 show another method for weighting. The same components as those shown in FIG. 1 are denoted by the same reference numerals, and description thereof will be omitted.

In the configuration example shown in FIG. 2, a weighting pattern is input to the contour extraction unit 26A, and the weighting pattern is used as a judgment level pattern. That is, the determination level is variably set according to the weighting pattern, and as a result, each pixel is weighted. In the configuration example shown in FIG. 3, the weighting pattern is sent to the transmission / reception unit 12A, and the weighting pattern is used as a variable gain function. That is, in the gain control of the received signal in the transmission / reception unit 12A, gain control is performed according to the weighting pattern. Weighting can also be performed on each pixel by such control. However, in the configuration examples shown in FIGS. 1 and 2, the tomographic image itself is created by a method similar to the conventional method, and as a result, the tomographic image can be displayed on the display 18 as it is. However, in the configuration example shown in FIG. The weighting is reflected on the image itself. Therefore,
It is desirable to adopt the configuration example shown in FIG. 1 or FIG.

Next, the weighting method according to the present invention will be described in detail with reference to FIG.

FIG. 4A shows a tomographic image 100. For example, a cross section of the left ventricle is displayed in the tomographic image 100, and the contour of the left ventricle is indicated by reference numeral 102.
(A) clearly shows the contour 102, but in an actual ultrasonic image, such a contour 102 is generally unclear, so that the weighting according to the present invention is effective.

As described above, the roughly recognized outline 1
ROI is set to include it outside 02. In the example shown in (A), the ROI is set to an elliptical shape. Of course, the present invention is not limited to such a shape. This ROI defines a recognition range that is a processing range for contour extraction, and also functions as a reference shape of a weighting pattern. In the present embodiment, one or more similar figures L2, L3, L4 are placed inside the ROI.
Is set, and a weighting value is individually assigned to a band-like region between these lines.

FIG. 4B shows the relationship between the pixel values on the line AA 'shown in FIG. In the configuration example shown in FIG. 1, an offset is added to each pixel value. Here, as the offset amount, S1, S2, S3, 0 are set stepwise from the ROI to the inside. This is added to the pixel value of each pixel, and the result of the addition is compared with a determination level K serving as a determination reference in contour extraction. Then, the comparison between the pixel value and the determination level K is not performed outward from the center point of the ROI, and the contour is determined when the determination level K exceeds the pixel value. Therefore, according to the addition of the offset as shown in (B), there is an advantage that a higher weight is applied to a place where the existence probability of the contour is high, so that the accuracy of the contour determination can be improved.

FIG. 4C shows a setting pattern in which the determination level K is varied according to the weighting pattern. As shown in the figure, the determination levels K4 and K4 are stepwisely determined for each band-shaped region from the ROI to the inside.
K3, K2, and K1 are set. With such a level setting, a portion having a high probability of existence of a contour can be weighted more.

In the configuration example shown in FIG. 3, for example, a gain offset corresponding to the offset shown in FIG. 4B is performed. Incidentally, although a stepwise pattern is shown as the weighting pattern in FIG. 4, the weighting process may be performed using a pattern in which the weighting value changes continuously.

FIG. 5 shows such a continuous weighting pattern, as shown in FIG.
When viewed on the line AA 'shown in FIG. 5A, a weighting pattern having a local maximum at or near the outer edge of the ROI and having a smaller weighting value inward of the ROI is used as shown in FIG. May be. FIG. 6 illustrates another configuration example of the weighting pattern. In the example illustrated in FIG. 6, an example in which a weight value is set for each pixel is illustrated. Of course, any weighting pattern is based on the outer edge of the ROI.

FIG. 7 shows an example of the configuration of the weighting pattern generator 22. For example, when the ROI is set by the ROI setting unit 20, the weighting pattern generation unit 2
2 is a stepwise similar shape L based on the outer edge of the ROI.
2, L3 and L4 are generated. Then, a weighting pattern is finally generated by synthesizing them and setting weighting values S1 to S4 for a band-like region surrounded by each line. Of course, a weighting pattern may be generated using a method other than such a method. According to the method shown in FIG. 7, there is an advantage that a weighting pattern can be defined by a simple method based on the outer edge of the ROI. In particular, even when the ROI is set as a complicated figure, the method shown in FIG. 7 can be easily applied.

[0036]

As described above, according to the present invention,
The contour of a predetermined organ can be accurately extracted on an ultrasonic image.

[Brief description of the drawings]

FIG. 1 is a diagram showing a preferred embodiment of an ultrasonic diagnostic apparatus according to the present invention.

FIG. 2 is a diagram showing a preferred embodiment of an ultrasonic diagnostic apparatus according to the present invention.

FIG. 3 is a diagram showing a preferred embodiment of the ultrasonic diagnostic apparatus according to the present invention.

FIG. 4 is an explanatory diagram showing a weighting process according to the present invention.

FIG. 5 is a diagram illustrating an example of a weighting pattern.

FIG. 6 is a diagram illustrating an example of a weighting pattern.

FIG. 7 is a diagram illustrating a configuration example of a weighting pattern.

[Explanation of symbols]

Reference Signs List 10 probe, 12 transmitting / receiving unit, 14 tomographic image forming unit, 16 display control unit, 18 display, 20 ROI
(Region of interest) setting unit, 22 weighting pattern generation unit,
26 contour extraction unit, 28 operation unit.

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-8-206117 (JP, A) JP-A-7-255703 (JP, A) JP-A 8-96143 (JP, A) JP-A 9-206 285465 (JP, A) JP-A-4-122359 (JP, A) JP-A-6-292667 (JP, A) JP-A-11-164834 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) A61B 8/00-8/15 G06T 1/00

Claims (7)

(57) [Claims] 1. An ultrasonic diagnostic apparatus including a contour extracting means for extracting a contour of a predetermined tissue from a two-dimensional ultrasonic image, wherein: a region for interest is set on the ultrasonic image according to a shape of the contour. Region of interest setting means; weighting pattern setting means for setting a two-dimensional weighting pattern for emphasizing the contour based on the shape of the region of interest; and weighting the ultrasound image according to the weighting pattern. An ultrasonic diagnostic apparatus, comprising: weighting means, wherein the contour extracting means performs contour extraction based on the weighted ultrasonic image. 2. The apparatus according to claim 1, wherein the weighting means performs weighting by adding an offset value based on the weighting pattern to a pixel value of each pixel forming the ultrasound image. Ultrasound diagnostic device characterized by the following. 3. The apparatus according to claim 1, wherein the weighting means controls an amplification degree of a reception signal for forming the ultrasonic image, and thereby each pixel constituting the ultrasonic image as a result. An ultrasonic diagnostic apparatus, wherein weighting is performed on pixel values based on the weighting pattern. 4. The apparatus according to claim 1, wherein the weighting unit performs weighting by variably setting a determination level to be compared with a pixel value of each pixel of the ultrasonic image in the contour extraction according to the weighting pattern. An ultrasonic diagnostic apparatus characterized by the above-mentioned. 5. The apparatus according to claim 1, wherein the weighting pattern setting means includes a pattern in which a weighting value decreases continuously or stepwise from the outer edge of the region of interest or its vicinity to the inside thereof as the weighting pattern. An ultrasonic diagnostic apparatus characterized by setting: 6. The apparatus according to claim 1, wherein the weighting pattern setting unit generates a similar shape of an outer edge of the region of interest,
A similarity generating means for stepwise generating a plurality of similarities inside the region of interest, and a means for assigning a weight value to each of the band-shaped regions divided by the outer edge of the region of interest and the plurality of similarities. An ultrasonic diagnostic apparatus comprising: 7. An image processing method for extracting a contour of a predetermined tissue from a two-dimensional ultrasonic image, wherein a step of user setting a region of interest on the ultrasonic image so as to include the contour, A step of setting a two-dimensional weighting pattern in which the weighting value is gradually reduced from at least the outer edge to the inner part based on the outer edge of the region; and, using the weighting pattern, weighting processing for a pixel value, gain for a received signal. Performing one of the variable control of (a) and the variable control of the contour determination criterion in the contour extraction processing.