JP5406484B2 - Ultrasonic data processor - Google Patents

Description

  The present invention relates to an ultrasonic data processing apparatus, and more particularly to a technique for tracing a contour of a target tissue.

  An ultrasonic data processing apparatus such as an ultrasonic diagnostic apparatus has a function of measuring the area and volume of a target tissue based on ultrasonic data. The target tissue is, for example, the left ventricle, placenta, tumor, and the like. The area calculation is executed based on two-dimensional ultrasonic data (tomographic data), and the volume calculation is usually executed based on three-dimensional ultrasonic data (volume data). In order to measure the area and volume of the target tissue, it is necessary to trace the contour of the target tissue. In the area calculation, the outer shape of the target tissue is specified, and in the volume calculation, the surface of the target tissue is specified.

  Manual tracing and automatic tracing are known as trace methods. According to manual tracing, although it is possible to trace relatively accurately even for a structural part where there is not much difference in luminance based on visual judgment based on empirical rules, it can be pointed out that it is cumbersome and burdensome. It is also quite difficult to trace the entire target tissue surface. According to the automatic trace, there is no burden on the user, but in many cases, it is not possible to accurately trace a structural portion with a small luminance difference. Patent Documents 1 and 2 describe tissue tracing or tissue extraction.

JP 2002-291750 A JP-A-7-184892

  As described above, manual tracing and automatic tracing have their merits and demerits, but in that case, if the user decides which method to select, it must be an intuitive decision. In some cases, the advantages of each method cannot be used effectively. Therefore, there is a demand for obtaining or providing an objective judgment index.

  An object of the present invention is to obtain an objective index when selecting manual tracing and automatic tracing.

  The present invention relates to an ultrasonic data processing apparatus for processing ultrasonic data acquired from a three-dimensional living body space, in a three-dimensional data space corresponding to the three-dimensional living body space, outside the target tissue. Setting means for setting a three-dimensional region of interest that wraps the image, and determination means for determining automatic trace processing or manual trace processing as a trace method for the ultrasonic data based on a luminance distribution in the three-dimensional region of interest The determination means determines an evaluation value calculation means for calculating an evaluation value indicating the bimodality of the luminance distribution, and determines automatic trace processing or manual trace processing based on the magnitude of the evaluation value. And a tracing method determining means.

  According to the above configuration, in the three-dimensional data space, a three-dimensional region of interest that encloses the target tissue (target tissue data) (desirably including the surrounding tissue) is set. The setting is preferably performed by the user while viewing a three-dimensional image or one or a plurality of tomographic images. When a three-dimensional region of interest is set, the luminance of each voxel existing inside the region of interest is referred to, thereby forming a luminance distribution. The luminance distribution reflects the tissue structure or tissue mode in the three-dimensional region of interest. Therefore, if the form of the luminance distribution is evaluated, specifically, the degree of bimodality (roughly not a form that rises like a mountain, but a shape that is separated into two peaks with a valley portion in between. It is possible to determine the trace method from the magnitude of the evaluation value. The luminance distribution may be formed not on the entire three-dimensional region of interest set outside the target tissue but on a portion including the surface of the target tissue. In that case, a second three-dimensional region of interest may be set inside the target tissue, and a hollow portion sandwiched between the two three-dimensional regions of interest may be set as the calculation target of the luminance distribution. Note that even if a part of the target tissue protrudes from the three-dimensional region of interest set outside the target tissue, if the majority of the surface of the target tissue belongs to the three-dimensional region of interest, a trace method is appropriately applied to some extent. Can be determined. When a three-dimensional region of interest is set outside the target tissue, it is desirable to set the three-dimensional region of interest with a certain margin between the target tissue. This is because it is considered that the discriminability between the target tissue and the others can be more accurately determined if the surrounding tissue of the target region is reflected to some extent in the luminance distribution.

  Preferably, the automatic trace processing is determined when the luminance distribution has a shape of two peaks as a whole and the evaluation value is larger than a predetermined value, and the luminance distribution is a shape of a single mountain as a whole. And the manual trace processing is determined when the separation evaluation value is smaller than a predetermined value.

  Preferably, the evaluation value calculating means changes the threshold value with respect to the luminance distribution in the three-dimensional region of interest, while varying the degree of variation of the portion on the higher luminance side than the threshold value and the portion on the lower luminance side of the threshold value. Ratio calculating means for calculating a ratio to the degree of variation of the difference, and maximum ratio specifying means for specifying a maximum ratio as the evaluation value in a series of ratios obtained by changing the threshold value. For example, a technique used in discriminant analysis may be used. Various statistical processing methods can be used.

  Preferably, the setting unit includes a unit that generates a plurality of tomographic images representing a plurality of cross-sections with respect to the target tissue based on the ultrasound data, and the three-dimensional image on the plurality of tomographic images. Receiving a user designation of a three-dimensional spherical body as the region of interest.

  As described above, according to the present invention, the advantages of both manual tracing and automatic tracing can be achieved. In particular, an objective index for selecting them can be obtained.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of the invention will be described with reference to the drawings.

  FIG. 1 shows a preferred embodiment of an ultrasonic data processing apparatus according to the present invention, and FIG. 1 is a block diagram showing the overall configuration thereof. This ultrasonic data processing apparatus is constituted by an ultrasonic diagnostic apparatus in the present embodiment. However, all or part of the configuration shown in FIG. 1 may be configured by an information processing device (computer).

  The 3D probe 10 is a transducer that transmits and receives ultrasonic waves to and from a three-dimensional space in a living body. In this embodiment, the 3D probe 10 has a 2D array transducer. The 2D array transducer is constituted by a plurality of vibration elements arranged two-dimensionally. Thereby, an ultrasonic beam is formed, and the ultrasonic beam is electronically two-dimensionally scanned. As a result, a three-dimensional data capturing space is formed. As electronic scanning methods, electronic sector scanning, electronic linear scanning, and the like are known. Instead of the 2D array transducer, a 1D array transducer and a mechanism for mechanically scanning the 1D array transducer may be provided.

  The transmission / reception unit 12 functions as a transmission beamformer and a reception beamformer. That is, at the time of transmission, a plurality of transmission signals are supplied from the transmission / reception unit 12 to the 2D array transducer. As a result, a transmission beam is formed. At the time of reception, the reflected wave from the living body is received by the 2D array transducer, whereby a plurality of received signals are output from the 2D array transducer to the transmission / reception unit 12. The transmission / reception unit 12 performs phasing addition processing on a plurality of reception signals, and thereby outputs a reception signal (beam data) after phasing addition. The beam data is stored in the 3D memory 14.

  The 3D memory 14 has a three-dimensional data space corresponding to the three-dimensional space in the living body, and a plurality of input beam data are stored in the 3D memory 14. Coordinate conversion from the transmission / reception coordinate system to the data coordinate system is executed when the data is stored or when data is read. The beam data is composed of a plurality of voxel data arranged along the ultrasonic beam. In the present embodiment, each voxel data is mapped to an address corresponding to the voxel data at the time of writing to the 3D memory 14. Prior to writing to each voxel data in the 3D memory 14, signal processing is executed as necessary. The signal processing may include, for example, binarization processing.

  The data processing unit 16 is a module that executes various types of data processing based on a plurality of beam data (volume data) stored in the 3D memory 14. In this embodiment, the data processing unit 16 includes an image forming unit 40, an automatic tracing unit 42, and a volume / area calculating unit 44. The image forming unit 40 is a unit that applies a process such as a volume rendering method based on volume data, thereby forming a three-dimensional ultrasonic image. The image forming unit 40 also has a function of forming three tomographic images corresponding to three cut planes set in a three-dimensional space as so-called triplane images. Further, the image forming unit 40 has a function of forming an arbitrary tomographic image corresponding to a cutting plane arbitrarily set by the user. Those image data are output to the display unit 20 via the display processing unit 18.

  The automatic trace unit 42 is a unit that executes auto-trace processing for slice planes for which automatic tracing is determined when performing trace processing on a plurality of slice planes set in a three-dimensional space. Prior to the trace processing, binarization processing for volume data or slice data is executed. The auto trace process can be realized based on a so-called edge detection method or the like. Various methods have been proposed as the auto-trace method. On the other hand, in the present embodiment, manual trace processing can also be performed on each slice plane. The manual trace process manually traces the contour of the target tissue on the tomographic image using a pointing device on the operation panel 38 described later. The coordinate data is given to the data processing unit 16, and the data processing unit 16 executes measurement processing described below using the coordinate data. Automatic trace processing or manual trace processing may be set uniformly for target tissues that exist three-dimensionally, or automatic trace processing or manual trace processing is switched and applied to each slice plane or tomographic plane. Also good. Further, the selection of the tracing method may be performed in units of ultrasonic diagnosis, that is, examination.

  The volume / area calculation unit 44 has a function of calculating the area of the target tissue based on the result of the automatic trace process and / or the result of the manual trace process, and calculating the volume. For example, if the target tissue is the left ventricle, loop-shaped trace lines are formed on a plurality of slice planes crossing the left ventricle, and the area on each slice plane is calculated, and then a certain thickness is given to them. The volume of the left ventricle is calculated by adding while giving. It is also possible to measure a fetus or placenta. The calculation result of the volume / area calculation unit 44 is sent to the display unit 20 via the display processing unit 18, and the calculation result is displayed on the screen as a numerical value or as a graph. The trace result may be displayed on the screen of the display unit 20. In addition, on the screen of the display unit 20, a plurality of tomographic images and a graphic representing the ROI are displayed when setting an ROI (three-dimensional region of interest) described later.

  Next, the trace method determination unit 22 will be described. In this embodiment, the trace method determination unit 22 includes a histogram creation unit 24, a separation evaluation value calculation unit 33, and a determination unit 34.

  The histogram creation unit 24 is a unit that creates a histogram (luminance distribution) based on a plurality of voxel data existing inside the ROI when the ROI is set by the user outside the target tissue. The separation evaluation value calculation unit 33 calculates the separation evaluation value by performing the form analysis of the histogram created as described above, and the separation evaluation value is a value indicating the degree of bimodality in the histogram. . A specific example of the arithmetic expression will be described later. The determination unit 34 is a module that determines the trace method by comparing the separation evaluation value obtained as described above with a predetermined value α. Specifically, if the separation evaluation value is larger than the predetermined value α, a tendency that the histogram has a two-peak shape is recognized, so that automatic tracing is determined as the tracing method, while the separation evaluation value is larger than the predetermined value α. If it is smaller, the histogram tends to have one mountain shape, so that manual tracing is determined as the tracing method. Such a determination result is passed to the data processing unit 16.

  The data processing unit 16 determines whether to execute automatic tracing for a target slice plane or to accept manual trace input according to the determination result of the determination unit. When the determination result of the determination unit 34 indicates a manual trace, the data processing unit 16 displays on the display screen a prompt for manual tracing. In response to the display, the user performs manual tracing on the displayed target image. The data processing unit 16 recognizes the coordinate data input at that time. However, the user's will is respected at this stage, and the actual trace may be executed according to the method selected by the user. In that case, it is only necessary to provide information on a determination method to be recommended to the user.

  The control unit 36 includes a CPU and an operation program, and performs operation control of each configuration illustrated in FIG. An operation panel 38 is connected to the control unit 36. The operation panel 38 is composed of a keyboard, a trackball, and the like. Using the operation panel 38, the user can set the ROI, perform manual tracing, and can further specify the threshold value α to be input to the determination unit 34 as necessary.

  Next, the tracing method determination method according to the present embodiment will be described in detail with reference to FIGS.

  FIG. 2 shows the setting of the ROI. Reference numeral 100 represents a display screen. The display screen 100 includes triplane images 102, 104, and 106 and a three-dimensional image or wire frame image 108. The triplane images 102, 104, and 106 are three tomographic images corresponding to three cut planes that are orthogonal to each other. On each image, lines 112, 113, and 114 representing the ROI are represented. The ROI has an elliptical sphere shape in this embodiment, the length of the major axis and the length of the minor axis can be arbitrarily set by the user, and the center coordinates are also arbitrarily set by the user. can do. Reference numeral 110 represents a target tissue. It is desirable that the three-dimensional region of interest is set so that the target tissue 110 is completely wrapped and a part of the background is included.

FIG. 3 shows a histogram 116 representing the luminance distribution for a plurality of voxels present in the region of interest. The horizontal axis of the graph shown in FIG. 3 represents luminance, and the vertical axis represents the number of pixels at each luminance, that is, the frequency. In the present embodiment, the separation evaluation value is calculated by statistically analyzing the form of the histogram. The specific contents will be described with reference to FIGS. FIG. 4 shows a histogram 120, which is generally composed of two peaks, which are broadly represented as class 1 and class 2. These two peaks are connected to each other, and a valley exists between the two peaks. While changing the threshold value t, the calculation of the dispersion ratio, which will be described later, is repeated, and the separation evaluation value F is regarded as the maximum value among the series of dispersion ratios obtained as a result. The maximum brightness is indicated by t _max in FIG. In this embodiment, if the separation evaluation value F is equal to or greater than the predetermined value α (see reference numeral 124), automatic tracing is determined (see reference numeral 126).

  On the other hand, a histogram 128 is shown in FIG. The histogram as a whole has one mountain shape, and is divided into two classes, namely, class 1 and class 2 for convenience when viewed from a certain threshold value. The separation evaluation value F is obtained as the maximum value among the results by calculating the dispersion ratio described later while changing the threshold value for the histogram. If the separation evaluation value F is equal to or less than the predetermined value α (see reference numeral 130), manual tracing is determined (see reference numeral 132). As described above, an advantage is obtained in that the determination can be performed accurately by determining the trace method using the separation evaluation value F reflecting the form of the histogram as a determination index.

Next, a method for calculating the separation evaluation value will be described. The separation evaluation value is an evaluation value indicating the degree of bimodality regarding the luminance distribution. Therefore, as long as such an evaluation value is required, several known methods can be used. In this embodiment, a discriminant analysis method is used. More specifically, in this method, when the luminance distribution, that is, the histogram is separated into two classes with a certain threshold t, the maximum value of the variance ratio of the inter-class variance and the intra-class variance is determined as the separation degree (separation evaluation value). It is what. The threshold at which the maximum value occurs is t _max .

The following equation (1) represents the interclass variance σ _B ² (t) at the threshold value t, and the following equation (2) represents the intraclass variance σ _I ² (t) at the threshold value t.

Here, the luminance range of the pixel is 0 to D, and when each pixel is binarized by the threshold value t, the average luminance value of the pixel group within the luminance range of 0 to t−1 is defined as f ₀ (bar). , The average luminance value of the pixels in the range from t to D is f ₁ (bar), the average luminance value of all the pixels is f (bar), and the number of pixels having luminance k is n _k. Yes. The dispersion ratio is defined by σ _B ² (t) / σ _I ² (t), and the maximum value F ₀ (t) is the separation evaluation value. This is shown by the following equation (3).

  An automatic trace process or a manual trace process is selected by comparing the maximum value of the above-mentioned degree of separation with a predetermined value. Specifically, if the separation evaluation value is large, the degree of bimodality is high, and it can be said that it is easy to automatically perform tissue discrimination. Therefore, automatic trace processing is determined. Since it is difficult to perform tissue discrimination (discrimination between the target tissue and the background tissue), manual trace processing is determined. It is not restricted to the said method, You may make it utilize the other calculated value which shows a bimodal degree as a separated evaluation value. In any case, appropriate determination of the trace method can be performed from the form analysis of the histogram reflecting the luminance variation relationship between the target tissue and its surroundings.

It is a block diagram which shows the whole structure of the apparatus which concerns on this invention. It is a figure which shows a tri-plane image. It is a figure which shows a histogram. It is a figure which shows the specific example in which an automatic trace is determined. It is a figure which shows the specific example in which a manual trace is determined.

Explanation of symbols

  10 3D probe, 12 transmission / reception unit, 14 3D memory, 16 data processing unit, 22 trace method determination unit, 24 histogram creation unit, 33 separation evaluation value calculation unit, 34 determination unit.

Claims (4)

In an ultrasonic data processing apparatus that processes ultrasonic data acquired from a three-dimensional living space,
In a three-dimensional data space corresponding to the three-dimensional in-vivo space, setting means for setting a three-dimensional region of interest that wraps it outside the target tissue;
Based on the luminance distribution in the three-dimensional region of interest, as a trace method for the ultrasound data, a determination unit that determines automatic trace processing or manual trace processing,
Including
The determination means includes
Evaluation value calculation means for calculating an evaluation value indicating the bimodality of the luminance distribution;
Trace method determination means for determining automatic trace processing or manual trace processing based on the size of the evaluation value;
An ultrasonic data processing apparatus comprising: The apparatus of claim 1.
The automatic trace processing is determined when the luminance distribution has a shape of two peaks as a whole and the evaluation value is larger than a predetermined value,
The manual trace processing is determined when the luminance distribution has a single mountain shape as a whole and the evaluation value is smaller than a predetermined value.
An ultrasonic data processing apparatus. The apparatus of claim 1.
The evaluation value calculation means includes:
A ratio for calculating a ratio between a variation degree of a portion on the higher luminance side than the threshold value and a variation degree of a portion on the lower luminance side than the threshold value while changing the threshold value with respect to the luminance distribution in the three-dimensional region of interest. Computing means;
Maximum ratio specifying means for specifying a maximum ratio as the evaluation value in a series of ratios obtained by changing the threshold;
An ultrasonic data processing apparatus comprising: The device according to any one of claims 1 to 3,
The setting means includes
Means for generating a plurality of tomographic images representing a plurality of cross sections in an intersecting relationship with respect to the target tissue based on the ultrasound data;
On the plurality of tomographic images, means for accepting a user designation of a three-dimensional spherical body as the three-dimensional region of interest;
An ultrasonic data processing apparatus comprising:

Images