JP5210960B2 - Ultrasonic diagnostic equipment - Google Patents

Description

  The present invention relates to an ultrasonic diagnostic apparatus, and more particularly to a technique for automatically detecting a boundary line (contour line) of a living tissue on an ultrasonic image.

  Ultrasound diagnostic apparatuses are used in the medical field. The ultrasonic diagnostic apparatus is an apparatus that transmits and receives ultrasonic waves to and from a living body and forms an ultrasonic image based on reception data obtained thereby. The ultrasonic diagnostic apparatus has a plurality of operation modes (B mode, Doppler mode, etc.) and a plurality of measurement functions. Examples of such a measurement mechanism include an NT (Nuchal Translucency) measurement function and an IMT (Intima-Media Thickness) measurement function. They are common in that the distance is measured between two points (or between two boundary lines) on the tomographic image.

  More specifically, in NT measurement (NT thickness measurement), the size (thickness) of posterior cervical edema (NT) in the fetus is measured. The edema can be generally recognized on the back side of the neck in a fetus about 10 to 14 weeks of gestation. Usually, the edema is observed as an elongated bag-like tissue on a tomographic image. When the outline is roughly divided into two boundary lines running in parallel, the interval between the two boundary lines (contour lines), in particular, the maximum width of edema (body fluid part) in the direction perpendicular to the longitudinal direction of the edema (hereinafter referred to as “NT thickness”). ")" Is known to be one index value for diagnosing fetal conditions or diseases. For example, when the NT thickness is larger than the reference value, more congenital diseases such as Down's syndrome are recognized than when the NT thickness is not. Therefore, it is desired to check the NT thickness using ultrasonic diagnosis. In the NT measurement, ultrasonic waves are transmitted / received to / from the fetus, whereby a tomographic image is displayed on the screen. On the tomographic image, two marks are manually accurately positioned on two boundary lines indicating the fetal cervical edema. At that time, the direction and position for distance measurement are determined by visual observation. Thereafter, distance measurement is automatically performed, that is, the distance between two marks is measured as NT thickness. Similar measurements are performed at a plurality of locations as necessary.

  On the other hand, in IMT measurement, one or both of the front wall and the rear wall (viewed from the probe) in the carotid artery is a measurement target. The blood vessel wall (that is, the anterior wall and the posterior wall) is composed of an intima, an intima and an adventitia when viewed from the blood flow side, that is, has a three-layer structure. In the IMT measurement, the thickness (IMT) of the composite including the inner membrane and the inner membrane is measured. This is because, on the tomographic image, the blood vessel wall is displayed quite thin, but the inner boundary line inside the intima and the outer boundary line between the media and the outer membrane are relatively likely to appear. IMT is measured as the distance between the two boundaries. It is known that when arteriosclerosis is promoted, IMT increases. By measuring IMT, one index value for diagnosing the degree of arteriosclerosis in a subject can be obtained. Specifically, in IMT measurement using an ultrasonic diagnostic apparatus, ultrasonic waves are transmitted to and received from the carotid artery, thereby displaying a tomographic image. Next, the markers are manually positioned on the inner and outer boundary lines on the front or rear wall, respectively. At this time, the two markers are positioned so that the two markers are aligned in a direction orthogonal to the blood vessel axis as much as possible. The distance between the two markers is then automatically measured and taken as IMT. The IMT may be measured at a plurality of locations, and in such a case, the maximum value, the average value, etc. are calculated.

  The following Patent Documents 1 to 6 disclose an ultrasonic diagnostic apparatus that performs IMT measurement. Patent Document 7 describes a stepwise setting of a threshold based on a histogram. Patent Document 8 describes the calculation of the standard deviation for the histogram.

JP 2008-168016 A JP 2007-319255 A JP 2000-271117 A JP 2006-51285 A JP 2006-122686 A JP-A-2005-268 Japanese Patent Publication No. 7-32773 JP 2008-253379 A

  In the manual thickness measurement as described above, the burden on the inspector is large and the reproducibility is not very good. Differences in measurement results tend to occur between inspectors. Moreover, since it is based on visual judgment, when the boundary line is unclear, the reliability of measurement tends to be extremely lowered. Therefore, automatic measurement or semi-automatic measurement of the boundary line is required. However, since a tomographic image contains a lot of noise, it is a major factor in erroneous detection. In detecting the boundary line, a threshold method is usually used. In this case, in order to increase the reliability of measurement, it is required to set a threshold value suitable for the image situation.

  An object of the present invention is to increase the automatic detection accuracy of the boundary line.

  Alternatively, an object of the present invention is to be able to appropriately set a threshold for detecting a boundary line.

  Alternatively, an object of the present invention is to make it possible to adaptively set an image range to be referred to when setting a threshold for detecting a boundary line.

  Alternatively, it is an object of the present invention to improve the detection accuracy of the boundary line and thus the thickness measurement accuracy in NT measurement or IMT measurement on a tomographic image.

  (1) An ultrasonic diagnostic apparatus according to the present invention includes a region-of-interest setting unit that sets a region of interest including at least one target boundary line for a target tissue on a tomographic image obtained by transmission and reception of ultrasonic waves. Within the region of interest, a reference point is searched for on a search line extending in a search direction intersecting the base line from each point on the base line of the region of interest, and based on the reference point on each search line Image processing for detecting the target boundary line by referring to a plurality of pixel values belonging to the band setting section and section setting means for configuring a band area as a set of a plurality of sections, and a plurality of pixel values belonging to the band area Determining means for determining a parameter.

  According to the above configuration, the region of interest is set on the tomographic image. In this case, preferably, the position and rotation angle of the region of interest are designated by the user in accordance with the traveling form (particularly the traveling direction) of the boundary of interest, and the region of interest is set based on the designation. However, the region of interest may be automatically set by analyzing the tomographic image (or other ultrasonic image). A plurality of regions of interest may be set on the tomographic image, and image processing may be executed independently in parallel in each region of interest. The attention boundary line is a boundary line (contour line, edge) to be detected on the ultrasonic image. In the detection of such a boundary line, it is necessary to adaptively set parameters used in the detection, but according to the present invention, when setting the parameters, the band region is adaptively set, That is, since the reference range for parameter setting can be narrowed down (irrelevant image portions that cause an error can be excluded from the reference range), as a result, the set parameter can be improved. Here, the parameter is preferably a threshold value to be compared with the pixel value in edge detection. The band region is preferably a partial region including the target boundary line to be detected and extending along the direction. Prior to determining such a band region, a reference point is detected (specified) on each search line, and a section (reference section) is set on the search line with the reference point as a reference. The reference point is preferably an edge point or a rising point on the near side of the boundary point to be detected. However, the reference point only needs to be an objectively determined point according to the image content (particularly according to the luminance distribution on the search line). For example, the boundary point on the target boundary line itself or a point close thereto is the reference point. May be specified. In that case, it is desirable to define the section around the reference point. For example, a point far from the baseline can be used as the reference point, but it is desirable to determine a point close to the baseline as the reference point from the viewpoint of speeding up the calculation. When the threshold method is used for detecting the reference point, the entire region of interest may be used as a reference object in determining the threshold (temporary threshold, threshold for detecting the reference point). In such a configuration, the threshold method is used in stages, that is, the temporary threshold based on the overall reference and the main threshold based on the local reference are set in stages. . The search direction is preferably a direction orthogonal to the base line, but the search direction can also be determined according to a background coordinate system (for example, a display coordinate system).

  As described above, according to the present invention, it is possible to adaptively determine a local region called a band region and improve image processing parameters by referring to the local region. Therefore, the detection accuracy can be improved and the reproducibility can be improved when detecting the target boundary line.

  Preferably, the determination unit includes a histogram generation unit that generates a histogram based on a plurality of pixel values belonging to the band region, and a threshold determination unit that determines a threshold as the image processing parameter based on the histogram. . As is well known, the histogram represents a frequency distribution for each pixel value and indicates an image state or an image tendency. Various methods are conventionally known as a method for determining a threshold value from a histogram. In the present invention, such various methods can be used. When determining a threshold value, a plurality of feature amounts are extracted from the histogram, and an optimum condition is selected from a plurality of threshold value determination conditions using them, and a threshold value (temporary threshold value, main threshold value) is determined according to the optimum condition. You may do it. The threshold may be determined directly from the histogram, or indirectly from the histogram (for example, a threshold determination function is defined from the histogram, and the threshold is actually determined from the threshold determination function). Also good). Note that a correction function for an image to be subjected to boundary detection may be defined based on a histogram.

  Preferably, the section setting means sequentially references pixel values from the base line on each search line and compares the pixel values with a temporary threshold to detect the reference point, and A section setting unit that sets the section on the basis of the reference point on each search line. In setting the temporary threshold, a histogram generated by referring to the region of interest may be used. The reference point serves as an objective reference when setting the section, and the reference point may constitute one end of the section, or the reference point may constitute an internal point of the section. If a plurality of sections (for example, first and second sections for detecting the first and second boundary lines) are set based on one reference point, a quick calculation can be expected.

  Preferably, the temporary threshold is a threshold for detecting a point on the near side of the target boundary line. Preferably, the section setting unit determines the reference point as one end of the section, and determines a point away from the reference point forward in the search direction as the other end of the section. Preferably, the section setting unit determines a point away from the reference point forward in the search direction as one end of the section, and determines a point further away from the reference point forward in the search direction as the other end of the section. According to this configuration, for example, in IMT measurement, both the inner boundary line detection section and the outer boundary line detection section can be set based on the reference point, which is convenient. From the viewpoint of reducing the amount of computation and improving the detection accuracy, it is desirable that the boundary line (boundary point) is searched only within the band region (that is, the section) on each search line.

  Preferably, input means for designating a position and an angle of the region of interest on the tomographic image by a user is included. That is, while viewing the tomographic image, it is desirable to determine the rotation angle of the region of interest (that is, the measurement coordinate system) so that the baseline direction is as parallel as possible to the traveling direction of the target boundary line.

  Preferably, the region of interest is a rectangular region, the upper side or the lower side of the region of interest is the base line, and each search line is set in parallel with the left and right sides of the region of interest. Preferably, the at least one target boundary line includes an inner boundary line and an outer boundary line in the blood vessel wall.

  Preferably, the region of interest includes an upper sub-region of interest and a lower sub-region of interest connected via an intermediate line functioning as the base line, and a boundary of interest is detected in each sub-region of interest. Is called. Each of the two sub-regions of interest forms part of the region of interest. According to the above configuration, it is simple because it is only necessary to set one (composite) region of interest for two boundary lines running in parallel up and down. It is desirable to set the region of interest so that the intermediate line falls between the two boundary lines. Using the three lines of the upper side, the middle line, and the lower side as a guide, the region of interest can be easily set at an appropriate position and rotation angle with respect to the two boundary lines.

  Preferably, the at least one attention boundary includes an upper boundary and a lower boundary that represent a fetal cervical edema outline, the upper boundary is detected in the upper sub-region of interest, and The lower boundary line is detected in the lower sub region of interest.

  Preferably, the threshold value determination unit includes a histogram analysis unit that calculates a plurality of feature amounts from the histogram, and the threshold value is determined using the plurality of feature amounts. Preferably, the plurality of feature amounts include a first feature amount defined as a type to which a shape of the histogram belongs, a second feature amount defined from a magnitude relationship of average luminance before and after the peak in the histogram, At least one of the third feature amount defined from the deviation on the luminance axis and the fourth feature amount defined from the degree of spread of the histogram is included. For example, in order to obtain the first feature value, for example, in order to obtain the first feature amount, a histogram having a bimodal shape, a horizontally flat shape having a unimodal property, and a vertically having a unimodal property are used. Which type belongs to the sharp shape is automatically determined. In order to obtain the second feature amount, for example, the average frequency before the peak and the average frequency after the peak are compared, and it is automatically determined which is greater. In order to obtain the third feature amount, for example, it is automatically determined which of the high luminance tendency and the low luminance tendency is applicable. In order to obtain the fourth feature amount, it is automatically determined which of the high-frequency tendency and the low-frequency tendency is applicable.

  (2) An ultrasonic diagnostic apparatus according to the present invention includes a region-of-interest setting unit that sets a region of interest including an inner boundary line and an outer boundary line in a blood vessel wall on a tomographic image obtained by transmitting and receiving ultrasonic waves. In the region of interest, a reference point is searched on a search line extending in a search direction intersecting the base line from each point on the base line of the region of interest, and based on the reference point on each search line Section setting means for setting a first zone and a second zone, thereby forming a first zone region as a set of a plurality of first zones, and forming a second zone region as a set of a plurality of second zones; , By referring to a plurality of pixel values belonging to the first band region, an inner threshold value determining means for determining an inner threshold value for detecting the inner boundary line, and a plurality of pixel values belonging to the second band region. Refer It allows characterized in that it comprises a an outer threshold determination means for determining the outer threshold for detecting said outer boundary. A plurality of first sections are arranged in the direction of the baseline, thereby defining a first band region. Similarly, a plurality of second sections are arranged in the baseline direction, and the second band region is defined by them. Desirably, the first belt region and the second belt region have a positional relationship where they are adjacent (parallel running) without overlapping. However, a partial overlap may occur between them.

  Preferably, means for generating an inner trace line along the inner boundary line, means for generating an outer trace line along the outer boundary line, and the blood vessel between the inner trace line and the outer trace line Means for measuring the thickness of the wall.

(3) The ultrasonic diagnostic apparatus according to the present invention includes an upper sub-region of interest including an upper boundary line of fetal posterior cervical edema on the tomographic image obtained by transmitting and receiving ultrasonic waves and the posterior cervical edema. A region-of-interest setting means for setting a composite region of interest comprising a lower sub-region of interest including a lower boundary line, and an intermediate point from each point on an intermediate baseline of the composite region of interest within the upper sub-region of interest The upper reference point is searched on the upper search line extending in the upper search direction intersecting the base line, and the upper section is set based on the upper reference point on each upper search line, whereby a set of a plurality of upper sections is set. An upper section setting means that constitutes an upper band area as an intermediate base line from each point on the intermediate baseline of the composite region of interest within the lower sub-region of interest Search for the lower reference point on the lower search line extending in the intersecting lower search direction, and set a lower section based on the lower reference point on each lower search line, thereby An upper side for determining an upper threshold for detecting the upper boundary line by referring to a lower section setting means constituting a lower band area as a set of side sections and a plurality of pixel values belonging to the upper band area A threshold value determining means; and a lower threshold value determining means for determining a lower threshold value for detecting the lower boundary line by referring to a plurality of pixel values belonging to the lower band region. To do.

  Preferably, means for generating an upper trace line along the upper boundary line, means for generating a lower trace line along the lower boundary line, and between the upper trace line and the lower trace line. And means for measuring the thickness of the posterior cervical edema.

  (4) A program according to the present invention is a program executed in an ultrasonic diagnostic apparatus or an information processing apparatus that processes an ultrasonic image, and based on user designation on the ultrasonic image, at least one for a target tissue A function of setting a region of interest including two attention boundary lines, and a search for a reference point on a search line extending in a search direction intersecting the base line from each point on the base line of the region of interest within the region of interest And setting a section based on a reference point on each search line, thereby forming a band area as a set of a plurality of sections arranged in the direction of the baseline, and a plurality of members belonging to the band area A function for determining a threshold value for detecting the target boundary line by referring to the pixel value of the image, and the determination on the ultrasonic image. Characterized in that it comprises a, a function of detecting the target boundary line with the threshold.

  According to the present invention, the automatic detection accuracy of the boundary line can be increased. Or according to this invention, the threshold value for detecting a boundary line can be set appropriately. Alternatively, according to the present invention, it is possible to adaptively set an image range that is referred to when setting a threshold for detecting a boundary line. Or according to this invention, in NT measurement or IMT measurement on a tomographic image, the detection accuracy of a boundary line, and thickness measurement accuracy can be improved.

1 is a block diagram showing an overall configuration of an ultrasonic diagnostic apparatus according to the present invention. It is a figure which shows IMT measurement on the tomographic image of a blood vessel. It is a figure which shows NT measurement (NT thickness measurement) on the tomographic image of a fetus. It is a figure which shows the setting of three area points on each search line in IMT measurement. It is a figure which shows the other setting method of three area points in IMT measurement. It is a figure which shows the two belt | band | zone area | regions set in the region of interest in IMT measurement. It is a figure which shows the distance measurement (IMT measurement) between two trace lines. It is a flowchart for demonstrating the operation example in IMT measurement. It is a figure for demonstrating the section setting on each search line in NT measurement. It is a figure which shows the belt | band | zone area | region set in each sub-region of interest in NT measurement. It is a figure which shows the distance measurement (NT thickness measurement) between two trace lines. It is a flowchart for demonstrating the operation example in NT thickness measurement. It is a figure for demonstrating the 1st example of the edge detection method using a threshold value. It is a figure which shows the 2nd example of the edge detection method using a threshold value. It is a figure which shows the 3rd example of the edge detection method using a threshold value. It is a figure which shows the 4th example of the edge detection method using a threshold value. It is a figure for demonstrating selection of the threshold value determination conditions based on several histogram feature-value.

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

(1) Configuration of Ultrasonic Diagnostic Device FIG. 1 shows a preferred embodiment of an ultrasonic diagnostic device according to the present invention, and FIG. 1 is a block diagram showing the overall configuration thereof. The probe 10 is an ultrasonic probe that is brought into contact with the body surface, and transmits ultrasonic waves to the living body and receives reflected waves from within the living body. The probe 10 has a 1D array transducer (not shown). The 1D array transducer is composed of a plurality of transducer elements, whereby an ultrasonic beam B is formed. The ultrasonic beam B is electronically scanned, thereby forming a scanning surface S that is a two-dimensional data capturing area. As the electronic scanning method, electronic linear scanning, electronic sector scanning, and the like are known. A 2D array transducer for forming a three-dimensional data capture area may be provided in the probe 10. In the present embodiment, when IMT measurement is performed, the probe unit 10 is brought into contact with the neck, thereby acquiring tomographic image information of the carotid artery. On the other hand, when NT measurement is performed, the probe 10 is brought into contact with the abdomen of a pregnant woman, whereby tomographic image information of the fetus is acquired.

  The transmission unit 12 functions as a transmission beam former. The transmitter 12 supplies a plurality of transmission signals to the 1D array transducer during transmission. As a result, a transmission beam is formed. On the other hand, when a reflected wave is received by the probe 10 from the living body, a plurality of reception signals are output in parallel from the 1D array transducer, and are processed by the reception unit 14. The reception unit 14 is configured as a reception beam former, and performs phasing addition processing on a plurality of reception signals, thereby outputting a reception signal (beam data) after phasing addition.

  The signal processing unit 16 performs some signal processing on the beam data. Such signal processing may include processes such as detection, logarithmic compression, noise removal and the like. Beam data output from the signal processing unit 16 is sent to the image forming unit 17. The image forming unit 17 is configured by a digital scan converter (DSC) in this embodiment. The image forming unit 17 has a coordinate conversion function, an interpolation processing function, and the like. In the image forming unit 17, a tomographic image is constructed based on the plurality of beam data. The image data is sent to the display processing unit 18 and also to the measurement unit 22.

  The display processing unit 18 has an image composition function and the like, and outputs display image data to the display unit 20. In the display processing unit 18, processing for combining a graphic image with a tomographic image is performed. The display unit 20 displays a tomographic image as an ultrasonic image. An ultrasonic image (such as a two-dimensional color Doppler image) other than the tomographic image may be displayed.

  The control unit 24 performs operation control of each configuration shown in FIG. 1, and the control unit 24 includes a CPU and an operation program. An operation panel 26 is connected to the control unit 24. The operation panel 26 includes a keyboard and a trackball and functions as a user input device. Using the operation panel 26, the user can set a region of interest (ROI). Specifically, while viewing a tomographic image displayed on the screen, an arbitrary position and an arbitrary rotation angle on the tomographic image. It is possible to specify a region of interest with an arbitrary size. Based on this user input, the control unit 24 sets (recognizes) the region of interest.

  The measurement unit 22 functions as an IMT measurement unit and an NT measurement unit in this embodiment. Specifically, the measurement unit 22 traces two boundary lines (contour lines) in the region of interest to generate two trace lines, and measures a distance, that is, a thickness between them. Incidentally, the coordinate information of the region of interest is given from the control unit 24 to the measurement unit 22. The measurement unit 22 is actually realized as a calculation function of software, and it may constitute a part of the control unit 24. The measurement result of the measurement unit 22 and necessary graphic data are sent from the measurement unit 22 to the display processing unit 18.

  FIG. 2 shows the state of IMT measurement. The image 28 is a tomographic image (B-mode image) of the blood vessel. A cross section of the blood vessel is represented on the tomographic image 28, and the blood vessel has a front wall 30 and a rear wall 32. Incidentally, the side closer to the probe side is the front wall 30 and the side farther from the probe side is the rear wall 32. Between the front wall 30 and the rear wall 32 is a blood flow part 34. As is well known, the front wall 30 includes an inner membrane 36, a middle membrane 38 and an outer membrane 40 as viewed from the blood flow portion 34 side. Here, the intima 36 is a very thin layer, which appears substantially on the image as a single line. In the IMT measurement, the thickness of the complex between the inner boundary line 42 and the outer boundary line 44, that is, the inner membrane 36 and the inner membrane 38 is measured. Similar to the front wall 30, the rear wall 32 includes an inner membrane 46, a middle membrane 48, and an outer membrane 50 as viewed from the blood flow portion 34. Reference numeral 52 represents an inner boundary line, and reference numeral 54 represents an outer boundary line.

  When performing IMT measurement on the front wall 30 on the tomographic image as described above, a region of interest (ROI) 56 is set so as to include the front wall 30 in order to limit the calculation range. Similarly, when performing IMT measurement on the rear wall 32, the region of interest (ROI) 60 is set so as to include the rear wall 32. In the region of interest 56, as indicated by reference numeral 58, the distance between the intima boundary line 42 and the epicardial boundary line 44 is measured as IMT 58. Similarly, in the region of interest 60, the distance between the inner boundary line 52 and the outer boundary line 54 is measured as the IMT 62.

  In this embodiment, the regions of interest 56 and 60 are specified by the user, and at that time, the horizontal lines (upper side and lower side) of the regions of interest 56 and 60 are as parallel as possible to the respective boundary lines. The tilt angle of the area is specified. In the present embodiment, the lower side of the region of interest 56 is the base line 56A, and the base line 56A functions as a start line when a plurality of search lines are set. On the other hand, in the region of interest 60, the upper side is defined as a baseline 60A, and the baseline 60A also functions as a start line when defining a plurality of search lines. Incidentally, when performing IMT measurement, the region of interest 56, 60 is designated by the user so that both the inner boundary line and the outer boundary line are included in the region of interest 56, 60. Of course, such a setting can be automated. In FIG. 2, the X direction represents the horizontal direction on the display screen, and the Y direction represents the vertical direction on the display screen. The x direction described later is the horizontal direction of the region of interest, and the y direction is the vertical direction of the region of interest. The former is an absolute display coordinate system, while the latter is a relative coordinate system based on the region of interest.

  FIG. 3 shows a state where NT measurement (that is, NT thickness measurement) is performed. The tomographic image 164 represents a cross section of the fetus 166. Reference numeral 168 represents the placenta. Reference numeral 170 represents a fetal cervical edema. The edema 170 appears as an elongated cavity on the tomographic image 164. Specifically, the edema is a region surrounded by an upper boundary line 174 and a lower boundary line 176 that run side by side on the tomographic image, and the inside 172 is a body fluid part.

  In this embodiment, when performing NT measurement, the region of interest 178 is designated by the user as shown. The region of interest 178 is configured as a combination of two sub-regions of interest 180 and 182, and a base line 178 </ b> A serving as an intermediate line is formed between the sub-regions of interest 180 and 182. The upper boundary line 174 is traced in the upper sub region of interest 180, and similarly, the lower boundary line 176 is traced in the lower sub region of interest 182. When two trace lines are drawn by these processes, the NT thickness 184 is measured between them. In such NT measurement, similarly to the above-described IMT measurement, when the user designates the region of interest 178, the calculation process basically proceeds automatically after the designation, thereby reducing the burden on the user. . Moreover, since it does not depend on subjectivity, objectivity or reproducibility can be improved. Incidentally, when specifying the region of interest 178 by the user, the position and inclination of the combined region of interest 178 are specified so that the boundary lines 174 and 176 are included in the sub regions of interest 180 and 182, respectively. In particular, the region of interest 178 is defined such that the base line 178A, which forms an intermediate line, is located exactly between the two boundary lines 174, 176. Here, the base line 178A forms a start line when a plurality of search lines are set in each of the sub-regions of interest 180 and 182. Therefore, it is desirable to define the region of interest 178 so that the base line 178A is as parallel as possible to the two boundary lines 174 and 176.

(2) Specific Example of IMT Measurement Next, a specific example of IMT measurement will be described with reference to FIGS.

  In FIG. 4, a region of interest 56 is shown. The region of interest 56 includes a part of the inner boundary line 42 and a part of the outer boundary line 44. In FIG. 4, the boundary lines 42 and 44 are represented as straight lines for convenience. Incidentally, the code | symbol 30 represents the front wall and the code | symbol 34 represents the blood-flow part. That is, the region of interest 56 is a region of interest set for the front wall 30.

  In FIG. 4, the lower side of the region of interest 56 functions as a baseline 56A. That is, the base line 56A is a line extending in the x direction, and a search line 58 is formed in the y direction at each position in the x direction. Specifically, each search line 58 is a line extending in the y direction orthogonal to the base line 56A, starting from each point (each pixel) on the base line 56A. That is, the direction becomes the search direction. If the ROI is rectangular, the search direction is parallel to the left and right sides of the region of interest 56. As shown in FIG. 4, pixel value reference and edge detection are sequentially performed from the start point 58A toward the search direction (forward) on each search line. That is, the pixel value of the target pixel is referred to, and the pixel value is compared with a predetermined threshold value (provisional threshold value). If the pixel value is larger than the threshold value, it is identified as an edge point (rising point), and The point becomes the reference edge point. In the figure, reference edge points are indicated by reference numeral 60. The reference edge point 60 functions as a first section point in the present embodiment, and the second section point 62 and the third section point 64 are determined based on the reference edge point 60. The temporary threshold may be a fixed value, but is preferably a threshold determined from a histogram formed by referring to the entire region of interest. Various methods are conventionally known as a method for determining a threshold value from a histogram.

  More specifically, the reference edge point 60 is a point detected in front of the inner boundary line 42. In other words, a temporary threshold is set so that such a rising point on the near side is detected. ing. The reference edge point 60 serves as a reference point when the two sections are defined, that is, the origin, and the second section point 62 is defined as a point obtained by adding a certain distance W1 forward in the search direction with respect to the reference edge point 60. Thus, the third section point 64 is determined as a point obtained by adding a certain distance W2 forward in the search direction with respect to the second section point 62. The third section point 64 may be determined by adding the distance W2 to the reference edge point 60. W2 is larger than W1. A section between the first section point 60 and the second section point 62 is a first section, and a section between the second section point 62 and the third section point 64 is a second section. The first section constitutes an element of a first belt region described later, and the second section constitutes an element of a second belt region described later. As a result, each first section is set to straddle the inner boundary line 42, that is, the sizes of the reference edge points 60 and W1 are determined so that such a condition is satisfied. Similarly, the position of the third section point 64, that is, the size of W2 is determined so that each second section straddles the outer boundary line 44.

  In the specific example shown in FIG. 4, the reference edge point 60 is detected on the front side of the inner boundary line, but instead, a method as shown in FIG. 5 can be adopted. In the method shown in FIG. 5, edge detection processing is sequentially executed from the start point along the search line 58. Here, the temporary threshold value is set to a value that can detect the inner boundary line 42 that is the original edge or a position close thereto, and as a result, the temporary edge point 66 is found on or near the inner boundary line 42. A first section point 68 is set at a predetermined distance on the near side of the temporary edge point 66, and a second section point 70 is set at a predetermined distance on the far side of the temporary edge point 66. A portion between the first section point 68 and the second section point 70 is determined as the first section. On the other hand, the search is started again from the temporary edge point 66 forward in the search direction. In that case, another temporary threshold value is set so that the outer boundary line 44 or its vicinity is detected, and by repeating the comparison between such another temporary threshold value and each pixel value, the temporary edge point 72 is obtained. Detected. With this as a reference, a first segment point 74 is defined on the front side thereof, and a second segment point 76 is defined on the back side, that is, the front side of the temporary edge point 72. Between the first section point 74 and the second section point 76 is the second section. As described above, an edge to be detected or a point close thereto may be detected, and the section may be determined based on the detected edge. However, according to the method shown in FIG. 4, the second interval point 62 and the third interval point 64 can be set immediately by simply detecting the reference edge point that is relatively close to the start point. The advantage is that it can be reduced to achieve rapid processing.

  In FIG. 6, the reference edge point is detected on the search line 58 at each coordinate in the x direction and two section points are set based on the reference edge point, thereby arranging a plurality of first section points on the plurality of search lines 58 first. As shown in FIG. Similarly, a second section point line 80 is configured as an array of second section points on the plurality of search lines 58. Further, a third section point line 82 is configured as an array of a plurality of third section points on the plurality of search lines 58. A region between the first segment point line 78 and the second segment point line 80 is defined as an inner band region 84. Similarly, the outer zone area 86 is defined between the second segment point line 80 and the third segment point line 82.

  The inner band region 84 includes the inner boundary line 42 therein, which is a region near the inner boundary line. The outer belt region 86 includes an outer boundary line 44 therein, which is a region near the outer boundary line. The inner band area 84 is a local area referred to for determining a threshold (main threshold) used when detecting the inner boundary line 42, and at the same time, the inner band area 84 is a search area for the inner boundary line 42. But there is. The outer band region 86 is a local region that is referred to in order to determine a threshold value (this threshold value) used when detecting the outer boundary line 44, and the outer band region 86 is a search region for the outer boundary line 44. It is what you make.

  Specifically, by referring to the inner band region 84, a histogram is created based on a plurality of pixel values included therein, and a threshold for edge detection (this threshold) is directly or indirectly from the histogram. Is determined. Also for the outer band region 86, a histogram is created by referring to the internal pixel value group, and a threshold value (this threshold value) is determined based on the histogram. Conventionally, various methods have been proposed as a method for determining a threshold value based on a histogram. The inner boundary point 88 is detected on each search line 58 using the threshold value determined by referring to the inner band region 84. In that case, on each search line, the first interval point line 78 in the inner band region 84 is used as a start line, and edge detection processing using a threshold value is sequentially executed from there to the second interval point line 80 side. Similarly, in the outer band region 86, the detection of the outer boundary point 90 is executed on each search line using the threshold value determined by referring to the outer band region 86. Also in this case, edge detection is sequentially performed on the side of the third section point line 82 on each search line using the second section point line 80 as a start line.

  FIG. 7 shows the inner trace line 92 and the outer trace line 94 formed as described above. The inner trace line 92 is constituted by an array of a plurality of inner boundary points 88 obtained by edge detection with respect to the inner boundary line. The outer trace line 94 includes a plurality of outer boundary points obtained by edge detection with respect to the outer boundary line. Of course, a smooth trace line may be configured by applying interpolation processing, smoothing processing, or the like to the arrangement of a plurality of boundary points. When the two trace lines 92 and 94 are automatically extracted and drawn as described above, IMT measurement is performed between the two trace lines 92 and 94. Specifically, IMT measurement is performed at one or more points in the x direction. Here, IMT is represented by reference numeral 96. In this case, the direction in which the thickness, that is, the distance is measured is the y direction in the example shown in FIG. The distance measurement direction may be adaptively determined based on the shapes of the two trace lines.

  According to the above-described method, the inner boundary line and the outer boundary line can be accurately and accurately detected after determining appropriate threshold values for each band area by exploratory setting of the two band areas. There is an advantage that the trace line can be accurately drawn. In the past, the entire region of interest was always referred to when setting the threshold. However, such a method refers to a lot of information other than the information near the target boundary line, and in some cases it is affected by noise. However, in this embodiment, it has been confirmed by experiments that such a problem can be greatly alleviated or avoided. In particular, in the present embodiment, since the band regions are defined for two boundary lines, the threshold value suitable for each boundary line can be set to greatly increase the reliability of boundary detection.

FIG. 8 shows an example of an IMT measurement method as a flowchart. In S100, a region of interest (ROI) is designated by the user on a tomographic image representing a blood vessel. In that case, it is desirable to set the ROI so that the horizontal axis in the region of interest is as parallel as possible to the two boundary lines. In S102, the entire region of interest is referred to determine a provisional threshold, and a histogram (overall histogram) is created based on a plurality of pixel values included therein. In S104, the provisional threshold a ₀ is calculated based on the histogram. The temporary threshold a ₀ is for determining the reference edge point as described above and does not require strict boundary detection. Therefore, even when the entire region of interest is referred to when determining the temporary threshold a ₀ There is basically no problem.

In S106, the search for the reference edge point is started on any of the search lines designated in order. In S108, the reference pixel values is compared to the temporary threshold value a _0, the pixel value S110 proceeds if it is temporary threshold a ₀ or is executed. Incidentally, error processing may be applied when the reference edge point cannot be detected even when reaching the upper side of the region of interest, that is, the end line. Examples of such error processing include a method of determining a reference edge point on the search line by interpolation using a reference edge point detected on an adjacent search line.

  In S110, the reference edge point, that is, the first section point is specified. In S112, the second section point is specified based on the reference edge point, and in S114, the third section point is specified based on the reference edge point. As a result, three section points are determined on the search line currently being processed. In S116, it is determined whether there is an unprocessed search line, and if there is, the process from S106 is repeatedly executed. On the other hand, when it is determined in S116 that the process has progressed to the last search line, the steps after S118 and the steps after S130 described below are executed.

  A series of steps from S118 to S128 and a series of steps from S130 to S140 may be executed in order or may be executed in parallel. A series of steps from S118 to S128 corresponds to the tracing process for the inner boundary line, and a series of steps from S130 to S140 corresponds to the tracing process for the outer boundary line.

In S118, the first band region is referred to, and a first partial histogram is created based on the pixel group included therein. In S120, the threshold value a ₁ is determined based on the first portion histogram. In S122, by utilizing the threshold value a _1, which search is started on the search line, if found threshold a ₁ or more pixel values in S124, the pixel value is specified as the inner boundary point in S126. Until it is determined in S128 that there are no unprocessed search lines, the steps after S122 are repeated.

On the other hand, in S130, the second band region is referred to, and a second partial histogram is created based on the pixel group included therein. In S132 it is determined the threshold a ₂ based on the second portion histogram in S134, the search on the specified search line is started. If a pixel value greater than or equal to the threshold value a ₂ is found in S136, the point having the pixel value is identified as the outer boundary point in S138, and whether or not the process has progressed to the last search line in S140. To be judged. If the process has not reached the last search line, the processes after S134 are repeatedly executed.

  When a plurality of inner boundary points are detected in the first band region and a plurality of outer boundary lines are detected in the second band region, that is, when an inner trace line and an outer trace line are formed, in S142, IMT measurement is performed between the two trace lines. In this case, IMT measurement is performed at a plurality of locations in the x direction, and the maximum value, minimum value, average value, and the like are obtained from the measurement results.

(3) Specific Example of NT Measurement Next, a specific example of NT measurement will be described with reference to FIGS. 9 to 12.

  FIG. 9 shows a region of interest 178 set on the tomographic image. The region of interest 178 includes a cross section of the fetal posterior cervical edema (NT) 170, and the region of interest 178 includes an upper boundary line 174 and a lower boundary line 176. More specifically, the region of interest 178 is divided into an upper sub-region of interest 180 and a lower sub-region of interest 182 with a base line 178A forming an intermediate line as a connecting line. 174 is included, and a lower boundary line 176 is included in the lower sub-region of interest 182.

  Within the region of interest 178, the baseline 178A functions as a common start line for the two sub-regions of interest 180,182. That is, in the upper sub-region of interest 180, the search line 100 is set in the y direction at each coordinate in the x direction, and the edge search is repeatedly executed on each search line 100 from the start point 100A on the base line 178A toward the upper side. The The threshold used in that case is a temporary threshold. When the pixel value exceeds the temporary threshold, the point is determined as the reference edge point (first section point) 102. Then, the second section point 104 is determined as a point separated by a predetermined distance W3 with respect to the reference edge point 102. As a result, the first segment point 102 and the second segment point 104 are determined on the search line 100. Similarly, also in the lower sub region of interest 182, the search line 106 is set in the y direction for each coordinate in the x direction, and edge detection is performed from the start point on the base line 178A toward the lower side that is the end line. It is executed repeatedly. At a point where the pixel value exceeds the provisional threshold value, it is determined as a reference edge (first interval point) 108, and the second interval point 110 is specified as a point separated by a predetermined distance W4 with reference to that. As a result, the first section point 108 and the second section point 110 are determined on each search line 106. Each first section in the upper sub region of interest 180 is set across the upper boundary line 174, and each second section in the lower sub region of interest 182 is set across the lower boundary line 176.

  As shown in FIG. 10, if the above processing is performed for each x coordinate (for each pixel arranged in the X direction), it is possible to define the band region 120 in the upper sub region of interest 180. That is, the band region 120 is defined as a region between the first section point line 112 and the second section point line 114, and here, the first section point line 112 is configured as an array of a plurality of first section points 102. The second section point line 114 is configured as an array of a plurality of second section points 104. Similarly, in the lower sub region of interest, a band region 122 is defined, that is, the band region 122 is an area sandwiched between the first section point line 116 and the second section point line 118. The first section point line 116 is configured as an array of a plurality of first section points 108, and the second section point line 118 is configured as an array of a plurality of second section points 110.

  When the band region 120 is determined, the pixels in the band region 120 are referred to, and a histogram is generated based on a plurality of pixel values. Based on the histogram, a threshold value (main threshold value) for detecting the upper boundary line is calculated. Similarly, the pixels in the band region 122 are referred to, a histogram is generated based on a plurality of pixel values, and a threshold value (main threshold value) for detecting the lower boundary line is calculated from the histogram. In the belt region 120, the upper boundary point 124 is detected on each search line 110 using the threshold value obtained as described above. As a result, an upper trace line 126 is formed as an array of a plurality of upper boundary points 124. On the other hand, in the band region 122, edge detection is performed on each search line using the threshold value determined as described above, and thereby a plurality of lower boundary points 128 are detected. A lower trace line 130 is formed as an arrangement of them. Of course, smoother trace lines 126 and 130 may be configured by applying interpolation processing and smoothing processing to the plurality of upper boundary points 124 and the plurality of lower boundary points 128. Good.

  In FIG. 11, the NT thickness 132 is measured between the upper trace line 126 and the lower trace line 130. It is desirable to measure the NT thickness 132 at a plurality of coordinates in the x direction. Although the thickness measurement direction is parallel to the y direction in the embodiment shown in FIG. 11, it is desirable to adaptively determine the distance measurement direction in an appropriate direction according to the form of the trace lines 126 and 130.

  According to the NT measurement method described above, there is an advantage that each boundary line can be accurately detected by setting a threshold value independently in the two sub-regions of interest. In addition, there is an advantage that the NT thickness can be measured in the entire region of interest. Also in the NT measurement method described above, a band region that is a local region is defined in detecting each boundary line, and a threshold value is determined based on a pixel value group referred to in the band region. Can be optimized. In addition, since the boundary line is detected in the band region, there is an advantage that the influence of noise or the like can be eliminated as much as possible.

  FIG. 12 is a flowchart showing the NT measurement method described above. In S200, the combined region of interest is designated by the user on the tomographic image. In this case, the position and inclination of the region of interest are determined so that the boundary lines are included in the two sub regions of interest. When the region of interest is appropriately defined, a baseline that is an intermediate line is located between the two boundary lines. Next, the first trace line generation process S202 and the second trace line generation process S204 are executed. They may be executed in parallel or in order. In FIG. 12, the specific contents of the first trace line generation process S202 are shown. In the second trace line generation process S204, the same configuration as that of the first trace line generation process is adopted. The description of the second line trace generation process S204 is omitted.

  In S206, the pixel value group in the upper sub-region of interest, that is, in the half ROI is referred to, thereby creating a whole histogram. That is, the entire sub-region of interest is referred to as the local region, not the belt region.

In S208, the temporary threshold a ₀ is calculated based on the whole histogram generated in S206. In S210, edge detection is performed on the search lines designated in order. In S212, the reference pixel values if it is determined that the tentative threshold a ₀ above, in S214, the point is identified as the reference edge point (first interval point). In S216, the second segment point is specified based on the reference edge point, that is, the first segment point. Such a process is repeated until it is determined in S218 that the process for the last line has been completed.

In S220, a pixel value group is referred to in a band region composed of a plurality of first sections configured as described above, and thereby a partial histogram is created. In S222, the threshold value a ₁ is determined based on the partial histogram. In S224, the edge searching is started on the search line which is sequentially designated, in S226, if a reference pixel values is determined to be the threshold value a ₁ or more, when the point is a boundary point at S228 Identified. Such a process is repeated until the final search line.

  With the first trace line generation process S202 as described above, it is possible to generate the first trace line with respect to the upper boundary line. On the other hand, the second trace line generation process S204 is performed using the same method as described above. A second trace line can be generated for the lower boundary line. In S232, the NT thickness is measured between the upper trace line and the lower trace line formed as described above. In that case, the NT thickness is measured at one or a plurality of positions in the x direction, and the maximum value, the minimum value, the average value, etc. are calculated from the plurality of NT thicknesses as necessary.

(4) Threshold Determination Method Next, a threshold determination method based on a histogram will be described with reference to FIGS. The threshold value determination method shown in each figure can be used in determining both the temporary threshold value and the main threshold value.

  In the first method shown in FIG. 13, a histogram is generated in S300, and the threshold value a is calculated directly from the histogram in S302. Then, using the threshold value a, edge detection (boundary detection) is performed on the original image in S304. As a method for determining the threshold value a in S302, a method described in the literature referred to in the description of the prior art can be used.

  In the second method shown in FIG. 14, a histogram is generated in S400, and the histogram is processed in S402. This processing includes smoothing processing, discretization processing on the horizontal axis, and the like, and further includes normalization processing for the vertical axis. In S404, the threshold value a is determined based on the processed histogram, and in S406, edge detection is performed on the original image using the threshold value a determined in this way.

  In the third method shown in FIG. 15, a histogram is generated in S500, and the histogram is processed in S502. The processing may include smoothing, discretization processing on the horizontal axis, normalization processing in the vertical axis direction, and the like as described above. In S504, the threshold value a is determined based on the processed histogram. On the other hand, in S506, a correction function for image processing is determined based on the histogram or the processed histogram. For example, when tone correction is performed, parameters α and β included in a function for tone correction are determined. In S508, tone correction processing is performed on the original image using the correction function determined as described above. In S510, other image processing is performed on the original image as necessary. Such image processing may include noise removal, labeling processing, and the like. In S512, an edge detection process is executed using the threshold value a determined in S504 for the original image that has undergone the above-described processes.

  FIG. 16 shows the fourth method. In S600, a histogram is generated, and in S602, histogram processing is executed. The processing process may include a normalization process and a discretization process. A plurality of feature amounts are extracted based on the processed histogram. That is, the processed histogram is analyzed from multiple angles. In the example shown in FIG. 16, each process of S604-S610 is performed.

  Specifically, in S604, the shape of the histogram is evaluated. Here, it is first judged whether it has a bimodal or unimodal configuration, and the latter is a short type with a narrow distribution in the luminance axis direction or a long type that extends widely in the same direction. Is determined. In S606, the data specific gravity is evaluated. Specifically, a peak in the histogram is specified, a frequency average value in a predetermined range on the front side of the peak is calculated, and a frequency average value in a predetermined range on the rear side of the peak is calculated. The frequency average value before the peak and the frequency average value after the peak are compared, and it is evaluated which is larger. The larger one corresponds to the specific emphasis. In S608, the luminance tendency for the histogram is evaluated. That is, it is determined whether the distribution of the histogram as a whole is biased toward the low luminance side, the biased toward the high luminance side, or the other (whether it can be classified into neither). In S610, the frequency tendency is evaluated. It evaluates the variance or spread of the histogram, and it is evaluated whether the locally concentrated area is low frequency, which can be said to be small, or whether the area spread throughout the horizontal axis direction is high. The

  By integrating the multifaceted analysis results as described above, specifically, by determining optimum threshold calculation conditions from the multifaceted analysis results, the threshold value a is determined based on the histogram as shown in S612. Is done. A specific example of this will be described below.

  FIG. 17 is a conceptual diagram showing the specific contents of S612 shown in FIG. In S604, the shape of the histogram is evaluated, and reference numeral 134 indicates bimodality, that is, U-type. Reference numeral 136 represents a unimodal short type. Reference numeral 138 indicates a unimodal long type. Each histogram is classified into one of the types. In S606, the data specific gravity is evaluated, and reference numeral 140 indicates a type in which the data specific gravity (specific weight) is before the peak. Reference numeral 142 indicates a type in which the data specific gravity (specific weight) is other than that before the peak. In S608, the luminance tendency is evaluated, the reference numeral 144 indicates a low luminance tendency, and the reference numeral 146 indicates other trends. In S610, the frequency tendency is evaluated. Reference numeral 148 indicates a low frequency type, and reference numeral 150 indicates a high frequency type. Further, reference numeral 152 indicates other types.

  The selection of threshold determination conditions in consideration of the above-described multifaceted analysis results is represented by a tree structure in FIG. That is, based on the analysis results of S604 to S610, one of the conditions 1 to 12 is selected in S612, and a threshold value is calculated from the histogram based on the selected condition. Of course, the method shown in FIG. 17 is an example, and other methods may be employed. In any case, it is desirable to evaluate the feature amount of the histogram from a plurality of viewpoints, select the optimum condition for the histogram, and calculate the threshold value accordingly.

  DESCRIPTION OF SYMBOLS 10 Probe, 12 Transmission part, 14 Reception part, 16 Signal processing part, 17 Image formation part, 18 Display processing part, 22 Measurement part, 30 Front wall, 32 Rear wall, 34 Blood flow part, 56,60 Region of interest, 170 Fetal cervical edema (NT), 178 Complex region of interest.

Claims (18)

A region-of-interest setting means for setting a region of interest including at least one target boundary line for a target tissue on a tomographic image obtained by transmission and reception of ultrasonic waves;
Within the region of interest, a reference point is searched for on a search line extending in a search direction intersecting the base line from each point on the base line of the region of interest, and based on the reference point on each search line A section setting means for setting a section and thereby forming a band region as a set of a plurality of sections;
Determining means for determining an image processing parameter for detecting the target boundary line by referring to a plurality of pixel values belonging to the band region;
An ultrasonic diagnostic apparatus comprising: The apparatus of claim 1.
The determining means includes
A histogram generator that generates a histogram based on a plurality of pixel values belonging to the band region;
A threshold determining unit that determines a threshold as the image processing parameter based on the histogram;
An ultrasonic diagnostic apparatus comprising: The apparatus according to claim 1 or 2,
The section setting means includes
A reference point detection unit that sequentially refers to pixel values from the base line on each search line and detects the reference points by comparing the pixel values with a temporary threshold;
A section setting unit that sets the section based on the reference point on each search line;
An ultrasonic diagnostic apparatus comprising: The apparatus of claim 3.
The ultrasonic diagnostic apparatus, wherein the temporary threshold is a threshold for detecting a point on the near side of the target boundary line. The apparatus of claim 4.
The ultrasonic diagnostic apparatus, wherein the section setting unit determines the reference point as one end of the section, and determines a point away from the reference point forward in the search direction as the other end of the section. The apparatus of claim 4.
The section setting unit determines a point away from the reference point forward in the search direction as one end of the section, and determines a point further away from the reference point forward in the search direction as the other end of the section. Ultrasonic diagnostic equipment. The device according to any one of claims 1 to 6,
An ultrasonic diagnostic apparatus comprising input means for designating a position and angle of the region of interest on the tomographic image by a user. The device according to any one of claims 1 to 7,
The region of interest is a rectangular region;
The upper side or the lower side in the region of interest is the baseline,
Each search line is set in parallel with the left and right sides of the region of interest.
An ultrasonic diagnostic apparatus. The apparatus of claim 8.
The ultrasonic diagnostic apparatus according to claim 1, wherein the at least one target boundary line includes an inner boundary line and an outer boundary line in a blood vessel wall. The device according to any one of claims 1 to 7,
The region of interest is composed of an upper sub-region of interest and a lower sub-region of interest connected via an intermediate line that functions as the baseline,
Detection of the boundary of interest is performed in each of the sub-regions of interest,
An ultrasonic diagnostic apparatus. The apparatus of claim 10.
The at least one attention boundary includes an upper boundary and a lower boundary representing the outline of fetal cervical edema;
The upper boundary is detected in the upper sub-region of interest;
The lower boundary is detected within the lower sub-region of interest;
An ultrasonic diagnostic apparatus. The apparatus of claim 2.
The threshold determination unit includes a histogram analysis unit that calculates a plurality of feature amounts from the histogram,
The ultrasonic diagnostic apparatus, wherein the threshold is determined using the plurality of feature amounts. The apparatus of claim 12.
The plurality of feature amounts include a feature amount that is defined as a type to which the shape of the histogram belongs, a feature amount that is defined from the magnitude relationship of average luminance before and after the peak in the histogram, and a bias on the luminance axis of the histogram. An ultrasonic diagnostic apparatus comprising at least one of a defined feature quantity and a feature quantity defined from a spread degree of the histogram. Region of interest setting means for setting a region of interest including an inner boundary line and an outer boundary line in a blood vessel wall on a tomographic image obtained by transmitting and receiving ultrasonic waves;
Within the region of interest, a reference point is searched for on a search line extending in a search direction intersecting the base line from each point on the base line of the region of interest, and based on the reference point on each search line A section setting means for setting a first section and a second section, thereby configuring a first band area as a set of a plurality of first sections, and configuring a second band area as a set of a plurality of second sections;
An inner threshold value determining means for determining an inner threshold value for detecting the inner boundary line by referring to a plurality of pixel values belonging to the first band region;
An outer threshold determining means for determining an outer threshold for detecting the outer boundary line by referring to a plurality of pixel values belonging to the second band region;
An ultrasonic diagnostic apparatus comprising: The apparatus of claim 14.
Means for generating an inner trace line along the inner boundary line;
Means for generating an outer trace line along the outer boundary line;
Means for measuring the thickness of the vessel wall between the inner trace line and the outer trace line;
An ultrasonic diagnostic apparatus comprising: In the tomographic image obtained by transducing the ultrasonic, and the lower sub-region of interest including the lower boundary of the fetus after cervical edema and upper sub-region of interest including the upper side borders of the fetus after cervical edema A region of interest setting means for setting a composite region of interest consisting of:
In the upper sub-region of interest, the upper reference point is searched on the upper search line extending in the upper search direction intersecting the intermediate baseline from each point on the intermediate baseline of the composite region of interest, and On the search line, an upper section is set based on the upper reference point, whereby an upper section setting means for configuring an upper belt area as a set of a plurality of upper sections,
In the lower sub-region of interest, the lower reference point is searched for on the lower search line extending from each point on the intermediate baseline of the composite region of interest to the lower search direction intersecting the intermediate baseline. A lower section setting means for setting a lower section based on the lower reference point on each lower search line, thereby forming a lower band region as a set of a plurality of lower sections,
An upper threshold determining means for determining an upper threshold for detecting the upper boundary line by referring to a plurality of pixel values belonging to the upper band region;
A lower threshold value determining means for determining a lower threshold value for detecting the lower boundary line by referring to a plurality of pixel values belonging to the lower band region;
An ultrasonic diagnostic apparatus comprising: The apparatus of claim 16.
Means for generating an upper trace line along the upper boundary line;
Means for generating a lower trace line along the lower boundary line;
Means for measuring the thickness of the posterior cervical edema between the upper trace line and the lower trace line;
An ultrasonic diagnostic apparatus comprising: A program executed in an ultrasonic diagnostic apparatus or an information processing apparatus that processes an ultrasonic image,
A function of setting a region of interest including at least one attention boundary for a target tissue based on a user designation on an ultrasound image;
Within the region of interest, a reference point is searched for on a search line extending in a search direction intersecting the base line from each point on the base line of the region of interest, and based on the reference point on each search line A function of setting a zone and thereby forming a belt region as a set of a plurality of zones arranged in the direction of the baseline;
A function of determining a threshold for detecting the boundary line of interest by referring to a plurality of pixel values belonging to the band region;
On the ultrasonic image, a function of detecting the boundary of interest using the determined threshold value;
The program characterized by including.

Images