JP2006230910A - Image processor and image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enhance the precision of detection of an image area as a detection object in an image to be processed. <P>SOLUTION: This image processor 1 computes a Shape Index, which serves as the amount of characteristic, indicating the curved shape of the density distribution of a medical image, at each pixel of the medical image to be processed, by using a curvature filter, and creates the image (Shape Index image) which emphasizes the unevenness of the density distribution. An abnormal shadow candidate area is detected by binarizing the Shape Index image through the use of a threshold of the preset Shape Index, and the image area of an abnormal shadow is detected by deleting the false-positive image area from the abnormal shadow candidate area on the basis of the amount of the characteristic of the detected abnormal shadow candidate area. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus and an image processing method.

  In the medical field, digitization of medical images has been realized, medical image data generated by a CR (Computed Radiography) device or the like is displayed on a display monitor, and a doctor interprets the medical image displayed on the display monitor. Diagnosis is made by observing the state of the lesion and changes over time.

  Conventionally, for the purpose of reducing the burden on doctors' interpretation, computer diagnostic support apparatus that automatically detects the shadow of a lesion appearing on an image as an abnormal shadow candidate by performing image processing on the medical image data ( A medical image processing apparatus called Computed-Aided Diagnosis (hereinafter referred to as CAD) has been developed.

  The shadow of a lesion often has a characteristic density distribution, and the CAD detects an image area estimated as a lesion based on such density characteristics as an abnormal shadow candidate area. For example, tumors and microcalcification clusters can be mentioned as characteristic features of the cancerous part of breast cancer. On a medical image (which is called mammography) of the breast, the mass shadow shows a density change close to a Gaussian distribution. On the other hand, micro-calcified clusters are formed by collecting (clustered) micro-calcified portions and appear as whitish-round shadows having a substantially conical density change on mammography.

In the CAD, various detection algorithms have been developed according to the type of lesion to be detected, and as an optimal algorithm for detecting a tumor shadow, a method using an iris filter has been proposed (for example, a patent) Reference 1 and Patent Document 2). In addition, as an optimal algorithm for detecting a microcalcification cluster shadow, a method using a morphologic filter has been proposed.
Japanese Patent Application Laid-Open No. 8-263364 Japanese Patent Laid-Open No. 10-91758

  However, the iris filter generally reacts strongly to a region that is round and has a lower concentration than the surrounding area, such as a shadow of a tumor, so that a mammary gland with a high degree of circularity and a thickness appears on the image. In some cases, the mammary gland of the normal tissue is erroneously detected as an abnormal shadow. Similarly, when a high-frequency signal such as noise is generated on the image, it is difficult to discriminate because the micro-calcification cluster shadow and the noise signal change are similar, and the noise region is erroneously detected. Sometimes.

  When the image signal is viewed as a three-dimensional signal composed of an image position (two-dimensional coordinates) and a density component, the image signal forms a curved surface indicating a density distribution. As described above, since the abnormal shadow shows a change in density such as a conical structure, the image region of the abnormal shadow should form a characteristic curved surface. However, a method for detecting an abnormal shadow candidate based on density characteristics, such as detection based on the concentration degree of the density gradient on a pixel of interest such as an iris filter, has been developed. A method that takes into account the curved surface shape of the shadow when viewed as a curved surface representing the above has not yet been proposed.

  The subject of this invention is improving the detection accuracy which detects the image area | region made into a detection target in a medical image.

  According to the first aspect of the present invention, the first feature amount calculating means for calculating the feature amount representing the shape of the curved surface indicating the density distribution of the medical image, and the feature amount calculated by the first feature amount calculating means. A detecting unit for detecting an abnormal shadow candidate region from the medical image, a second feature amount calculating unit for calculating a feature amount of the abnormal shadow candidate region detected by the detecting unit, and the second feature amount. False positive deletion means for deleting false positives of the abnormal shadow candidate region based on the feature amount calculated by the calculation means.

  According to a second aspect of the present invention, in the image processing apparatus according to the first aspect, the first feature amount calculating means calculates a shape index as a feature amount representing the shape of the curved surface.

  According to a third aspect of the present invention, in the image processing apparatus according to the first or second aspect, the feature amount calculated by the second feature amount calculating means includes a shape index.

According to a fourth aspect of the present invention, in the image processing apparatus according to any one of the first to third aspects, the detection unit is based on the feature amount calculated by the first feature amount calculation unit. By detecting the abnormal shadow candidate area by binarizing the medical image using a threshold value of the feature amount set in advance,
The false positive deleting means deletes a false positive image region from the abnormal shadow candidate region obtained by the binarization.

  Invention of Claim 5 is provided with the display means which displays the image obtained by deletion of the false positive by the said false positive deletion means in the image processing apparatus as described in any one of Claims 1-4. It is characterized by.

  According to a sixth aspect of the present invention, in the image processing device according to any one of the first to fifth aspects, an arbitrary target pixel in a medical image is set, and an image region within a predetermined range from the target pixel is set. Setting means for setting is provided, wherein the first feature value calculating means calculates a feature value representing a curved surface shape of a density distribution in the set image area.

  A seventh aspect of the present invention is the image processing apparatus according to any one of the first to sixth aspects, wherein a normal plane is defined with a normal line at a target pixel on a curved surface indicating a density distribution of the medical image as a rotation axis. A function calculating unit that rotates a predetermined angle and calculates a function that approximates a curve obtained by cutting out a curved surface of an image area within a predetermined range from a target pixel by a normal plane for each rotation angle; The feature quantity calculating means calculates the feature quantity based on the approximate function calculated by the function calculating means.

  The invention according to claim 8 is the image processing apparatus according to claim 7, wherein the function calculating means calculates a function approximating the curve using a least square method.

  According to a ninth aspect of the present invention, in the image processing apparatus according to the seventh or eighth aspect, the function calculating unit calculates an approximate circle as the approximate function, and the first feature amount calculating unit includes a rotation angle. Each time, the curvature of the pixel of interest of the curve is calculated from the calculated radius of the approximate circle, and the feature amount is calculated based on the curvature calculated for each rotation angle.

  According to a tenth aspect of the present invention, in the image processing apparatus according to any one of the first to ninth aspects, the first feature amount calculating unit calculates a feature amount for all pixels of the medical image. It is characterized by that.

  The invention according to claim 11 is the image processing apparatus according to any one of claims 1 to 10, wherein the false positive deletion means is an image whose size is out of a predetermined range in the abnormal shadow candidate region. It is characterized in that the area is deleted as a false positive image area.

  The invention according to claim 12 is the image processing apparatus according to any one of claims 1 to 10, wherein the false positive deletion unit is an image having a shape other than a predetermined shape in the abnormal shadow candidate region. It is characterized in that the area is deleted as a false positive image area.

  According to a thirteenth aspect of the present invention, in the image processing device according to any one of the first to twelfth aspects, the false positive deletion unit is calculated by the first and / or second feature amount calculation unit. Using a multivariate analysis that uses as input data the various feature quantities of the medical image including the feature quantities that are included and uses the index value indicating the degree of true-positive abnormal shadows as output data. It is a feature.

  The invention according to claim 14 is the image processing apparatus according to any one of claims 1 to 13, wherein the first feature amount calculating unit changes a region range of the curved surface for calculating the feature amount, The feature amount is calculated in the changed region range.

  According to a fifteenth aspect of the present invention, in the image processing device according to any one of the first to thirteenth aspects, the first feature amount calculating unit changes a resolution of the medical image and changes the resolution. The feature is that the feature amount is calculated by the resolution.

  According to a sixteenth aspect of the present invention, in the image processing apparatus according to any one of the first to fifteenth aspects, a Gaussian distribution-like curved surface is formed as an abnormal shadow from the false positive deletion by the false positive deletion means. It is characterized in that an image area is detected.

  According to a seventeenth aspect of the present invention, in the image processing apparatus according to the sixteenth aspect, an image area of a tumor shadow is detected as an abnormal shadow by the false positive deletion by the false positive deletion means.

The invention according to claim 18 is a first feature amount calculating step for calculating a feature amount representing a shape of a curved surface indicating a density distribution of a medical image;
A detection step of detecting an abnormal shadow candidate region from the medical image using the feature amount calculated in the first feature amount calculation step;
A second feature amount calculation step of calculating a feature amount of the abnormal shadow candidate region detected in the detection step;
And a false positive deletion step of deleting a false positive of the abnormal shadow candidate region based on the feature amount calculated in the second feature amount calculation step.

  According to a nineteenth aspect of the present invention, in the image processing method according to the eighteenth aspect, in the first feature amount calculating step, a shape index is calculated as a feature amount representing the shape of the curved surface. .

  According to a twentieth aspect of the present invention, in the image processing method according to the twentieth or nineteenth aspect, the feature amount calculated in the second feature amount calculating step includes a shape index.

  The invention according to claim 21 is the image processing method according to any one of claims 18 to 20, wherein, in the detection step, based on the feature amount calculated in the first feature amount calculation step, An abnormal shadow candidate region is detected by binarizing the medical image using a threshold value of the feature amount set in advance. In the false positive deletion step, the abnormal shadow candidate region obtained by the binarization is detected. It is characterized in that false positive image regions are deleted from.

  The invention according to claim 22 includes a display step of displaying an image obtained by deleting false positives by the false positive deletion step in the image processing method according to any one of claims 18 to 21. It is characterized by.

  According to a twenty-third aspect of the present invention, in the image processing method according to any one of the eighteenth to twenty-second aspects, an arbitrary target pixel in a medical image is set, and an image region within a predetermined range from the target pixel is set. Including a setting step for setting, and in the first feature amount calculating step, a feature amount representing a curved surface shape of a density distribution in the set image region is calculated.

  The invention according to claim 24 is the image processing method according to any one of claims 18 to 23, wherein a normal plane is defined with a normal line at a target pixel on a curved surface indicating a density distribution of the medical image as a rotation axis. A function calculating step of calculating a function approximating a curve obtained by rotating a curved surface of an image area within a predetermined range from a target pixel by a normal plane for each rotation angle; In the feature amount calculating step, the feature amount is calculated based on the approximate function calculated in the function calculating step.

  According to a twenty-fifth aspect of the present invention, in the image processing method according to the twenty-fourth aspect, in the function calculating step, a function that approximates the curve is calculated using a least square method.

  According to a twenty-sixth aspect of the present invention, in the image processing method according to the twenty-fourth or twenty-fifth aspect, an approximate circle is calculated as the approximate function in the function calculating step, and a rotation angle is calculated in the first feature amount calculating step. Each time, the curvature of the target pixel of the curve is calculated from the calculated radius of the approximate circle, and the feature amount is calculated based on the curvature calculated for each rotation angle.

  According to a twenty-seventh aspect of the present invention, in the image processing method according to any one of the eighteenth to twenty-sixth aspects, in the first feature amount calculating step, feature amounts are calculated for all pixels of the medical image. It is characterized by that.

  The invention according to claim 28 is the image processing method according to any one of claims 18 to 27, wherein, in the false positive deletion step, an image whose size is out of a predetermined range in the abnormal shadow candidate region. The region is deleted as a false positive image region.

  The invention according to claim 29 is the image processing method according to any one of claims 18 to 27, wherein, in the false positive deletion step, an image having a shape other than a predetermined shape in the abnormal shadow candidate region. The region is deleted as a false positive image region.

  The invention according to claim 30 is the image processing method according to any one of claims 18 to 29, wherein the false positive deletion step is calculated in the first and / or second feature amount calculation step. False positive image regions are deleted using multivariate analysis with various feature values of the medical image including the feature values as input data and index values indicating the degree of true positive abnormal shadows as output data. It is characterized by.

  The invention according to claim 31 is the image processing method according to any one of claims 18 to 30, wherein, in the first feature amount calculating step, a region range of the curved surface for calculating the feature amount is changed, The feature amount is calculated in the changed region range.

  According to a thirty-second aspect of the present invention, in the image processing method according to any one of the eighteenth to thirty-third aspects, in the first feature amount calculating step, the resolution of the medical image is changed and changed. It is characterized in that the feature amount is calculated with the resolution.

  According to a thirty-third aspect of the present invention, in the image processing method according to any one of the eighteenth to thirty-second aspects, a Gaussian-distributed curved surface is formed as an abnormal shadow from the false positive deletion in the false positive deletion step. It is characterized in that an image area is detected.

  According to a thirty-fourth aspect of the present invention, in the image processing method according to the thirty-third aspect, an image area of a tumor shadow is detected as an abnormal shadow from the false positive deletion in the false positive deletion step.

  According to the present invention, the feature amount representing the shape of the curved surface of the density distribution of the medical image is calculated, the abnormal shadow candidate region is detected from the medical image based on the calculated feature amount, and the detected abnormal shadow candidate region is detected. By deleting false positives of the abnormal shadow candidate area based on the feature amount of the abnormal shadow, it is possible to accurately detect the abnormal shadow image area having a characteristic density distribution by distinguishing it from the normal tissue, improving the detection accuracy Can be made.

  In particular, based on the feature amount calculated by the first feature amount calculation means, the abnormal image candidate region is detected by binarizing the medical image using a preset threshold value of the feature amount, and binary By deleting the false-positive image area from the abnormal shadow candidate area obtained by the conversion, it becomes possible to detect the abnormal shadow image area more reliably.

  In addition, by deleting an image region whose size (area, radius, etc.) is outside a predetermined range and / or an image region other than a predetermined shape (for example, a circle) from the abnormal shadow candidate regions as a false positive image region. Thus, it is possible to detect the image area of the abnormal shadow more accurately.

  Furthermore, it is possible to further improve the detection accuracy of abnormal shadows by deleting false positive image regions using multivariate analysis.

  In addition, by making it possible to change the range of the curved surface for calculating the feature amount or the resolution of the medical image, it is possible to accurately detect the image region of the abnormal shadow regardless of the curvature of the curved surface indicated by the density distribution. It becomes possible.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

First, the configuration in the present embodiment will be described.
FIG. 1 shows a main part configuration of an image processing apparatus 1 according to an embodiment of the present invention. As shown in FIG. 1, the image processing apparatus 1 includes a CPU (Central Processing Unit) 2, an I / F (InterFace) 3, an operation unit 4, a display unit 5, a communication unit 6, a RAM (Random Access Memory) 7, and a storage. The apparatus 8 and the program memory 9 are included.

  The CPU 2 develops various programs such as a system program and an abnormal shadow detection processing program stored in the program memory 9 in the RAM 7, and controls the operation of each unit of the image processing apparatus 1 in cooperation with these programs.

  The CPU 2 uses a curvature filter to describe a shape index (hereinafter, “Shape Index”) that is a feature amount representing the shape of the curved surface of the density distribution of the medical image signal for each pixel of the processing target image (medical image). .) And curvature filter image conversion processing (see FIG. 8) for creating an image (Shape Index image) in which the unevenness of the density distribution is emphasized is executed.

  In the curvature filter image conversion process, the CPU 2 selects a pixel of interest on a curved surface indicating a density distribution of a medical image signal composed of signal components in three directions of image position (two-dimensional coordinates) and density, and a predetermined centering on the pixel of interest. Set the image area within the range, calculate the curvature of the curved surface in the set image area using the least square method using the curvature filter having the size of the predetermined range, and from the calculated curvature Then, a Shape Index that is a feature amount of the curved surface is calculated. The CPU 2 calculates the Shape Index of each pixel by causing the above-described curvature filter to scan the entire processing target image. A feature amount calculation method using the curvature filter of the present embodiment will be described in detail later.

  Further, the CPU 2 binarizes the shape index image obtained by the curvature filter image conversion processing using a preset shape index threshold value, so that an abnormal shadow candidate region (hereinafter simply referred to as “candidate region”) is obtained. ) Is detected. Then, by deleting the false positive image area from the candidate area obtained by binarization, an abnormal shadow image area is detected (see FIG. 7).

  In the process of deleting the false positive image area, the CPU 2 deletes the image area whose size (area, radius, etc.) is out of the predetermined range and / or image area other than the predetermined shape among the candidate areas. For example, in the case of a mass shadow, since there are many circular shapes with a diameter of about 5 mm to 30 mm, the CPU 2 calculates a circularity (a value normalized to 0 to 1) indicating the circularity, and the circularity is An image area that is lower than a preset threshold and has a diameter of less than 5 mm is deleted.

  In addition, the CPU 2 uses multivariate analysis to calculate an index value indicating the degree of true-positive abnormal shadow for each candidate region (an index value indicating how much the true-positive abnormal shadow has characteristics). Then, based on the calculated index value, it is determined whether or not it is false positive, and the image area determined to be false positive is deleted. In multivariate analysis, multivariate analysis is constructed so that various feature amounts such as circularity, contrast, texture, and shape index are input data, and an index value indicating the degree of true positive abnormal shadow is output data. As a method of multivariate analysis, any one of an artificial neural network, principal component analysis, discriminant analysis, or the like may be used, or a method other than these may be used.

  Further, the CPU 2 performs various image processes on the input medical image signal before executing the curvature filter image conversion process. In this various image processing, gradation processing for adjusting contrast, density gradation in low density regions of breasts and tumors where contrast tends to be small are expanded, and conversely, there is little possibility that images of microcalcification clusters exist Contrast correction processing that performs correction to compress the density gradation of the fat area, unsharpness mask processing that adjusts the sharpness of the image, and within a density range that is easy to see without reducing the contrast of details of subjects with a wide dynamic range Dynamic range compression processing and the like.

  The I / F 3 is an interface for connecting to the image generation apparatus G. The medical image signal generated by the image generation device G is input to the image processing device 1 via the I / F 3.

  As the image generation device G, for example, a film on which a medical image is recorded is recorded on the film by a laser digitizer that scans a laser beam and reads a medical image signal, or a sensor that includes a photoelectric conversion element such as a CCD (Charge Coupled Device). A film scanner or the like that reads the medical image signal that has been processed can be applied.

  Also, instead of reading a medical image recorded on a film, an imaging device that captures a medical image using a stimulable phosphor, and a radiation detection element and a capacitor that generate charges according to the intensity of irradiated radiation A flat panel detector or the like can be connected, and the method for inputting the medical image signal is not particularly limited.

  The operation unit 4 includes a keyboard composed of cursor keys, numeric keys, various function keys, and the like, and outputs an operation signal corresponding to a key pressing operation to the CPU 2. The operation unit 4 may include an input device such as a mouse or a touch panel as necessary.

  The display unit 5 includes a display screen such as an LCD (Liquid Crystal Display), and displays various display information according to a display control signal input from the CPU 2.

  The communication unit 6 includes a communication interface such as a network interface card, a modem, and a terminal adapter, and transmits and receives various types of information to and from external devices on the communication network. For example, it is good also as a structure which receives a medical image signal from the image generation apparatus G via the communication part 6, and connects to the server in a hospital via the communication part 6, etc., or the medical terminal installed in each clinic It is good also as a structure which connects to and transmits an image processing result.

  The RAM 7 forms a work area for temporarily storing various programs executed by the CPU 2 and data processed by these programs.

  The storage device 8 is a memory that stores data processed by the CPU 2 and includes a feature amount file 81. The feature amount file 81 stores a curved surface feature amount (Shape Index in this embodiment) representing the density distribution of the medical image signal calculated by the CPU 2.

  The program memory 9 stores various processing programs such as a system program and an abnormal shadow detection processing program.

  Next, a feature amount calculation method using the curvature filter of the present embodiment will be described.

  FIG. 2A shows a curved surface E representing an image position represented by two-dimensional coordinates and a density distribution of a medical image signal composed of signal components in three directions of density. In FIG. 2A, an arbitrary pixel on the curved surface E is set as the target pixel p, a plane spanned by the normal vector m and the tangent vector t at the target pixel p is defined as a normal plane F, The intersecting line with the curved surface E (that is, the curved surface E cut by the normal plane F) is defined as a normal section J.

  In FIG. 2A, for convenience of explanation, the curved surface E is shown as a smooth curved surface. However, since the digital image is actually handled, the curved surface E has a discrete density for each pixel as shown in FIG. It has a staircase pattern indicating the value (pixel signal value).

  In order to calculate the curvature at the target pixel p, as shown in FIG. 2B, the normal plane F (or tangent vector t) is rotated by a predetermined angle β with the normal line at the target pixel p as the rotation axis. For each rotation angle, the normal curvature at the target pixel p of the curve γ represented by the normal section J may be calculated. Here, the predetermined angle β is determined based on the processing speed of the image processing apparatus 1, and can be set to π / 2, π / 8, or the like, for example. When the predetermined angle β is increased, the calculation time can be shortened.

  The normal curvature at the target pixel p on the normal section J is obtained by calculating a function that approximates the curve γ represented by the normal section J by the least square method, and calculating the curvature at the target pixel p of the calculated approximate function. It is done. In the present embodiment, as shown in FIG. 4, the curve γ represented by the normal section J (concentration profile in FIG. 4) is approximated by a circle. Hereinafter, a process of calculating an approximate circle of the curve γ (a circle that approximates the curve γ) using the least square method will be described.

次に、曲線γ上における注目画素ｐの近傍（所定範囲内の画像領域）のｎ個の画素信号値を（Ｘ_i,Ｙ_i）（i＝1,…,n）とし、式（２）に示すように、ｎ次元ベクトルＡ、ｎ行３列の行列Ｂ、３次元ベクトルＣを定義する。

Assuming that the center coordinates of the approximate circle are (a, b) and the radius is r, the approximate circle is expressed as in equation (1) using two-dimensional coordinates (X, Y).

Next, n pixel signal values in the vicinity (image region within a predetermined range) of the target pixel p on the curve γ are defined as (X _i , Y _i ) (i = 1,..., N), and the equation (2) As shown, an n-dimensional vector A, an n-by-3 matrix B, and a 3-dimensional vector C are defined.

In order to approximate the curve γ with the circle of the expression (1), C that minimizes L = | A−BC | ² may be obtained. That is, it is only necessary to obtain C such that L is partially differentiated by the vector C and becomes zero. C satisfying ∂L / ∂C = 0 is expressed by Equation (3).
C = (B ^T B) ^-1 BA (3)
In Expression (3), T represents a transposed matrix. (B ^T B) ⁻¹ B is a pseudo inverse matrix of B.

A curve γ represented by the normal section J is approximated by a circle determined by C satisfying the expression (3). The reciprocal of the radius of this approximate circle is the normal curvature of the target pixel p of the curve γ. That is, when the radius of the approximate circle of the curve γ at the rotation angle θ is r (θ), the normal curvature k (θ) at the rotation angle θ is expressed as shown in Expression (4).
k (θ) = 1 / r (θ) (4)
When the center coordinates (a, b) of the approximate circle are above the curved surface E, the sign of the right side of Equation (4) is negative, and the center coordinates (a, b) of the approximate circle are below the curved surface E. In this case, the sign on the right side of Equation (4) is positive.

式（５）のShape Index ＳＩは、注目画素ｐを中心とした所定範囲内の画像領域における曲面Ｅの曲面形状を表す。 When the normal plane F is rotated about the normal line of the pixel of interest p as the rotation axis, the shape of the normal section J (curve γ) changes, so the value of the normal curvature k (θ) also changes according to the rotation angle θ. become. That is, the normal curvature k (θ) has a maximum value and a minimum value depending on the rotation angle θ of the normal plane F. Of the normal curvature k (θ) calculated at each rotation angle θ (0 ≦ θ <π), assuming that the maximum value is k _max and the minimum value is _kmin , the Shape Index SI is as shown in Equation (5). Defined.

Shape Index SI in Expression (5) represents the curved surface shape of the curved surface E in an image region within a predetermined range centered on the pixel of interest p.

  FIG. 5 shows a curved surface shape corresponding to the value of Shape Index. In the white part of the image to be processed (that is, the part having a low density signal value), the value of Shape Index approaches 1, and the curved surface shape is concave. On the other hand, in the black portion of the processing target image (that is, the portion where the signal value of density is high), the value of Shape Index approaches 0 and the curved surface shape is convex.

  FIG. 6 shows a curvature filter and a processing target image (medical image) of this embodiment. As shown in FIG. 6, by scanning this curvature filter over the entire processing target image, the shape index value at each pixel is calculated, and a shape index image in which the unevenness of the density distribution is emphasized is created.

  In this embodiment, the curved surface shape of the density distribution when the image is output to the film is handled. However, the curved surface shape of the luminance distribution when the image is output to the display monitor may be handled. In the case of the luminance distribution, the relationship between the shape index value and the curved surface shape is opposite to that in the case of the density distribution. That is, in the case of luminance, the signal value is white when the signal value is high, and black when the signal value is low. Accordingly, in the case of the luminance distribution, when the Shape Index approaches 1, the curved surface shape becomes convex, and when the Shape Index approaches 0, the curved surface shape becomes concave.

Next, the operation in this embodiment will be described.
As an operation in the present embodiment, an abnormal shadow detection process executed in the image processing apparatus 1 will be described with reference to a flowchart of FIG.

  First, when a medical image signal is input from the image generation device G via the I / F 3, as a pre-process of the abnormal shadow detection process, a gradation conversion process, an eye sharpness process, a dynamic process is performed on the input medical image signal. Various image processing such as range compression processing is performed (step S1).

  Next, for each pixel of the pre-processed medical image signal, a shape index that represents the shape of the curved surface of the density distribution of the medical image signal is calculated, and an image in which the unevenness of the density distribution is emphasized (shape index image) ) Is generated (step S2). The curvature filter image conversion process in step S2 will be described in detail later with reference to FIG.

  Next, the shape index image obtained by the curvature filter image conversion process is binarized using a preset shape index threshold value, so that an abnormal shadow candidate region (hereinafter referred to as an abnormal shadow candidate region) is obtained. Is detected (step S3).

  For example, since a tumor tends to be a concave with a gentle Gaussian distribution, an image region having a Shape Index value of 0.75 to 1.00 is a tumor shadow candidate region, as shown in FIG. Therefore, when a tumor shadow candidate region is detected as an abnormal shadow candidate region, in step S3, a binarization threshold is set to 0.75, an image region having a Shape Index value of 0.75 or more is set as a white region, and the Shape Index is selected. Binarization is realized with an image area having a value of less than 0.75 as a black area, and a white area is detected as a tumor shadow candidate area.

  Next, the abnormal shadow candidate areas obtained by the binarization in step S3 are labeled (numbered) (step S4), and the false positive image areas are deleted from each labeled abnormal shadow candidate area. Are detected (step S5).

  When the abnormal shadow image area is detected, the detection result is displayed on the display unit 5 (step S6), and the abnormal shadow detection process is terminated. In step S6, specifically, a medical image input from the image generating device G is displayed on the display unit 5, and an abnormal shadow image area is identified by an arrow or color display on the medical image. Is displayed.

  Next, the curvature filter image conversion process (step S2 in FIG. 7) will be described with reference to the flowchart in FIG.

  First, the target pixel p is set in the pre-processed medical image signal (step S10), and an image region within a predetermined range centered on the target pixel p is set (step S11). Next, the rotation angle θ with the normal line of the target pixel p as the rotation axis is set to 0 (step S12).

  Next, a curved surface E indicating the density distribution in the image area set in step S11 is cut out by the normal plane F of the rotation speed θ, thereby extracting the curve γ represented by the normal section J of the rotation angle θ (step S13). The pixel signal value (density value) on the curve γ is stored in the storage device 8. Next, using the least square method, the curve γ at the rotation angle θ is approximated by a circle, and the normal curvature k (θ) at the rotation angle θ of the pixel of interest p is calculated from the radius of the approximate circle (step S14).

  Next, a value obtained by adding β to the rotation angle θ is set as a new rotation angle θ (step S15). Next, it is determined whether or not the current rotation angle β is greater than or equal to π (step S16).

When it is determined in step S16 that the rotation angle θ is less than π (step S16; NO), the process returns to step S13, and the processes of steps S13 to S16 for the rotation angle θ are repeated. When it is determined in step S16 that the rotation angle θ is equal to or greater than π (step S16; YES), the maximum value k _max and the minimum value k _min are selected from the normal curvatures k (θ) calculated for each rotation angle. Is calculated (step S17).

Next, using the maximum value k _max and the minimum value k _min of the normal curvature calculated in step S17, a shape index is calculated by equation (5) (step S18), and the calculated shape index value is the current value. The feature amount file 81 stores the feature amount of the pixel of interest p.

  Next, it is determined whether or not the shape index calculation has been completed for all pixels of the processing target image (step S19). If it is determined in step S19 that the calculation of Shape Index has not been completed for all pixels (step S19; NO), the process returns to step S10, and the next pixel of interest p is set in the processing target image (step S10). The processes of steps S11 to S18 are repeated for the set target pixel p. If it is determined in step S19 that the shape index calculation has been completed for all the pixels (step S19; YES), a shape index image in which the unevenness of the density distribution is emphasized is created (step S20), and the curvature filter image conversion is performed. The process ends.

  In the curvature filter image conversion process of FIG. 8, by changing the size of the curvature filter or changing the resolution of the processing target image, it is possible to deal with density distributions having various sizes of curvature.

Hereinafter, a change in the size of the curvature filter will be described.
When the size of the curvature filter is changed, the range of the image signal on the curve γ represented by the normal section J is changed to the target of the approximate circle. That is, when the size of the curvature filter is changed, the radius of the approximate circle changes, and thus the calculated normal curvature k (θ) also changes.

  As shown in FIG. 9, when the parameter n indicating the size of the curvature filter (that is, the range in which the curve γ is approximated by a circle) is changed to 3, 5, 7,..., The number of signals used for approximation increases. Therefore, as shown in FIG. 10, the size of the approximate circle changes. The curvature is an index value indicating the degree of curvature of the curved surface. The smaller the radius rn (θ) of the approximate circle, the larger the curvature and the greater the degree of curvature. Therefore, as shown in FIG. 9, since the radius of the approximate circle increases as the mask size n × n indicating the size of the curvature filter is increased, the shape of the curve γ centered on the target pixel p is the target pixel p. It can be inferred that as the distance from the distance increases, the curve becomes gentler.

Thus, by using the curvature filter mask size n × n, is calculated radius rn of the approximate circle curve gamma (theta) is normal curvature k _n at each rotational speed θ (θ) = 1 / rn (θ ) Is calculated from the maximum value and the minimum value, and a shape index image is created.

  As described above, according to the image processing apparatus 1 of the present embodiment, the Shape Index is calculated as a feature amount representing the shape of the curved surface of the density distribution of the medical image signal for each pixel of the medical image, and the calculated Shape is calculated. By detecting the abnormal shadow candidate area by binarizing the medical image based on the index value, the false positive image area is deleted from the abnormal shadow candidate area based on the feature amount of the detected abnormal shadow candidate area. Thus, by detecting the abnormal shadow image region, it is possible to accurately detect the abnormal shadow image region having a characteristic density distribution by distinguishing it from the normal tissue.

  In particular, by calculating the Shape Index as a feature value representing the curved surface shape of the density distribution, it is possible to easily and reliably detect the image area of the concave Gaussian distribution corresponding to the tumor shadow candidate, and improve the processing speed. be able to.

  For example, in mammography, even a linear normal tissue such as the mammary gland has various thicknesses from thin to thick, so the shadow of the mammary gland tissue that has become thick and massive has a high roundness mass or roundness. In some cases, it is difficult to distinguish from the shadow of a microcalcification cluster having a spread. However, by calculating the shape index of the curved surface represented by the concentration distribution, it is possible to prevent false positive shadows such as linear normal tissues from being erroneously detected as abnormal shadows, and to improve detection accuracy. .

  Further, when a false-positive image region is deleted from the abnormal shadow candidate region obtained by binarization, among the abnormal shadow candidate regions, an image region whose size (area, radius, etc.) is outside a predetermined range and / or a predetermined region Delete the image area other than the shape of the image, further determine whether it is false positive based on the index value obtained by multivariate analysis, and delete the image area determined to be false positive As a result, the abnormal shadow image area can be detected more reliably.

  In addition, by making it possible to change the size of the curvature filter (mask size) or the resolution of the medical image, it is possible to accurately detect an image region of an abnormal shadow regardless of the curvature of the curved surface indicated by the density distribution. It becomes possible.

  Note that the description in the present embodiment can be changed as appropriate without departing from the spirit of the present invention.

  For example, the curvature filter image conversion processing of the present embodiment is not only for detecting a shadow of a tumor or a microcalcification cluster from mammography, but also for detecting an abnormal shadow at that part from a medical image obtained by imaging other parts. Is also applicable. In addition, the processing target image in the present invention may be not only a radiographic image such as mammography but also an ultrasonic image or an MRI (Magnetic Resonance Imaging) image.

  Furthermore, the three-dimensional signal that is a target in the curvature filter image conversion processing is not limited to a medical image signal that includes three-direction components of the image position (two-dimensional coordinates) and density. For example, it may be a sound spectrogram composed of three axes of frequency, time, and frequency spectrum, or a color signal composed of a brightness component and two perceptual chromaticity components.

特徴量として平均曲率及びガウス曲率を用いる場合、曲面の形状は、平均曲率とガウス曲率の符号の正負により８種類に分類される。 In this embodiment, Shape Index is used as a feature amount representing the shape of a curved surface, but average curvature, Gaussian curvature, and Curvedness can also be used. The definitions of average curvature, gaussian curvature, and curvedness are shown in equations (6) to (8).
Average curvature = (k _max + k _min ) / 2 (6)
Gaussian curvature = k _max × _kmin (7)

When the average curvature and the Gaussian curvature are used as the feature amount, the shape of the curved surface is classified into eight types according to the sign of the average curvature and the Gaussian curvature.

1 is a block diagram showing a main part configuration of an image processing apparatus according to an embodiment of the present invention. The figure for demonstrating rotation of the normal plane which made the rotation axis the curved surface showing density distribution of a medical image signal, and the normal line of the attention pixel P as a rotation axis. The figure which shows the curved surface showing the density distribution in a digital image. The figure explaining the method of calculating the curvature of a curve by approximating the curve which shows density distribution with a circle using the least squares method. The figure which shows the curved surface shape corresponding to the value of Shape Index. The figure explaining a mode that a curvature filter is scanned within a process target image. 5 is a flowchart showing an abnormal shadow detection process executed in the image processing apparatus of the present embodiment. 6 is a flowchart showing curvature filter image conversion processing executed in the image processing apparatus of the present embodiment. The figure which shows distribution of the pixel signal value in the normal cross section J on the normal plane F. FIG. The figure which shows the approximate circle in each mask size at the time of changing the magnitude | size (mask size) of the image area | region which calculates a normal curvature.

Explanation of symbols

1 Image processing device 2 CPU
3 I / F
4 Operation unit 5 Display unit 6 Communication unit 7 RAM
8 Storage device 81 Feature file 9 Program memory

Claims (34)

A first feature amount calculating means for calculating a feature amount representing a shape of a curved surface indicating a density distribution of a medical image;
Detecting means for detecting an abnormal shadow candidate region from the medical image using the feature amount calculated by the first feature amount calculating unit;
Second feature amount calculating means for calculating a feature amount of the abnormal shadow candidate region detected by the detecting means;
False positive deletion means for deleting false positives of the abnormal shadow candidate region based on the feature quantity calculated by the second feature quantity calculation means;
An image processing apparatus comprising:   The image processing apparatus according to claim 1, wherein the first feature amount calculating unit calculates a shape index as a feature amount representing the shape of the curved surface.   The image processing apparatus according to claim 1, wherein the feature quantity calculated by the second feature quantity calculation unit includes a shape index. The detection unit binarizes the medical image using a preset threshold value of the feature amount based on the feature amount calculated by the first feature amount calculation unit, so that an abnormal shadow candidate region is obtained. Detect
The image processing apparatus according to claim 1, wherein the false positive deletion unit deletes a false positive image region from the abnormal shadow candidate region obtained by the binarization.   The image processing apparatus according to claim 1, further comprising a display unit that displays an image obtained by deleting false positives by the false positive deleting unit. A setting unit for setting an arbitrary pixel of interest in the medical image and setting an image region within a predetermined range from the pixel of interest,
The image processing apparatus according to claim 1, wherein the first feature amount calculation unit calculates a feature amount representing a curved surface shape of a density distribution in the set image region. . The normal plane is rotated by a predetermined angle around the normal line of the pixel of interest on the curved surface showing the density distribution of the medical image as a rotation axis, and the curved surface of the image area within the predetermined range from the pixel of interest is plotted on the normal plane for each rotation angle. A function calculating means for calculating a function approximating a curve obtained by cutting out,
The image processing apparatus according to claim 1, wherein the first feature amount calculating unit calculates a feature amount based on the approximate function calculated by the function calculating unit.   The image processing apparatus according to claim 7, wherein the function calculating unit calculates a function that approximates the curve using a least square method. The function calculating means calculates an approximate circle as the approximate function,
The first feature amount calculation means calculates a curvature at the target pixel of the curve from the calculated approximate circle radius for each rotation angle, and calculates a feature amount based on the curvature calculated for each rotation angle. The image processing apparatus according to claim 7, wherein the image processing apparatus is an image processing apparatus.   The image processing apparatus according to claim 1, wherein the first feature amount calculating unit calculates feature amounts for all pixels of the medical image.   The said false positive deletion means deletes the image area | region outside a predetermined range as a false positive image area among the said abnormal shadow candidate area | regions, The Claim 1 characterized by the above-mentioned. Image processing apparatus.   The false positive deletion means deletes an image area whose shape is other than a predetermined shape from the abnormal shadow candidate areas as a false positive image area. Image processing apparatus.   The false positive deletion means uses various feature quantities of the medical image including the feature quantity calculated by the first and / or second feature quantity calculation means as input data, and indicates an index indicating the degree of true positive abnormal shadow The image processing apparatus according to claim 1, wherein false-positive image regions are deleted using multivariate analysis using values as output data.   The first feature amount calculation unit changes a region range of a curved surface from which a feature amount is calculated, and calculates a feature amount in the changed region range. An image processing apparatus according to 1.   The image processing according to any one of claims 1 to 13, wherein the first feature amount calculating unit changes a resolution of the medical image and calculates a feature amount with the changed resolution. apparatus.   The image processing apparatus according to claim 1, wherein an image area constituting a curved surface with a Gaussian distribution is detected as an abnormal shadow from the deletion of false positives by the false positive deletion unit. .   The image processing apparatus according to claim 16, wherein an image area of a tumor shadow is detected as an abnormal shadow from the false positive deletion by the false positive deletion means. A first feature amount calculating step of calculating a feature amount representing a shape of a curved surface indicating a density distribution of a medical image;
A detection step of detecting an abnormal shadow candidate region from the medical image using the feature amount calculated in the first feature amount calculation step;
A second feature amount calculation step of calculating a feature amount of the abnormal shadow candidate region detected in the detection step;
A false positive deletion step of deleting false positives of the abnormal shadow candidate region based on the feature amount calculated in the second feature amount calculation step;
An image processing method comprising:   The image processing method according to claim 18, wherein in the first feature amount calculating step, a shape index is calculated as a feature amount representing the shape of the curved surface.   20. The image processing method according to claim 18, wherein the feature quantity calculated in the second feature quantity calculation step includes a shape index. In the detection step, based on the feature amount calculated in the first feature amount calculation step, the medical image is binarized using a preset threshold value of the feature amount, whereby an abnormal shadow candidate region is obtained. Is detected,
21. The image processing method according to claim 18, wherein, in the false positive deletion step, a false positive image region is deleted from the abnormal shadow candidate region obtained by the binarization. .   The image processing method according to any one of claims 18 to 21, further comprising a display step of displaying an image obtained by deleting false positives in the false positive deletion step. A setting step of setting an arbitrary target pixel in the medical image and setting an image region within a predetermined range from the target pixel;
The image processing according to any one of claims 18 to 22, wherein, in the first feature amount calculation step, a feature amount representing a curved surface shape of a density distribution in the set image region is calculated. Method. The normal plane is rotated by a predetermined angle around the normal line of the pixel of interest on the curved surface showing the density distribution of the medical image as a rotation axis, and the curved surface of the image area within the predetermined range from the pixel of interest is plotted on the normal plane for each rotation angle. Including a function calculation step of calculating a function approximating a curve obtained by cutting,
The image processing method according to any one of claims 18 to 23, wherein in the first feature amount calculation step, a feature amount is calculated based on the approximate function calculated in the function calculation step. .   The image processing method according to claim 24, wherein in the function calculating step, a function that approximates the curve is calculated using a least square method. In the function calculating step, an approximate circle is calculated as the approximate function,
In the first feature amount calculation step, the curvature of the target pixel of the curve is calculated from the calculated radius of the approximate circle for each rotation angle, and the feature amount is calculated based on the curvature calculated for each rotation angle. The image processing method according to claim 24 or 25, wherein:   The image processing method according to any one of claims 18 to 26, wherein in the first feature amount calculation step, a feature amount is calculated for all pixels of the medical image.   28. The false positive deletion step, wherein, among the abnormal shadow candidate regions, an image region whose size is outside a predetermined range is deleted as a false positive image region. The image processing method as described.   28. The false positive deletion step, wherein an image region whose shape is other than a predetermined shape is deleted as a false positive image region among the abnormal shadow candidate regions. The image processing method as described.   In the false positive deletion step, various feature amounts of the medical image including the feature amount calculated in the first and / or second feature amount calculation step are used as input data, and an index indicating the degree of true positive abnormal shadow The image processing method according to any one of claims 18 to 29, wherein a false positive image region is deleted using multivariate analysis using a value as output data.   31. The method according to claim 18, wherein in the first feature amount calculation step, a region range of the curved surface for calculating the feature amount is changed, and the feature amount is calculated in the changed region range. The image processing method according to item.   The image according to any one of claims 18 to 30, wherein, in the first feature amount calculation step, the resolution of the medical image is changed, and the feature amount is calculated with the changed resolution. Processing method.   The image processing method according to any one of claims 18 to 32, wherein an image region constituting a Gaussian distributed curved surface is detected as an abnormal shadow from the false positive deletion in the false positive deletion step. .   34. The image processing method according to claim 33, wherein an image region of a tumor shadow is detected as an abnormal shadow from the false positive deletion in the false positive deletion step.

Images