JP2005034473A - Irregular shadow detecting apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an irregular shadow detecting apparatus which can be applied to the whole body of a patient and can easily confirm variation with time of the shadow. <P>SOLUTION: Tomograms for sections comprising the head, the neck, the breast, the abdomen, and the leg of the same subject which are acquired at the diverse time are stored with relating to the acquisition times. Also, irregular shadow detection algorithm for each section is memorized. Detection of the irregular shadow is carried out based on the algorithm which after the section of the tomogram is recognized, is automatically selected according to the section. In this case, the tomogram is interpolated at intervals below its slice interval to create tomograms and the tomograms at the same position which have diverse acquisition time are chosen from them to perform a differential operation. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an abnormal shadow detection apparatus, and more particularly to an abnormal shadow detection apparatus applicable to the whole body.

An abnormal shadow of a patient is detected using a computer-aided diagnosis apparatus (hereinafter referred to as CAD). In the detection of abnormal shadows, different CADs are used depending on parts such as the chest, liver, and large intestine (for example, see Non-Patent Document 1).
"OmniCAD (TM)", [online], 2003, R2 Technology, Inc., [searched July 1, 2003], Internet <URL: HYPERLINK "http://www.r2tech.com/prd/shared /pdf/OmniCad.pdf "http://www.r2tech.com/prd/shared/pdf/OmniCad.pdf>

  However, in the conventional abnormal shadow detection as described above, it is necessary to use different CADs for different parts in order to diagnose the whole body of the patient, and it is difficult to capture changes in the shadow over time.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an abnormal shadow detection apparatus that can be applied to the whole body of a patient and can easily confirm a change with time of the shadow.

  In order to achieve the above object, the abnormal shadow detection apparatus according to claim 1 is a medical image of a part including the head, neck, chest, abdomen, and legs of the same subject acquired at different dates and times. When detecting an abnormal shadow based on the medical image stored in the image storage means, the storage means for storing the abnormal shadow detection algorithm for each part of the subject, and the medical image stored in the image storage means. Identification means for identifying a part of the medical image, selection means for selecting an abnormal shadow detection algorithm for the identified part of the medical image from the abnormal shadow detection algorithm stored in the storage means, and storage in the image storage means Detecting means for detecting an abnormal shadow using the obtained medical images having different acquisition dates and times and an abnormal shadow detection algorithm selected based on a part of the medical image, and the detection And display means for displaying the result, the.

  Since the abnormal shadow detection apparatus according to claim 1 detects an abnormal shadow using a medical image acquired at different dates and times and an abnormal shadow detection algorithm corresponding to a part of the medical image, it can be applied to the whole body of a patient. The change with time of shadow can be easily confirmed.

  The abnormal shadow detecting apparatus according to claim 2 is the abnormal shadow detecting apparatus according to claim 1, wherein the medical image is a tomographic image acquired at a predetermined slice interval, and the tomographic image is stored in the image storage means. Is stored.

  The abnormal shadow detection apparatus according to claim 3 is the abnormal shadow detection apparatus according to claim 2, wherein the detection means interpolates the tomographic image at an interval equal to or less than the slice interval when detecting the abnormal shadow. It is characterized by creating an image.

  In the abnormal shadow detecting apparatus according to the third aspect, the abnormal shadow can be detected at intervals equal to or smaller than the slice thickness of the tomographic image. In addition, it is not necessary to store a large number of tomographic images in advance, and a tomographic image of a necessary part may be created when detecting an abnormal shadow, so that the storage capacity can be reduced.

  The abnormal shadow detection apparatus according to claim 4 is the abnormal shadow detection apparatus according to claim 3, wherein when the detection unit detects the abnormal shadow, the tomograms at the same position having different acquisition dates and times from the created tomographic image An image is selected, and the selected tomographic image is subjected to a difference calculation.

  In the abnormal shadow detection apparatus according to the fourth aspect, the difference calculation is performed by selecting the tomographic images at the same position with different acquisition dates and times, so that it is possible to more easily check the time variation of the abnormal shadow.

  The abnormal shadow detection apparatus according to the present invention can be applied to the whole body of a patient, and the change with time of the shadow can be easily confirmed.

  Hereinafter, preferred embodiments of an abnormal shadow detection apparatus according to the present invention will be described in detail with reference to the accompanying drawings.

  FIG. 1 shows a configuration of an abnormal shadow detection system 10 to which the present exemplary embodiment is applied. The abnormal shadow detection system 10 includes an abnormal shadow detection device 20, a CT device 40, and an MRI device 50, and these devices are connected via a LAN 60.

  The abnormal shadow detection apparatus 20 relates to the abnormal shadow detection apparatus of the present invention, and includes a central processing unit (hereinafter referred to as CPU) 22, a main memory 24, a magnetic disk 26, a display memory 28, a CRT 30, a controller 32, a mouse 34, and a keyboard. 36, which are connected via a common bus 38.

  The magnetic disk 26 stores a database 100 in which data of images taken by the CT apparatus 40 and the MRI apparatus 50 are stored, and an abnormal shadow detection program 200, and the CPU 22 performs predetermined processing according to this program. In processing, the main memory 24 is used as an area for temporary storage and processing of images and data. The processing result is displayed on the CRT 30 via the display memory 28 and stored in the magnetic disk 26, and is used for re-displaying and referring to the result.

  The CT apparatus 40 and the MRI apparatus 50 capture medical images of patients such as tomograms and DR images, and the captured image data can be stored in the database 100 stored in the magnetic disk 26 via the LAN 60. It has become.

  Next, the abnormal shadow detection procedure in the abnormal shadow detection system 10 will be described with reference to FIG. In the following, a case where abnormal shadow detection is performed on a tomographic image is described.

  First, as shown in FIG. 2, a tomographic image of the whole body is taken at the age of N1 using the CT apparatus 40 and the MRI apparatus 50, and the image data is stored in the database 100 of the abnormal shadow detection apparatus 20 (step 30).

  Next, a tomographic image of the whole body is taken at the age of N2 using the CT apparatus 40 and the MRI apparatus 50, and the image data is set and stored in the database 100 (step 31). Thereafter, data of tomographic images obtained by imaging the whole body using the CT apparatus 40 and the MRI apparatus 50 are stored as a set regularly or irregularly (when Nx). Using these data, the abnormal shadow detection apparatus 20 detects an abnormal shadow.

  When detecting an abnormal shadow, a plurality of sets of tomographic images stored over time are used to perform subtraction between the corresponding tomographic images of each set. The slice interval of the tomographic images (for example, about 5 mm to 10 mm). May differ between the sets, so before subtraction of tomographic images, the tomographic images are interpolated within the slice interval at the time of imaging (step 32). The interval (for example, 1 mm) between the tomographic images created by interpolation is set between the sets to be subtracted.

  Here, as shown in the example of FIG. 3, the rate of growth depending on the section and age, such as between the acromion and the radius edge, between the radius edge and the upper pelvis, between the upper edge of the pelvis and the tailbone, and from the coccyx to the toe Therefore, the number of tomographic images created by interpolation differs. For example, in FIG. 3, the number of tomographic images is 1.03 times for the head and 1.07 times for the legs. At this time, for example, it is checked in advance that the chest algorithm is effective from the acromion to the radius end in FIG.

  After interpolation calculation, subtraction is performed between tomographic images (step 33). Subtraction of tomographic images is performed between tomographic images at the same anatomical position. After the subtraction, abnormal shadow detection is performed with an algorithm corresponding to the site of the tomographic image (step 34), and the result is displayed on the CRT 30 (step 35). In step 34, an abnormal shadow may be detected using a feature quantity such as circularity according to the part, or an abnormal shadow may be detected by a method similar to CAD for each part.

  Even if the position is anatomically the same, the position and size of the subject in the tomographic image may differ depending on the imaging time and imaging conditions. For this reason, position adjustment (movement, rotation) and size adjustment (enlargement, reduction) may be performed when subtracting images.

  As described above, the abnormal shadow detection apparatus 20 performs subtraction between tomographic images created by the interpolation calculation, so that the temporal change of the abnormal shadow can be easily confirmed.

  The above is the description of the abnormal shadow detection for each part. Next, a procedure for continuously processing whole body data will be described with reference to FIG.

  First, a set of tomographic images taken at different dates (for example, N1 years old and N2 years old) is input (step 47). Next, each tomographic image is interpolated at intervals (for example, 1 mm) that are equal to or less than the slice interval to create a tomographic image (step 48). This interpolation interval is set between the sets.

  As described above, the abnormal shadow detection apparatus 20 does not need to store a large number of tomographic images in advance, and it is only necessary to create a tomographic image of a necessary part at the time of abnormal shadow detection, so that the storage capacity can be reduced. it can. In addition, since it is only necessary to align the intervals by interpolation, abnormal shadows can be detected even when the slice intervals of tomographic images differ between sets.

  From the tomographic images created for each set, a tomographic image having the same anatomical position is selected (step 49). At this time, the correlation between the tomographic images of each set is obtained, and two images having the highest correlation can be set at the same position. The following processing is performed on the tomographic image selected in step 49.

  It is determined whether or not the image is classified for the head (step 50). While an oval region with a high CT value corresponding to the skull is present in the image, it may be determined as the head, or the eye appears in the image after examining from the image showing the end of the skull The head may be determined until it can be recognized.

  Which image is determined to be the head depends not on medical distinction but on the nature of the algorithm used. Specifically, it is possible to determine whether or not the head is a head based on whether or not a hollow closed region exists with a CT value of 1000 or more corresponding to the skull.

  If the determination is affirmative, the routine proceeds to step 55, where calculation is performed using the head algorithm, and then the routine proceeds to step 60. In the calculation in step 55, as described in steps 30 to 34 above, tomographic images taken at different times are interpolated to create tomographic images that are anatomically located at the same position, and subtraction is performed between these tomographic images. After that, the abnormal shadow is detected by the head algorithm. On the other hand, if the determination is negative, the process proceeds to the next step 51.

  In step 51, it is determined whether or not the image is a neck. The cervical region can be recognized because it is a continuous region from the head slice. For example, the cervical region is determined until the shoulder ridge shown in FIG. 7 is recognized in the image. If the determination is affirmative, the routine proceeds to step 56, where calculation is performed using the cervical algorithm, and then the routine proceeds to step 60. In the calculation of step 56, tomographic images taken at different times are interpolated to create anatomical tomographic images, and after subtraction between these tomographic images, an abnormal shadow is detected by the cervical algorithm. I do. On the other hand, if the determination in step 51 is negative, the process proceeds to the next step 52.

  In step 52, it is determined whether or not the image is a chest. Judgment as to whether or not it is a chest can be made based on whether or not a CT value is minus 900 or less and a hollow closed region exists. If the determination is affirmative, the routine proceeds to step 57, where calculation is performed using the chest / abdominal algorithm, and then the routine proceeds to step 60. In the calculation of step 57, tomograms taken at different times are interpolated to be applied to the chest and abdomen to create a tomogram that is anatomically located at the same position, and after subtraction between those tomograms, Abnormal shadows are detected by the abdominal algorithm. On the other hand, if the determination in step 52 is negative, the process proceeds to the next step 53.

  In step 53, it is determined whether or not the image is an abdomen. The abdominal region is known because it is a continuous region from the chest slice. If the determination is affirmative, the routine proceeds to step 58, where calculation is performed using the chest / abdominal algorithm, and then the routine proceeds to step 60. In the calculation of step 58, tomographic images taken at different times are interpolated to create anatomical tomographic images, and after subtraction between those tomographic images, an abnormal shadow is detected by the abdominal algorithm. Do. On the other hand, if the determination in step 53 is negative, the process proceeds to the next step 54.

  In step 54, it is determined whether or not the image is a leg. The leg portion is determined by recognizing that a region having a low CT value corresponding to muscles surrounds a region having a CT value of 1000 or more corresponding to a foot bone. If the determination is affirmative, the routine proceeds to step 59, where calculation is performed using the leg algorithm, and then the routine proceeds to step 60. In the calculation of step 59, tomographic images at different positions are interpolated to create tomographic images that are anatomically the same position, and after subtracting between these tomographic images, abnormal shadows are detected by the leg algorithm. Do. On the other hand, if the determination in step 54 is negative, the process proceeds to the next step 60.

  The abnormal shadow detection from step 55 to step 59 may be performed using a feature amount such as circularity according to the region, or may be performed by the same method as the CAD for each region.

  In step 60, it is determined whether or not the whole body data processing has been completed. If the determination is affirmative, the process proceeds to step 61 to display the abnormal shadow detection result, and the present processing routine is terminated. If the determination is negative, the process returns to step 49 to repeat the tomographic image selection and the abnormal shadow detection process.

  As described above, the abnormal shadow detection apparatus 20 identifies an imaging region with respect to tomographic images acquired at different dates and detects abnormal shadows using an algorithm selected based on the result. Applicable and it is possible to easily confirm the temporal change of abnormal shadows.

  Although the present embodiment describes the case where abnormal shadow detection is performed on a tomographic image, the abnormal shadow detection method and apparatus according to the present invention is not only a tomographic image but also other medical devices such as an X-ray image such as a DR image. It can also be applied to images.

It is a block diagram which shows the structure of the abnormal shadow detection system which concerns on one embodiment of this invention. It is a figure showing an outline of an abnormal shadow detection system concerning one embodiment of the present invention. It is a figure which shows the setting of the interpolation ratio according to one embodiment of this invention according to the site | part of a patient. It is a figure which shows the flow of a process in the abnormal shadow detection system concerning one embodiment of this invention. It is a figure which shows selection of the abnormal shadow detection algorithm according to one embodiment of this invention according to the site | part of a patient. It is a figure which shows data processing of the whole body according to one embodiment of this invention. It is a figure which concerns on one embodiment of this invention and shows an acromion. It is a figure which shows the coccyx concerning one embodiment of this invention. It is a figure which shows the outer half moon of a leg part concerning one embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Abnormal shadow detection system, 20 ... Abnormal shadow detection apparatus, 40 ... CT apparatus, 50 ... MRI apparatus, 60 ... LAN, 100 ... Database

Claims (4)

Image storage means for storing medical images of parts including the head, neck, chest, abdomen, and legs of the same subject acquired at different dates and times in association with the acquisition date and time;
Storage means for storing an abnormal shadow detection algorithm for each part of the subject;
Identification means for identifying a part of the medical image when detecting an abnormal shadow based on the medical image stored in the image storage means;
Selection means for selecting an abnormal shadow detection algorithm for a part of the identified medical image from the abnormal shadow detection algorithm stored in the storage means;
Detecting means for detecting an abnormal shadow using a medical image having a different acquisition date and time stored in the image storage means and an abnormal shadow detection algorithm selected based on a part of the medical image;
Display means for displaying the detection result;
An abnormal shadow detection apparatus comprising:   The abnormal shadow detection apparatus according to claim 1, wherein the medical image is a tomographic image acquired at a predetermined slice interval, and the tomographic image is stored in the image storage unit.   The abnormal shadow detection apparatus according to claim 2, wherein when detecting the abnormal shadow, the detection unit creates a tomographic image obtained by interpolating the tomographic image at an interval equal to or less than the slice interval. 4. The detection unit according to claim 3, wherein when detecting an abnormal shadow, the detection unit selects a tomographic image at the same position having a different acquisition date and time from the created tomographic image, and calculates a difference between the selected tomographic images. The abnormal shadow detection apparatus described.

Images