JP2008173213A - Device for supporting medical image diagnosis - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide the techniques to support diagnoses by giving information useful in grasping the timing of tumor treatments. <P>SOLUTION: A CPU 11 searches for the past tomogram of a patient bearing the same patient ID as the tomogram presently selected from the image attached information in an image database 4, makes a three-dimensional alignment between the searched past image and the presently selected image, sets up with a mouse 17 an abnormal shade on the picture showing the present image aligned with the past image, calculates the growth degree of the abnormal shade set up by the mouse 17, and displays on a display 15 the calculated degree of growth of abnormal shade in comparison with the growth in the present or past image. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention extracts abnormal shadows from tomographic images including the same part of the subject obtained at different dates and times by the medical image diagnosis support apparatus, and abnormal shadows from the size and shape of the extracted abnormal shadows. The present invention relates to a medical image diagnosis support apparatus capable of presenting a degree of growth to a radiogram interpretation doctor.

Conventionally, there is a technique called comparative interpretation that efficiently interprets medical images such as tomographic images including the same part of a subject obtained at different dates and times by a medical image diagnostic apparatus. (For example, Patent Document 1)
JP 2003-284690 A

  However, the comparative interpretation of Patent Document 1 only discloses a technique for displaying a medical image including the same part of a subject obtained at different dates and times so that it can be efficiently compared. The progress must be judged by the interpreter.

  By the way, although it is a ground glass-like shadow at an early stage such as high-grade lung cancer among abnormal shadows, there is a thing that changes from the inside of a tumor shadow to a solid shadow as it progresses. That is, the area of the solid shadow increases, but the size of the abnormal shadow does not change.

  For this reason, there has been an unsolved problem that it is difficult to determine the change in the future size of the tumor due to the normal tumor doubling time, that is, the timing of treatment.

  An object of the present invention is to provide a diagnosis support technology capable of providing information for determining the timing of treatment of a tumor.

  The object is to search for a past image of the same patient from the image supplementary information including the patient ID of the tomogram currently selected from the image database, the searched past image, and the currently selected current image. Positioning means for matching the three-dimensional position, abnormal shadow setting means for setting an abnormal shadow on the screen displaying the current image where the past image is aligned, and the degree of growth of the set abnormal shadow This is achieved by including an abnormal shadow growth degree calculating means for calculating the image and a display means for displaying the calculated abnormal shadow growth degree in association with the current image or the past image.

  According to the present invention, it is possible to provide information for determining the timing of tumor treatment.

  Hereinafter, preferred embodiments of a medical image diagnosis support method and apparatus according to the present invention will be described in detail with reference to the accompanying drawings.

  FIG. 1 is a hardware configuration diagram showing a configuration of 1 of a medical image diagnosis support system according to an embodiment of the invention.

  1 includes a medical tomographic imaging apparatus 2 such as an X-ray CT apparatus and an MR apparatus, an image database 4 that stores medical images obtained by the medical tomographic imaging apparatus 2, and a medical The medical tomographic imaging apparatus 2, the image database 4, and the medical image display apparatus 10 are connected to each other via a network such as a LAN 3.

  The medical image display apparatus 10 includes a central processing unit (CPU) 11 that mainly controls the operation of each component, a memory 12 that stores a control program for the apparatus and that serves as a work area when the program is executed, and an operating system. System (OS), peripheral device drive, magnetic disk 13 that stores various application software including programs for quantitatively evaluating the region of interest from medical images, etc., and display that temporarily stores display data Detects the state of the memory 14, the display 15 such as a CRT monitor or a liquid crystal monitor that displays images based on the data from the display memory 14, the keyboard 16, the mouse 17 as a position coordinate input device, and the state of the mouse 17. Connect the controller 18 that outputs signals such as the position of the mouse pointer on the display 15 and the state of the mouse 17 to the CPU 11 and the above components. The common bus 19 is provided.

  In the present embodiment, the medical image display device 10 reads a medical image from the image database 4 via the LAN 3, but a storage device connected to the medical image display device 10, such as an FDD, CD-RW drive, or MO drive Medical images may be read from a ZIP drive or the like. Alternatively, a medical image may be acquired directly from the medical tomographic imaging apparatus 2 via the LAN 3.

  FIG. 2 is a diagram showing a main flow executed by the medical image diagnosis support apparatus of FIG. The CPU 11 in FIG. 1 operates according to this main flow. Hereinafter, details of the main flow will be described in the order of steps.

  First, since the subject ID input screen is displayed on the display 14 of the medical image diagnosis support apparatus, the operator inputs the patient ID number. Then, the CPU 11 reads from the magnetic disk 12 a tomographic image corresponding to the ID number of the patient to be diagnosed from the tomographic images captured in advance by the medical image modality. When the reading is completed, the CPU 11 displays a tomographic image (current image).

[Step S20]
From the DICOM information in the read image data, images (past images) taken in the past of the same patient are read in the image database.

[Step S21]
The CPU 11 matches the position of the past image so that the position matches the current image. As a technique at this time, a technique of matching based on slice position information of the bronchial bifurcation may be used.

[Step S22]
After completing the alignment, the CPU 11 performs a growth degree calculation process. FIG. 3 is a diagram showing details of the growth degree calculation process. Details of this process will be described below in the order of steps.

[Step S30]
After designating the shadow of interest by the operator, the CPU 11 extracts the designated shadow area. As a technique at this time, a threshold value for extracting a shadow is determined based on the designated area and its peripheral area information, and binarization processing is performed using the threshold value. Then, a method of extracting a shadow using a region expansion method with a designated point as a starting point may be used.

[Step S31]
The CPU 11 analyzes the shadow existing at the same position in the past image and the extracted shadow inside from the extracted area, and calculates the feature amount. FIG. 4 is a diagram showing details of the feature amount calculation processing. Details of this process will be described below in the order of steps.

[Step S40]
The CPU 11 calculates the roughness of the CT value inside the shadow.
This roughness is the extracted shadow, and when divided into a high CT value region (high CT region) and a low CT value region (low CT region), the run length of the high CT region and the run length of the low CT region It is a ratio. If the roughness is low, the distribution of CT values inside the shadow is uniform, and if it is high, the distribution of CT values inside the shadow is not uniform. FIG. 5 shows a conceptual diagram regarding this roughness. This is also applied to past images.

[Step S41]
The CPU 11 analyzes an area that has changed over time.
At this time, the solid shadow mixed ground glass-like shadow tends to increase the size of the high CT value region in the shadow center. Using this feature, the number of pixels and the like in which the shadow center has changed is used as a feature amount.

[Step S42]
The CPU 11 calculates the following feature quantities in addition to the feature quantities extracted in step S40 and step S41.
Feature 1: Long shadow diameter Feature 2: Short shadow diameter Feature 3: Total shadow volume Feature 4: Shooting interval (time) from the last shooting to the present
Feature amount 5: A conventional tumor doubling time for abnormal shadows may be used, so add it.
Tumor doubling time is the time required for tumor volume to double. In order to calculate the doubling time from the image information, the following formula (1) is used.

ここで、V₁ とは1回目(前回)撮影時の体積、V₂ とは2回目(今回)撮影時の体積、Δt とは経時変化(時間)を表している。

Here, V ₁ represents the volume at the first (previous) shooting, V ₂ represents the volume at the second (current) shooting, and Δt represents a change with time (time).

[Step S32]
The CPU 11 calculates the growth degree of the shadow using a neural network based on the feature amount obtained in steps S40 to S42. An explanatory diagram is shown in FIG. As a technique at this time, the CPU 11 may perform multivariate analysis and statistically determine and calculate a coefficient value to be multiplied to each feature quantity, or may calculate an average value of each feature quantity. Good.

[Step S23]
The CPU 11 displays the degree of growth of the abnormal candidate shadow on the screen as illustrated in FIGS. Here, it has a two-screen configuration of a screen that displays the entire current image and a screen that displays an enlarged image of a designated part of the current image. FIG. 8 shows an example in which the calculated growth degree of the shadow is displayed on the enlarged image. FIG. 9 shows a display example in which the shadow feature values used for calculating the degree of growth are displayed as a list. Thus, it became possible to determine the change in the future size of the tumor, that is, the timing of treatment due to the normal tumor doubling time.

Next, a second embodiment of the medical image diagnosis support method and apparatus according to the present invention will be described in detail with reference to FIGS.
First, since the subject ID input screen is displayed on the display 14 of the medical image diagnosis support apparatus, the operator inputs the patient ID number. Then, the CPU 11 reads a tomographic image corresponding to the ID number of the patient to be diagnosed from the magnetic disk 12 from the tomographic images taken in advance by the medical image modality. When the reading is completed, the CPU 11 displays a tomographic image (current image).

[Step S20]
The CPU 11 reads images (past images) taken in the past of the same patient in the image database from the DICOM information in the read image data.

[Step S21]
The CPU 11 matches the position of the past image so that the position matches the current image. As a technique at this time, a technique of matching based on slice position information of the bronchial bifurcation may be used.

[Step S100]
The CPU 11 performs an abnormal candidate shadow detection process on the registered current image and the past image found from the image database.
The CPU 11 performs abnormal candidate shadow detection processing. In the abnormal candidate shadow detection process, for example, a multi-valued image is created from the original image 100, and the center coordinates of the shadow are obtained based on the multi-valued image. Then, the CPU 11 sequentially samples the shadow pixel values by rotating a radius of a predetermined length with reference to the center coordinates, and detects by determining whether the shadow is an abnormal candidate shadow based on the sampling pixel values. be able to.

[Step S22]
The CPU 11 performs a growth degree calculation process on the abnormal candidate shadow detected in step S90. This growth degree calculation process is the feature amount described in the first embodiment.

[Step S101]
The CPU 11 rearranges the growth degrees of the detected abnormal candidate shadows obtained in step S22 in descending order (or ascending order). Here, the rearranged order information is added as image supplementary information of the current image or the past image, and the supplementary information can also be displayed when the current image or the past image is displayed on the monitor.

[Step S102]
The CPU 11 displays the degree of growth of the abnormal candidate shadow on the screen as illustrated in FIG. Here, there are two screen configurations: a screen that displays the entire current image and a screen that displays an enlarged image of a designated part of the current image. FIG. 11 shows a display example in which the growth degrees of detected abnormal candidate shadows are displayed as a list. The operator can interpret images in the order of high degree of growth (or low order).
As a result, the affected area to be preferentially treated can be efficiently displayed.

1 is an overall configuration diagram of an image diagnostic system according to an embodiment of the present invention. It is a flowchart which shows 1st embodiment of the medical image diagnosis assistance method which concerns on this invention. FIG. 3 is a diagram showing details of a growth degree calculation process in the main flow of FIG. FIG. 3 is a diagram showing details of a feature amount calculation process in the main flow of FIG. FIG. 4 is an explanatory diagram of feature amounts in the CT value information analysis processing in the flow of FIG. FIG. 4 is an explanatory diagram of a feature amount in an analysis process of a temporally changing site in the flow of FIG. FIG. 4 is an explanatory diagram of a neural network used in the growth degree calculation process in the flow of FIG. 3 is a screen display example in a measurement result display process in the main flow of FIG. 6 is another screen display example in the measurement result display process in the main flow of FIG. It is a flowchart which shows 2nd embodiment of the medical image diagnosis assistance method which concerns on this invention. 9 is a screen display example in the measurement result display process in the main flow of FIG.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Medical image display system, 2 ... Medical tomography apparatus, 3 ... LAN, 4 ... Image database, 10 ... Medical image display apparatus, 11 ... CPU, 12 ... Main memory, 13 ... Magnetic disk, 14 ... Display memory, 15 ... monitor, 16 ... keyboard, 17 ... mouse, 18 ... controller, 19 ... common bus, 50 ... specified shadow area, 51 ... high CT value area, 52 ... run length of high CT value, 60 ... in past image Abnormal shadow 61: Abnormal shadow in the current image, 62: Time-dependent change site analysis processing result, 63: Time-varying curve representing the number of pixels that have changed over time with respect to the distance from the shadow center, 110 ... Original image (current image), 81/91/111 ... Enlarged image of abnormal shadow, 92 ... List of feature values, 112 ... List of growth degree in applied image

Claims (3)

Image search means for searching for past images of the same patient from the image supplementary information including the patient ID of the tomogram currently selected from the image database;
Alignment means for aligning a three-dimensional position between the searched past image and the currently selected current image;
An abnormal shadow setting means for setting an abnormal shadow on the screen displaying the current image in which the past image is aligned;
Abnormal shadow growth degree calculating means for calculating the growth degree of the set abnormal shadow;
Display means for displaying the calculated degree of growth of the abnormal shadow in association with the current image or the past image;
A medical image diagnosis support apparatus comprising:   The medical image diagnosis support apparatus according to claim 1, wherein the display unit displays a list of abnormal shadow growth degrees calculated by the abnormal shadow growth degree calculation unit.   The display means rearranges the list display of the calculated abnormal shadow growth degree in a predetermined order in descending or ascending order, and displays the rearranged order information in association with the current image or the past image. The medical image diagnosis support apparatus according to claim 2.

Images