JP2004290570A - Image diagnosis supporting device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image diagnosis supporting device by which diagnosis efficiency can be improved. <P>SOLUTION: The image diagnosis supporting device comprises a memory 101b reading in medical purpose tomogram data of a subject acquired from a medical purpose image diagnosing device, a CPU 101a detecting an abnormal shade candidate within the medical purpose tomogram data read in by an image read-out means and having a mounted program for making a stacking-up image stacking up the read-in medical purpose tomogram images in the body axial direction of the subject, and an image displaying device 102 displaying a mark of a position of the abnormal shade candidate detected by the abnormal shade candidate detecting means on the stacking-up image made by the stacking-up image forming means and the medical purpose tomogram. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an image diagnosis support apparatus that performs CAD (Computer Aided Detection) processing for detecting a lesion (abnormal shadow) having a predetermined feature amount in a medical image by a computer.
[0002]
[Prior art]
A conventional image diagnosis support apparatus analyzes a shadow of an image captured by a CT apparatus or an MRI apparatus using a computer as described in [Patent Document 1], and narrows down a lesion candidate from the shadow to obtain a doctor. To support the diagnosis by asking the doctor. As a medical image of the lung field as an example of narrowing the focus candidates from shadows, in the lung field, in addition to shadows such as cancer, blood vessels, cross-sections of blood vessels, cross-sections of bronchi are mixed. Then, a shadow that is considered to be a cancer candidate is extracted from these images and presented to the doctor. As a method of presenting to a doctor, a method of encircling a shadow considered to be a cancer candidate on a medical image with, for example, a circle is generally used.
[0003]
[Patent Document 1]
WO 02/02002
[0004]
[Problems to be solved by the invention]
However, since the mark indicating the cancer candidate is marked on the tomographic image, the mark indicating the cancer candidate must be found while looking at each tomographic image. In addition, when a doctor who has already performed one interpretation has to refer to the analysis result, the same operation as in the interpretation must be performed, which may hinder efficient diagnosis.
[0005]
An object of the present invention is to provide an image diagnosis support device with improved diagnosis efficiency.
[0006]
[Means for Solving the Problems]
The object is to read medical tomographic image data of a subject obtained from a medical image diagnostic apparatus, to detect an abnormal shadow candidate in medical tomographic image data read by the image reading means, Means for creating a stacked image by stacking the medical tomographic images in the body axis direction of the subject, and creating the position of the abnormal shadow candidate detected by the abnormal shadow candidate detecting means by the medical tomographic image and the stacked image creating means Means for displaying a mark on the stacked image thus obtained.
[0007]
BEST MODE FOR CARRYING OUT THE INVENTION
First, a hardware configuration of the image diagnosis support apparatus of the present invention will be described with reference to FIG.
[0008]
The image diagnostic apparatus operates an image processing apparatus 101, an image display apparatus 102 for displaying a medical image on which image processing has been performed by the image processing apparatus 101, and buttons on a screen displayed on the image display apparatus such as a keyboard and a mouse. The external input device 103 includes
[0009]
The image processing apparatus 101 includes a CPU 101a for performing an operation for detecting an abnormal shadow candidate, a magnetic disk 101c for storing a plurality of tomographic images, and a temporary storage for an image to be displayed on an image display device and an image for performing an operation. A memory 101b for performing dynamic development and an interface (I / F) 101d for inputting and outputting tomographic images.
[0010]
Next, processing contents of the image diagnosis support apparatus of the present invention will be described with reference to FIGS.
Here, an example in which 30 chest slice images having a slice thickness of 10 mm are captured by an X-ray CT apparatus is considered.
[0011]
(Step 1)
A chest slice image captured by the X-ray CT apparatus is read and stored in the memory 101b of the image processing apparatus 101.
[0012]
(Step 2)
An abnormal shadow candidate detection process (hereinafter, CAD process) is performed on the image data stored in the memory 101b of the image processing apparatus 101. In the CAD processing, as shown in FIG. 3, a slice position (D1, D2 in the figure) where an abnormal shadow candidate is detected by calculation and a position of the abnormal shadow candidate in the slice, such as coordinates, are stored in the memory 101b.
[0013]
(Step 3)
An interpolation image is created for the 30 chest slice images read in step 1. Here, as an example, a principle of creating a 1 mm interpolation image (hereinafter, interpolation data) between the first slice and the second slice of a read image (hereinafter, actual data) will be described. Since the interval between the actual data is 10 mm and the interval between the interpolation images to be created is 1 mm, nine interpolation images are created between the actual data. As shown in FIG. 4, the image data of the first slice is S1 (x, y), the image data of the second slice is S2 (x, y), and the interpolated image data at the slice position of -3 mm is H3 (x, y). . In this case, the interpolation image data H3 (x, y) can be obtained by the following equation.
H3 (x, y) = 0.3 × S1 (x, y) + (1-0.3) × S2 (x, y)
Generally, the following equation is obtained.
Hn (x, y) = (A / T) × Pn × Sm (x, y) + (1− (A / T) × Pn) × Sm + 1 (x, y)
Here, A is the interpolation slice thickness, T is the slice thickness of the actual data, and Pn is the slice position of the interpolation image.
If actual data is obtained with a small slice thickness, an interpolation image need not be created.
[0014]
(Step 4)
Based on the interpolation data created in step 3, a stacked image (here, also referred to as a scout image) such as a chest simple X-ray image shown in FIG. 5 is created.
As shown in FIG. 6, the scout image scans the slice image in the direction of the arrow (v direction), and calculates the average value of the scanned pixels. This average value is obtained for all u-directions and used as column data. Further, the column data is obtained for all slice images, and the column data for the obtained number of images is stacked in the body axis direction.
Scanning in the v direction may be performed only on data within the field of view of the slice image. Further, the column data may be a maximum value instead of the average value of the scanned columns.
[0015]
(Step 5)
In the row on the scout image corresponding to the slice position (D1, D2) where the abnormal shadow candidate is stored in the memory in step 2, a mark for presenting to the operator that there is an abnormal shadow candidate is marked. The mark attached to the line on the scout image may be a line as shown in FIG. 7 or the line may be colored, but the type of the mark is not limited to the line. Alternatively, in addition to the slice position where the abnormal shadow candidate is located, the slice position where the abnormal shadow candidate is stored in the memory in step 2 and the position of the abnormal shadow candidate within the slice are marked and displayed as shown in FIG. You may.
[0016]
(Step 6)
The real data, CAD processing data, interpolation data, and scout data created in step 2, step 3, and step 4 are developed in the memory, and the real image, the CAD-processed image, the interpolation image, and the scout image are displayed on the image display device. In the image display, as shown in FIG. 9, a real image or an image after CAD processing and a scout image are displayed.
For example, as shown in FIG. 10, an interpolated image is provided with an interpolated / non-interpolated switching button so that it is possible to switch between displaying an interpolated image or displaying an actual image.
The real image and the CAD-processed image are arranged on the same screen as shown in FIG. 11, for example, so that they can be compared.
[0017]
(Step 7)
If an abnormal shadow candidate marked on the scout image described in step 5 is pointed out by an external input device such as a mouse, an image after CAD processing corresponding to the slice position is displayed. FIG. 12 shows a processing flow of this processing.
[0018]
(Step 7-1)
An arbitrary location on the scout image displayed on the image display device is indicated by an external input device such as a mouse as shown in FIG.
(Step 7-2)
As shown in FIG. 14, the top row of the scout image is set to the 0th row at the row position, and the number of the indicated row is calculated. Since the scout image is an image obtained by stacking images interpolated by 1 mm, for example, when the 120th column is pointed out on the scout image, the slice position can be associated as 120 mm.
(Step 7-3)
The plurality of abnormal shadow slice positions (D1, D2,...) Obtained in step 2 are compared with the slice position of the pointed point obtained in step 7-2, and the slice position of the pointed point in the abnormal shadow slice position is compared. Search for a slice position that matches.
(Step 7-4)
If there is an abnormal shadow slice position that coincides with the indicated slice position in step 7-3, the abnormal shadow slice is displayed on the image display device. If there is no matching slice position, no display is made.
[0019]
According to the embodiment described above, it is possible to provide an image diagnosis support apparatus capable of realizing quick image reading by reducing the labor of a doctor who refers to a CAD analysis result and improving operability.
[0020]
【The invention's effect】
The present invention can provide an image diagnosis support device with improved diagnosis efficiency.
[Brief description of the drawings]
FIG. 1 is a diagram showing a hardware configuration of the present invention.
FIG. 2 is a diagram showing a flowchart of a processing outline of the present invention.
FIG. 3 is a diagram showing a mode of obtaining a slice position having an abnormal shadow candidate as a result of CAD operation.
FIG. 4 is a diagram showing a mode of creating an interpolation image based on an actual image.
FIG. 5 is a schematic diagram illustrating a scout image.
FIG. 6 is a diagram showing a mode of creating row data of a scout image from a slice image.
FIG. 7 is a display example showing a slice position where an abnormal shadow candidate exists on a scout image.
FIG. 8 is a display example showing a slice position where an abnormal shadow candidate exists on a scout image.
FIG. 9 is a display example diagram for displaying an actual image or an image after CAD calculation and a scout image on an image display device.
FIG. 10 is a diagram showing a display example of an interpolation image and a mode of performing display switching between the interpolation image and the real image.
FIG. 11 is a display example diagram for displaying an actual image, an image after CAD calculation, and a scout image on an image display device.
FIG. 12 is a view showing a flowchart of slice display change processing.
FIG. 13 is a view showing a method of pointing out abnormal shadow candidates on a scout image in slice display change processing.
FIG. 14 is a diagram showing a mode of obtaining a slice position corresponding to a position pointed out on a scout image in a slice display change process.

Claims (1)

Means for reading medical tomographic image data of the subject obtained from the medical image diagnostic apparatus, and means for detecting an abnormal shadow candidate in the medical tomographic image data read by the image reading means,
Means for creating a stacked image in which the read medical tomographic images are stacked in the body axis direction of the subject,
Means for displaying a position of the abnormal shadow candidate detected by the abnormal shadow candidate detecting means on a medical tomographic image and a stacked image created by the stacked image creating means. apparatus.

Images