JPS5967938A - Computer tomographic apparatus - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a computed tomography (CT) apparatus,
Especially regarding the data processing section.

With the currently mainstream 3rd generation CT, it takes about 1 second at the earliest to create one image, but if you want an image of the heart, you need to create an image with data sampling of about 0.1 seconds. However, this is mechanically difficult, and even if it were possible, the power of the tube would be insufficient and the quality of the data would deteriorate, making it impossible to obtain a good image. In this way, projection data (complete data or complete angular data) obtained by scanning the rotational direction at the required angle corresponding to one reconstructed image
It takes time to collect.

The purpose of the present invention is to convert projection data (incomplete data or incomplete angular direction data) that lacks the required angle to a near real image.
It is an object of the present invention to provide a computerized tomography apparatus that can obtain better images and that is cyclical and has small movements, such as optimal for reducing the effects of heart movement and breathing.

To understand the purpose of this, let's take the case of obtaining a heart image as an example.In this case, a heart image is created from complete angle data as a base image, for example, by a 4-second scan, and in this case, this base image is Image data that contains a 4-second blur (approximately 4 cycles are performed in 4 seconds) and is obtained by time-sharing included in the base image scan or within a short period of time (approximately 0.1 seconds). This is achieved by sequentially correcting the base image using incomplete angular direction data and approaching the image captured within this short time.

The purpose of this configuration is to follow the image data memory with a data correction circuit and an image display device, and the arithmetic circuit inputs base image data in a time-division manner, and at the same time inputs incomplete data from the projection data buffer. This is achieved by inputting the data, performing correction calculations, and feeding back the output data to the image data memory.

The gist and preferred embodiments of the invention are also explained with reference to the drawings.

FIG. 3 is a schematic block diagram showing a configuration example of the present invention. 12 is an X-ray tube, 14 is a detector, 16 is an A/D converter,
18 is a projection data buffer, 20 is a convolution (CNV) calculation circuit, 22 is a backglossion (BP) calculation circuit, and the former two constitute an image reconstruction processor, 24 is an image data memory, and 26 is a A data correction calculation circuit, 28 an image display device, and 30 a subject.

(Figure 1 is a good idea for creating a base image)
be scanned.

An embodiment in which data is obtained by scanning is shown.

Example t C. A base image is created by scanning 360 pixels, and correction is performed using about 30 pieces of data included in it (Figure 2 i) using a data correction calculation circuit.
If there is no detection due to the movement of , there is no difference even if the correction is made. If there is movement, the image approaches the image at the moment (phase) when the correction data (within 30) was obtained and becomes b<.

Note that similar to the present invention, attempts to create images from incomplete data and provide CT equipment with high temporal resolution have already been proposed (for example, IEEEVOL, BME-
29No, 5 MAY 1982. rIterat
iveReconstruction-Reproj
ctionJ (IRR)).

However, in these methods, an image is created from incomplete data (measured data) and the image is corrected one by one using the incomplete data, so the more incomplete the data, the more the image quality will deteriorate. seems to be decreasing.

In contrast, in the present invention, a base image that includes blur due to movement but is reconstructed from an overwhelming amount of information is stored in advance, and that image is sequentially reconstructed using less blurred data, although it is incomplete. By making corrections, image quality can be greatly improved. This is E
CG)! It can also be applied to incomplete data obtained using J.

The above-mentioned IRR proposal is explained as follows with reference to FIG.

■ Consider an object with an absorption coefficient distribution as shown in Figure 4 (1).

■ It is assumed that there are six pieces of measurement data (incomplete data) as shown in Figure 4 (2).

■ Using these six pieces of data, perform convolution, backglossion, and image reconstruction. Here, since it is impossible, we assume that it is as shown in Fig. 4 (3).

■ Next, the image data in Figure 4 (3) is
Correction will be made based on the measurement data of 2). -1 in order of j,
6 and add it to each matrix of column j-1. Similarly, the final results for j=2 to 6 are shown in FIG. 4 (4).

■ Now, perform projection from the data shown in Figure 4 (4) to obtain glosijection data in other angular directions that are not included in the measurement, and combine them with the measured data to create a set of projection data. Volume and back projection 1c are performed and reconstructed. Image obtained in this way (close to Figure 4 (4))
Then, using the measured data again, correction is performed as in the calculation described above. By repeating this several times, the image will converge and will hardly change, and this image will be the final image.

The present invention differs significantly from the IRR proposal in the following two points.

(i) In above (■), the image is reconstructed using incomplete measurement data and is compared to the base image, but
In this proposal, when there is little measurement data, the difference between the base image and the ideal image becomes large and artifacts become strong.

Therefore, we selected a base image that included some blur but was close to the ideal image in Figure 1 (based on sufficient angular direction data), and used it as measured data with little blur, although it was incomplete as angular data. (ii) In addition, in the above-mentioned item (2), a projection is performed from the corrected image, a set of projection data is created, it is reconstructed, and the correction is repeated again, but in the present invention, this This process is not necessary, and it is sufficient to repeatedly correct J=1 to 6 in the above-mentioned step (2), so the configuration is simple.

According to the mode of operation of the present invention, based on the base image,
It becomes possible to take images of the heart divided into its phases.

In the configuration example shown in FIG. 3, the image data source of the image display device 28 is such that the data correction calculation circuit 26 according to the present invention is stored in the image data memory 24 only when obtaining an image of a subject that may be obtained in the image data memory 24. It would be better if it was connected to 24.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing a scan mode for creating a base image, FIG. 2 is a diagram showing an example of scanning incomplete data, and FIG. 3 is a block diagram showing an embodiment of the present invention. , FIG. 4 is an explanatory diagram of data correction according to a prior example. 12 is an X-ray tube, 14 is a detector, 20 and 22 are reconstruction processors, 24 is an image data memory, and 26 is a data correction calculation circuit.

Claims (1)

[Claims] The image data memory is followed by a data correction calculation circuit and an image display device, and the calculation circuit inputs data for one reconstructed image obtained by scanning in the rotational direction at the necessary angle in a short time, and simultaneously inputs data from the projection data output. A computerized tomography apparatus 0 characterized in that data of insufficient required angle is inputted, a correction calculation is performed, and the output data is fed back to an image data memory.