JPS5865143A - Computer tomography 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 so-called third generation OCT scanners, which arise in particular from variations in the characteristics of radiation detection probes. f4+m
This invention relates to a CT scanner designed to reduce ring and one bear artifacts on one station of the image.

The so-called CT scanner (Computeriz@dTo
For example, as shown in Figure 61JJ1, mography' (computer tomography) consists of an X-ray source 1 that emits flat fan-shaped fan beam X-rays Fx, and a plurality of radiation detection elements that detect the X-rays. The Xi sources 1 are placed facing each other with the subject P sandwiched between the multi-bunnel type radiation detectors 2 arranged in parallel according to the expansion of the rays Fx, and
Then, the radiation detector 2 is placed in the middle of the whole body of the temporary specimen P (rotate in the same direction at predetermined angle increments at the same angular speed, and then #i Projections were used to collect X-projection data in various directions on the fortifications, and once enough data had been collected, this data was analyzed with an electronic computer and the examinee's salary was determined. The X-win absorption rate corresponding to the position of 1 garden of H is exposed, and the absorption rate is given an engineered gradation, so that the previous ml legend training dedication W [picture W in the surroundings! ?This is a very stiff material, and it is possible to obtain a side-effect fault structure from soft tissue to comb kneading.

In this way, the OCT scanner uses the fan beam XIv, which contains the subject, to rotate the X-ray source and the enclosed ray detector, and collects the X-vehicle projection data. Generation CT and \L7 are included.

By the way, in such a conventional CT scanner,
? The positional relationship between the tM source and each radiation detection element is always constant, and since there are many radiation detection elements, if the characteristics of each radiation detection element are not uniform, the X-projection Differences occur between the data due to the characteristics of the radiation ice detection elements Dh.

In practice, it is difficult to make the characteristics of each radiation detection element the same, so correction is performed using various methods to prevent differences in characteristics from being translated into data. However, the correction may not be performed correctly due to the intuitive nature of the radiation r>I3I extractor system, environmental changes, etc.

Because of this, radiation? Independent changes in the valley channels (each radiation detection element D) of the IMbL output result in various addition noises in the Tan configuration tomographic image.

Among these channels, if there is a channel that has a constant difference from the adjacent channels, the data collection position of that channel is a specific beam position in the fan beam A ring-shaped or arc-shaped artifact (1) was produced in the area where the image was formed, which was a major hindrance to the image quality.

The present invention has been made in view of the above-mentioned circumstances, and focuses on the properties of the log-like artifact components included in radiation projection data on the sinobram, and calculates the radiation m m l> 74 for the data on the sinobram. A two-dimensional filter with a cutoff frequency distribution of an appropriate spatial frequency that can remove artifact components in the position and in the direction of the channel row of the radiation detector is used, and the data value of the radiation projection is corrected using this two-dimensional filter. Then, by removing all artifact components and performing image correction using the projection data of this corrected body, an AM composite image is obtained in which artifacts caused by variations in the characteristics of the radiation detector are removed. The purpose of the present invention is to provide a computer to which the computer can be used.

Hereinafter, a practical example of the present invention will be described with reference to FIG.

A radiation axial source that irradiates fan beam X-rays is placed between the entire center of the object, and radiation IfM't- passes through the object.
Radiation detector constructed by arranging multiple channels of radiation detecting elements in parallel and scanning at equal intervals angles around the subject while keeping the relative relationship value a constant.
/R type OCT scanner, front + iL each corner IA
Fan beam x emitted from the radiation source at tm, VM
It is assumed that the radiation projection data group t-P (0, 1) of the subject is obtained through the radiation detector. Here, 0 is the scanning angle, that is, the projection angle of the radiation
), t is the channel number (channel i) of the radiation detection element of the radiation detector that detected the data.

Each data P(
Let us consider what is expressed as data points on a two-dimensional matrix surface with the front +L channel direction as the horizontal axis and the glojection force 4 as the vertical axis at θ, 1). This is called a sinobram, and the second part shows how each point in the plane created by the X1m beam of the projection that traces to the subject' corresponds to the glojection data passing through that point on this sinobram. Figure 3. That is, due to radiation source 1! P1+P on object P when Rojek 7wn Kakuro is 00! #P3 position! Logexi and ndata are each S1 esz *S on the tightsgram.
Corresponding to g, when θ changes from 0 to 2π, on the sinobram, the dust resistance r from the rotation center of pI HP @ + Pl
x erg ersf: becomes sine wave data with amplitude.

In addition, in Figure 2! Logexy, nde tosl.

The data of 8m+Sl is the channel c1 of the radiation detector 2.
,c,,Cs.

Here, if the radiation detection line 1° of a certain channel has a constant density difference by a certain projection angle interval, then R1 + R1l on the sinobram as shown in Fig. 5g4 (a).
It will have a vertical edge like R3. This is reconstruction I! The Ii image results in a ring arch 7 act.

The abnormal data is for the abnormal channel in Cynobram soil.
It is a collection of points extending linearly in the θ direction, and the ring component is generated based on the ring component data. be done.

Analyzing the correspondence between the information on the sinobram and the ring component in this way, the abnormal data that causes ring-shaped artifacts in the composite image is distributed in a straight line extending in the θ direction with respect to the abnormal channel on the sinobram. 0, tt) Full two-dimensional Fourier transform Iik・”1
When photographed in the spatial frequency region, as shown in Fig. 5, the ring frequency component bθ in the force direction is in the low frequency region of υ15 degrees, and in the t direction, the high frequency is in the high frequency region, that is, in the diagonal line portion Dht. It will be distributed.

Therefore, for P(θ, t), the cutoff frequency in the θ direction is ωθ. The low-pass characteristic of 1 coo, the force in the direction, and the to-off frequency are sushi. The above-mentioned Dht part is extracted by adding a two-dimensional filter operation having a bypass characteristic of Modified projection data (modified projection data) PI (θ, 1) with the component cut, that is, with the ring-shaped arch 7 act component removed, can be obtained.

Here, if the two-dimensional filter is H(θ, 1), then PI(
θ, 1) can be determined by the following equation.

PI (θ, t) = p (θ, t) - p (θ, 1) rate H(
θ, 1) = p(θ, 1) h(θ, 1) = p(θ, 1)-, (2CP(θ, 1) h(θ-θ
', t-t'ndθIdtl...
・・・Song・(1) Here the car is 1Ilii image reconstruction performance J
Commerusion 111i [calculation shown] in HC.

The two-dimensional filter h (0, t) is a function of 0 force direction independent in direction, h (θ , t) = h (#) lines (t)
・・・・・・・・・・・・・・・(2) and Kinwasuko, l
! - is created, and for p(e,t)K, mafh(#),! :
- If the scene axto is not present, perform a union operation with h(t) on the result (h(θ) and h(t) may be reversed) to obtain a two-dimensional filter edge. )'l-; h can be done.

418th! /'l (a) shows the correspondence between the locus of the groso-ection data and the locus of the ring component at P2 in FIG. 2.

It is assumed that the ring component occurs for a certain range of θ. The sail-shaped θ direction of the two-dimensional filter in real space for completely removing the ring component is shown in FIG. 4(b), and the direction in the θ direction is shown in FIG. 4(c).

In the direction of , the ring component occurring in the center channel is more emphasized in the reconstructed image, so it is more effective to select the two-dimensional filter appropriately depending on the channel.

In reality, it can be counted as 1 with a finite number of discrete values, but r = jil
The calculation given by the first equation is before and after Noirta processing! Projection 7″-ta total 1′f(P(s,j)).

(Pl(s-j))s Also, the two-dimensional filter used for filter processing is (b(k, t)). However, &C as 1=0 to M-1°.

Expression of MXN of CP (1, j)) by i<-t
(4), and (pi(i=j)') can also be expressed in the same format.

Note that the two-dimensional filter (h(k, t)) is (K+1)X(
L+1) (K, L is the number 111111), and if we take an example of a #15X3 two-dimensional filter,
As shown in the diagram 1 This
For r-ta 1ζ, cut-off frequency sound 4 with appropriate spatial frequency for lronoeku/banghougang and nanno1 in the channel direction.
] Adding a two-dimensional filter operation to the data correction shift, resulting in a reduction of the /G component.
It is possible to obtain a marked reduction in the ring-shaped arch 7 act (reduced clinical diagnosis%) without sacrificing much of the emotional component.

As described above, the present invention includes a radiation beam that emits a fan-shaped radiation beam toward the center of a region 19 (>N) in which a subject is placed, and A multi-channel detector for obtaining a radiation-detection output 1 is provided, which is made up of a constant number of radiation detection elements facing the source and arranged in parallel in the direction of spread of the radiation brain beam; By rotating the center at a predetermined angle increments as the center of full rotation in the region of the previous day's previous day while maintaining the positional relationship between the detector and the detector, In a computer tomography apparatus, the radiation detection data projected from the radiation source is obtained and the data is processed by a processing unit to reconstruct a cross-sectional image of the object at the cross section. Projection data t1 The data is divided into two-dimensional distribution according to the projection angle for each channel, and =i1'8t2 For each channel, the amount equivalent to the characteristic variation occurring in the radiation output element of that channel kw#) is removed. Filter constant at - Praise for each channel and radiation particle projection angle
MIJ = α: For each of the radial edge projection data that has been distributed in dimension, filter processing is performed using the corresponding position of the two-dimensional filter - 〇 Noirta formula, and after this filter processing, Based on the radiation projection data association ll1111#! Since it is configured with a function to perform seven reconstructions, the radiation-projection data for each channel is subjected to correction 1 corresponding to the characteristic variation for each channel by quadratic filter processing, and artifacts contained in the data are Since it is the incident radiation-projection data with the component removed,
If this projection data is used to compose an image, a reconstruction image with reduced artifacts and no loss of information can be obtained, and an easy-to-read image that is useful for diagnosis can be obtained. All tomography equipment can be provided.

[Brief explanation of the drawing]

Figures 1 (a) and (b) are for explaining the principle of the 31st successful CT scanner. Figure 3 is a diagram showing the relationship with each channel of the radiation detector, and is a diagram for explaining how the radiation data passing through each point of the subject in Figure 2 corresponds to the radiation detector L. , Fig. 4(,)(b) (C) is a diagram to clarify the effect of the sinobram and the two-dimensional filter yt =
FIG. 5 is a diagram for explaining the distribution of ring components in the frequency domain and the meaning of a two-dimensional filter, and FIG. 6 is a diagram showing an example of a second-order filter. 1... Radiation? tM Keyaki, 2... Radiation gland Kyoubatake, P...
・Subject, D... Radiation spin detection system. 3rd v! J Figure 4

Claims (1)

[Claims] A radiation beam that spreads out in a fan-shape, the entire subject to be examined, a radiation source that radiates toward a predetermined area, and a radiation source that faces the radiation source through the fortress, so that the radiation beam spreads in nine directions. A multi-channel detector for obtaining radial leg projection data 'ft, which is constructed by arranging radial forest elements of a straw ridge in parallel, and t.
Relating the source and detector to each other'
By keeping the center of the H11ml predetermined area at a predetermined angle at a time while rotating, Ml
Obtain the radiation-projection data projected from each angular force direction on the plane of a predetermined region, and perform reconstruction processing using the iron metal that performs t calculations (↓F), mt
KL to reconstruct the cross-sectional image of the subject of the 5tJtr song.
In the i-computer tomography device, the arithmetic processing device divides the previous ml of each radial H carpet projection data t1 into two-dimensional distribution according to each channel and projection angle from which the data was obtained, and also divides the data into a two-dimensional distribution according to the projection angle. Setting all the two-dimensional filters assigned to each channel and radiation projection angle [''+J @α: Applying a filtering action using the two-dimensional filter to each of the dimensionally distributed radiation-projection data,
A computer tomography apparatus that performs a correction to remove characteristic fluctuations occurring in a radiation detection element of a channel, and reconstructs an image based on the corrected radiation projection data.