JP3400063B2 - Computer tomography equipment - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a computer tomography apparatus (hereinafter simply referred to as a CT apparatus), and more particularly to 36.
CT that can continuously collect projection data at angles of 0 ° or more
Regarding the device.

[0002]

2. Description of the Related Art In an X-ray diagnostic apparatus, a subject is continuously exposed to X-rays, an X-ray image is picked up by a television camera, and the X-ray image is continuously observed on a screen. However, examinations and treatments are often performed. This is commonly referred to as fluoroscopy.

In recent years, even in the CT apparatus, the development of the slip ring allows the X-ray tube and the detector to continuously rotate around the subject, and it is possible to continuously collect projection data of the same slice having different scan times. Because it is getting
It is desired to display tomographic images having different scan times in almost real time. This is called CT fluoroscopy.

However, in order to perform CT fluoroscopy, it is necessary to reconstruct a tomographic image in real time, and there is a problem that the load of hardware for reconstructing a large amount of data at high speed is heavy. To solve this problem, 36
A method for efficiently reconstructing a tomographic image from continuous projection data of 0 ° or more has been considered. An example of this is the continuous slice reconstruction method described in Japanese Patent Publication No. 1-213136 (US Pat. No. 4,495,645) filed by the present inventor. In this example, since the tomographic image can be reconstructed in about 1 second or less, the tomographic image can be displayed almost in real time with the scan.

Since projection data of 360 ° or more is continuously collected, there is a drawback that the exposure dose of the subject increases.
In order to solve this problem, the dose (mAs) should be reduced during CT fluoroscopy to observe images continuously, and the dose should be increased in important scenes so that the image should be stored (captured). It is possible to do It is desirable that the dose during CT fluoroscopy be as small as possible, but there is a problem that the image quality deteriorates when the dose is reduced. For example, it is necessary to irradiate the subject with X-rays with a dose of 50 mAs or more in order to obtain an observable CT fluoroscopic image. Below this, it is difficult to obtain a CT fluoroscopic image that can be used for diagnosis, because noise in the collected projection data increases. However, considering the total exposure dose during CT fluoroscopy, it is desirable to further reduce the dose.

[0006]

As described above, the conventional computer tomography apparatus deteriorates the image quality when performing so-called CT fluoroscopy for continuously collecting projection data and reconstructing and displaying a tomographic image in almost real time. The radiation dose to the subject could not be reduced in order not to allow it.

The present invention has been made to address the above-mentioned circumstances, and an object thereof is to provide a computer tomography apparatus capable of improving the image quality of a tomographic image without increasing the exposure dose.

Another object of the present invention is to improve the image quality of a tomographic image without increasing the exposure dose when performing so-called CT fluoroscopy for continuously collecting projection data and reconstructing and displaying the tomographic image in almost real time. An object of the present invention is to provide a computer tomography apparatus capable of performing reconstruction calculation at high speed with a small calculation amount.

Another object of the present invention is a computer tomography apparatus capable of performing efficient diagnosis when performing so-called CT fluoroscopy for continuously collecting projection data and reconstructing and displaying a tomographic image in substantially real time. Is to provide.

[0010]

A computer tomography apparatus according to the present invention is a computer tomography apparatus for continuously collecting projection data at an angle of 360 ° or more,
Recursive filter means for obtaining reconstruction data based on projection data of a predetermined slice and other projection data having the same slice and the same angle, but different scan times, and an output of the recursive filter means. Means for reconstructing a tomographic image of the predetermined slice based on the above.

[0011]

[0012]

[0013]

According to the computer tomography apparatus of the present invention, a recursive filter is applied to projection data and a tomographic image is reconstructed from the filtered data, thereby reconstructing a tomographic image with good image quality even with a small dose. can do.

[0014]

[0015]

[0016]

Embodiments of the computer tomography apparatus according to the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a portion related to a reconstruction process of the computer tomography apparatus according to the first embodiment.

An X-ray tube 26 for irradiating a subject X-ray beam placed on the bed 22 with a fan-shaped X-ray beam, and a detector array arranged in an arc shape are used to detect X-rays transmitted through the subject 24. The data acquisition unit 2 having a detector 28 for detecting and collecting projection data indicating the X-ray transmittance is provided. The data acquisition unit 2 is configured to continuously collect projection data at an angle of 360 ° or more. That is,
Although not shown, the rotating portion holding the X-ray tube 26 and the detector 28 is attached to the fixed portion via a slip ring.

The operation unit 1 is connected to such a continuous rotation type data acquisition unit 2, and a control signal such as an instruction of a scan condition is supplied from the operation unit 1 to the data acquisition unit 2. here,
The scan conditions include a tube voltage, a slice width, a dose per rotation (tube current x scan time) (mAs), and the dose is not limited to the end of one rotation and can be changed at any time during the rotation. This is to cope with the case where CT fluoroscopy is performed and an important tomographic image is obtained, the dose is increased to obtain a tomographic image with good image quality.

All the projection data output from the data acquisition unit 2 is stored in the raw data storage unit 4 via the preprocessing unit 3. The pre-processing unit 3 performs pre-processing such as logarithmic conversion and correction on the projection data collected by the data acquisition unit 2,
The processed data is written in the storage unit 4 as raw data. In addition to the raw data, the raw data storage unit 4 also stores attribute data such as a dose, a projection angle, and a detector number when the raw data was collected.

The raw data read from the raw data storage unit 4 is supplied to the recursive filter unit 5. The recursive filter unit 5 applies a recursive filter to the raw data to reduce noise in the raw data, and writes the noise-reduced data as the reconstruction raw data in the reconstruction raw data storage unit 6. Specifically, the recursive filter unit 5 weights the raw data read from the raw data storage unit 4 and the reconstruction raw data read from the reconstruction raw data storage unit 6 having the same projection angle as the raw data, respectively. Multiply the coefficients and add both. The reconstruction raw data storage unit 6 stores 360 ° data necessary for reconstructing a tomographic image of one slice.

The output of the reconstruction raw data storage unit 6 is supplied to the pre-reconstruction raw data storage unit 7 and the reconstruction unit 8. The pre-reconstruction raw data storage unit 7 stores one projection data one projection before the 360 ° data stored in the reconstruction raw data storage unit 6. The reconstruction unit 8 reconstructs a tomographic image using the reconstruction raw data read from the reconstruction raw data storage unit 6 and the pre-reconstruction raw data read from the pre-reconstruction raw data storage unit 7. Constitute. The reconstructed CT image is stored in the image storage unit 9 and then displayed on the display unit 10.

The computer 11 controls the above-mentioned respective parts. The operation of the first embodiment will be described below. First, the principle of the present invention will be described. When the subject is scanned with a small dose, the projection data contains a lot of noise. In order to reduce noise, the same slice is scanned many times and these projection data are added (averaged). However, according to this method, the followability with respect to the change over time becomes poor. In CT fluoroscopy in which tomographic images of a large number of slices are reconstructed in almost real time, it is not preferable that the real time property of image display deteriorates.
Therefore, in the present invention, noise is reduced by reconstructing a CT image based on the reconstruction raw data DR obtained by applying the non-linear recursive filter to the raw data DP as shown in the following equation. .

DR (m, n) = k × DR (m−M, n) + (1−k) × DP (m, n) (1) where m is a projection number (m = 1, 2, ,), N is a data number (detector number) (n = 1 to N), and M is one revolution (36
0 °) projection number, DP (m, n) is raw data of the n-th detector of m-th projection data, DR (m, n) is raw data for reconstruction of the n-th detector of m-th projection data, DR (m-M, n)
Is the raw data for the pre-reconstruction of the nth detector of the (m-M) th projection data 360 ° before DR (m, n), and k is the weighting coefficient (k = 0 to 1) of the recursive filter. Number).

As is clear from the equation (1), when k is increased, the influence of the raw data is reduced and the influence of the reconstruction raw data before 360 ° is increased, so that the noise can be reduced. However, conversely, if the effect of the filter is strengthened in this way, followability (real-time property) to changes in raw data deteriorates, and there is a possibility that an actual tomographic image cannot be faithfully reconstructed. On the other hand, if k is reduced and the influence of raw data is increased, the effect of the filter is weakened, and noise increases. In the present invention, the weighting coefficient k (filter strength) is changed according to changes in dose and raw data. k is determined as follows.

First, the magnitude of the dose is determined. When the dose is large, the noise of the raw data is small, so k is made small to increase the influence of the raw data (to weaken the effect of the filter). That is, (1-k) is increased to increase the weight of the raw data DP (m, n). For example, when the dose is 100 mAs or more at the time of photographing, set k = 0 and set C
When the dose is 50 mAs or less due to T-transmission or the like, k is determined according to the change in the raw data.

The change in raw data is, for example, as follows:
Current raw data DP (m, n) and pre-reconstruction raw data D
It is defined by the sum of squares E of the difference of R (m−M, n). When the change amount E is large, it indicates that the data change is large. Therefore, k is decreased to increase the influence of the raw data, that is, (1−k) is increased, and the raw data DP is increased.
Increase the weight (1-k) of (m, n). On the contrary, if the change amount E is small, it means that the change is small. Therefore, it is more necessary to suppress the noise than to increase the influence of the raw data. Therefore, k is increased and the raw data D is increased.
The weight (1-k) of P (m, n) is reduced to enhance the effect of the filter.

The amount of change for determining k can be defined in various ways. For example, a method of comparing the current raw data with the raw data 360 ° before and determining k according to the difference is also possible. It is possible.

The operation of the first embodiment thus constructed will be described with reference to the flow chart shown in FIG. When the operation is started, an initial process is performed in step S1. Details of the initial processing are shown in FIG.

As shown in FIG. 3, in the initial processing,
In step S15, the process waits until the first 360 ° scan (data collection, writing to the raw data storage unit 4) is completed. When completed, the initial value of the projection number m is set to 1 in step S16.

In step S17, the raw data DP (m, n) of the m-th projection is read from the raw data storage unit 4, and step S
In step 18, these raw data DP (m, n) are stored in the raw reconstruction data storage unit 6 for reconstruction m raw data DR (m, n).
Write as n). Here, n = 1 to N.

In step S19, m is incremented (1
However, it is determined in step S20 whether m> M. If m> M is not satisfied, the process returns to step S17, the raw data of the next projection is read from the raw data storage unit 4, and is written in the reconstruction raw data storage unit 6. If m> M, the process proceeds to step S21. In this way, the first three collected
The 60 ° raw data DP is stored in the reconstruction raw data storage unit 6 as the reconstruction raw data DR.

In step S21, the data in the pre-reconstruction raw data storage unit 7 is set to 0 (initialized). Step S
The reconstruction unit 8 is activated at 22 and the tomographic image of the first slice is reconstructed at step S23. In the reconstruction in step S23, the convolution operation and the back projection operation are performed as usual based on only the 360 ° data read from the reconstruction raw data storage unit 6. The tomographic image is stored in the storage unit 9 and also displayed on the display unit 1.
Displayed as 0. This completes the initial processing.

When the initial process shown in FIG. 3 is completed, the process shown in FIG.
Returning to the flowchart of FIG. 6, m = M + 1 is set in step S2, and ms = 1 is set in step S3. The following processing is repeatedly executed until the scan is completed, but here, for convenience of explanation, the first processing will be described as an example.

In step S4, the m-th data from the raw data storage unit 4 is read.
Read the raw projection data DP (m, n) (n = 1 to 1)
N). Here, since m = M + 1, the first projection data after 360 °, that is, the second rotation, is read.

At step S5, the raw data for reconstruction 6 is stored.
Reconstructing raw data DR (ms, n) of the msth projection is read from
(Ms, n) is written in the pre-reconstruction raw data storage unit 7 as the pre-reconstruction raw data of the msth projection. Where ms
Since = 1, the first projection data is read.

In step S7, the weighting coefficient k of the recursive filter is determined. The details of this weighting factor determination process are shown in FIG. At step S24, the current raw data DP (m,
It is determined whether the dose (mAs) when n) is obtained is the dose for imaging. In the present embodiment, the dose (mAs) is reduced (for example, 50 mAs) during CT fluoroscopy for continuously observing CT images, and the dose is increased in order to memorize (take) the image in an important scene (for example, 50 mAs). , 100mAs
Since the scanning conditions can be changed at any time as described above, it is determined for each data whether it is a fluoroscopic or radiographic dose. This determination is made based on the dose data in the attribute data stored together with the raw data in the raw data storage unit 4. When the raw data is collected at a dose of 100 mAs or more for imaging, the raw data has little noise, and therefore k = 0 in step S25. As a result, substantially no filtering is applied, the raw data for reconstruction becomes equal to the current raw data, the real-time property is improved, and the time resolution is improved. Even if the dose is for radiography, k = 0,100 mA when the dose is 200 mA or more.
In the case of about s, k may be varied depending on the dose, such as k = 0.1.

When the raw data is not collected with the dose for imaging, for example, with the dose for fluoroscopy of 50 mAs or less, the variation E of the data is calculated according to the equation (2) in step S26. calculate. In step S27, k is determined according to the amount of change E using the non-linear function f.
Here, the non-linear function f for determining k is basically determined by the relationship that k becomes large (the effect of the filter becomes strong) when the change amount E becomes small and k becomes small when the change amount E becomes large. However, it is set experimentally in consideration of the region to be imaged, the dose, the number of projection data, the scanning time, the purpose of imaging, etc.

When the process of determining k shown in FIG. 4 is completed,
Returning to the flowchart of FIG. 2, the formula (1) is calculated in step S8.
According to the above, recursive filter processing is performed to obtain raw data for reconstruction. Here, DR (m−M,
2 is DR (ms, n) in the flowchart of FIG. 2, and DR (m, n) of the equation (1) is a new DR (ms, n) in the flowchart of FIG.
Then, recursive filter processing is performed according to the following equation.

DR (ms, n) ← k × DR (ms, n) + (1-k) × DP (m, n) (3) Further, in step S8, a new DR (ms, n) is projected to the msth projection. The reconstructed raw data is written in the reconstructed raw data storage unit 6. Therefore, the reconstruction raw data of the first projection in the second rotation is obtained.

The reconstruction unit 8 is activated in step S9, and the reconstruction process is performed in step S10. Step S10
In the reconstruction of, the tomographic image is reconstructed using the reconstruction raw data of 360 ° of the second projection data of the first rotation to the first projection data of the second rotation. However, here
Since the reconstruction is performed by using the continuous reconstruction method described in Japanese Patent Publication No. 1-213136, the first projection data of the second rotation and the raw data storage unit for pre-reconstruction 7 are actually used. The difference from the first projection data for the first rotation stored in
All you have to do is convolution and back projection on top of the previous tomographic image.

The reconstructed tomographic image is displayed on the display unit 10. When the reconstruction of one tomographic image is completed, step S
In step 11, ms and m are each incremented (incremented by 1), and in step S12 it is determined whether ms> M. m
When s> M, ms = 1 is set in step S13, and ms>
If it is not M, ms is left as it is. Step S14
Then, it is determined whether or not the reconstruction of the entire photographing range is completed. If the projection number m has reached a predetermined number, it is determined that reconstruction is complete, and the operation ends. If the reconstruction is not finished,
The process from step S4 is repeated.

As described above, according to this embodiment, since the continuous reconstruction method is adopted, the tomographic images are continuously reconstructed in real time based on the projection data obtained by the continuous rotation. Therefore, the tomographic image is displayed on the display unit 10 in real time, and CT fluoroscopy is performed. Scanning is performed with a small dose during this CT fluoroscopy, but since raw reconstruction data with reduced noise can be obtained in the recursive filter unit 5, a tomographic image with good image quality can be obtained even with a small dose. Can be reconfigured.
When an important tomographic image is obtained, the dose can be increased to obtain a tomographic image with good image quality.

Next, a second embodiment of the present invention will be described. Since the configuration of the second embodiment is similar to that of the first embodiment shown in FIG. 1, detailed description will be omitted. The overall operation of the second embodiment is similar to that of the first embodiment shown in FIG. 2, but the determination process of the weighting coefficient k in step S7 is different from that of the first embodiment. In the first embodiment, the weighting coefficient k is determined according to the raw data change amount E and the magnitude of the dose, but in the second embodiment, it is simply based on the dose when the raw data DP (m, n) is collected. decide. For example, as shown in FIG. 5, the dose (mA
The coefficient k is determined according to s). Here, the non-linear function f for determining k is basically determined by the relationship that k increases as the dose decreases and k decreases as the dose increases. It is experimentally set in consideration of the scan time, the purpose of photography, etc. Needless to say, the relationship between the dose and the coefficient k is not limited to the relationship illustrated in FIG.

As described above, in the second embodiment, the method of determining the weighting coefficient k is simpler than in the first embodiment. Next, a third embodiment of the present invention will be described. Since the configuration of the third embodiment is similar to that of the first embodiment shown in FIG. 1, detailed description will be omitted.

The overall operation of the third embodiment is similar to that of the first embodiment shown in FIG. 2, but the display mode is different from that of the first embodiment. In the first embodiment, the real-time tomographic images are simply displayed continuously, but in the present embodiment, as shown in FIG. 6, both the freeze image at a specific timing and the current image of the real-time tomographic image are displayed. They are displayed side by side on one screen or on two screens arranged side by side. The freeze timing is specified by the operation unit 1. According to this, the comparative diagnosis is easy and the diagnostic ability is improved. It should be noted that, as shown in FIG. 7, not only one set of freeze image and current image, but also a plurality of shot images (shot with a large dose) are arranged in time series and displayed. You may also display side by side. All of these image displays are controlled by the computer 11.

The present invention is not limited to the above-described embodiments, but can be implemented with various modifications. For example, although only the case of the single slice scan has been described, the projection data of the same position before one rotation is used even in the case of a system in which a multi-slice scan by a plurality of detectors, a shuttle scan in which slice positions are alternately moved, and a helical scan are repeated. By performing recursive filter processing using
The present invention can be similarly implemented.

[0047]

As described above, according to the present invention, 3
In a computer tomography apparatus that continuously collects projection data at an angle of 60 ° or more, the collected raw data is subjected to recursive filter processing and then reconstructed, and the weighting factor of this recursive filter is used as a dose or raw data. Determine according to changes in data. This makes it possible to reduce the influence of the filter when the dose is large, and to reconstruct a tomographic image that has a large influence of the raw data.When the dose is small, the amount of change in the data is examined, and when the amount of change is large, the filter is also filtered. The effect of is weakened, the temporal followability to the change of the raw data is improved, and when the change amount is small, the effect of the filter is strengthened to suppress the noise. Therefore, it is possible to reconstruct a tomographic image with relatively good image quality even with a small dose. Further, in the case of the same image quality, it is possible to reduce the dose by reducing the dose .

[Brief description of drawings]

FIG. 1 is a first computer tomography apparatus according to the present invention.
The block diagram of an Example.

FIG. 2 is a flow chart for explaining the operation of the first embodiment.

FIG. 3 is a flowchart showing details of the initial processing shown in FIG.

FIG. 4 is a flowchart showing the details of the weighting factor determination process shown in FIG.

FIG. 5 is a second computer tomography apparatus according to the present invention.
The figure which shows the relationship between the weight coefficient and dose in an Example.

FIG. 6 is a third of the computer tomography apparatus according to the present invention.
The figure which shows an example of the display in an Example.

FIG. 7 is a diagram showing another example of display in the third embodiment.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Operation part, 2 ... Data acquisition part, 3 ... Preprocessing part, 4 ... Raw data storage part, 5 ... Recursive filter part, 6 ... Reconstruction raw data storage part, 7 ... Prereconstruction raw data Storage,
8 ... Reconstruction unit, 9 ... Image storage unit, 10 ... Display unit, 11 ...
calculator.

─────────────────────────────────────────────────── --Continued from the front page (56) References JP-A-3-85683 (JP, A) JP-A-57-134142 (JP, A) JP-A-3-246688 (JP, A) (58) Field (Int.Cl. ⁷ , DB name) A61B 6/00-6/14

Claims (5)

(57) [Claims] 1. A computer tomography apparatus for continuously collecting projection data at an angle of 360 ° or more, wherein projection data of a predetermined slice and another projection data of the same slice at the same angle but different scan time Computerized tomography, characterized by comprising: recursive filter means for obtaining reconstruction data based on the above, and means for reconstructing a tomographic image of the predetermined slice based on the output of the recursive filter means. Imaging device. 2. The computer tomography apparatus according to claim 1, wherein the recursive filter means comprises means for determining a weight according to a dose of projection data of the predetermined slice. 3. The computer tomography apparatus according to claim 1, wherein the recursive filter means comprises means for determining a weight in accordance with a variation amount and a dose of projection data of the predetermined slice. 4. The reconstructing means is necessary for reconstructing a means for storing a tomographic image of one slice and another tomographic image having different scan times corresponding to the tomographic image stored in the storing means. The difference between the projection data and the projection data necessary for reconstructing the tomographic image of the one slice is reconstructed and calculated, and the operation result and the tomographic image stored in the storage unit are combined to obtain the other data. 2. A computer tomography apparatus according to claim 1, further comprising means for obtaining a tomographic image. 5. A real-time tomographic image reconstructed on the basis of projection data currently collected by the reconstructing means and a tomographic image reconstructed on the basis of projection data previously collected in time. Further comprises means for displaying side by side
The computer tomography apparatus according to claim 1, wherein: