JP2005073283A - Processing unit of radiographic image signal - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform a simple operation processing of spatial frequency processing on a radiographic image, and to improve an image quality deterioration by noise. <P>SOLUTION: A processing unit of a radiographic image signal obtains a digital radiographic image signal So, corresponding to a transmission dosage of each part of an object. The processing unit obtains a non-sharp mask signal Su corresponding to a predetermined low frequency from the digital radiation image signal So. Further, the processing unit subtracts a value obtained by multiplying a weak factor K by this non-sharp mask signal Su from the digital radiographic image signal So of the original. The image signal S, obtained by such a calculation, is a signal obtained by reducing the amplitude of the component of the predetermined low frequency or lower, as compared with the original. Thus, the signal of the relatively high-frequency region is frequency-enhanced. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a radiographic image signal processing apparatus, and more particularly to image processing (frequency processing) in a medical radiographic system.

  Radiation images such as X-ray images are often used for medical purposes. As a method for obtaining this radiographic image, there is a so-called radiographic system in which a phosphor layer (phosphor screen) is irradiated with radiation passing through a subject, and this film is developed by irradiating a film coated with a silver salt photosensitive material. is there.

  In recent years, with the advancement of radiological image diagnostic technology, a method has been devised in which the above radiograph is scanned, the radiographic image information recorded therein is read, converted into a digital signal, and then reproduced on a CRT or photosensitive material. I came. As a result, more diagnostic information can be obtained from a single radiographing, which improves diagnostic performance and reduces exposure dose. This method is also expected from the viewpoint of efficiency of storing and retrieving radiation image information.

In the radiographic image information reading apparatus using the photographic film, a photographic film on which a radiographic image is recorded is exposed and scanned with reading light, and the reflected light or transmitted light is detected by a photodetector and converted into an electrical signal. Is called.
On the other hand, a method for obtaining radiographic image information without using a radiographic film made of a silver salt photosensitive material has been devised. In this method, the radiation passing through the subject is absorbed by a certain type of phosphor, and then the phosphor is accumulated by the absorption by exciting the phosphor with, for example, light or thermal energy. Some of them emit radiation energy as fluorescence, and detect and image this fluorescence. Specifically, for example, it is disclosed in U.S. Pat. No. 3,859,527 or JP-A-55-12144. These show the radiation image conversion method using a stimulable phosphor and using visible light or infrared as stimulating light, and use a radiation image conversion panel in which a stimulable phosphor layer is formed on a support. Then, radiation transmitted through the subject is applied to the stimulable phosphor layer of the radiation image conversion panel to accumulate radiation energy corresponding to the radiation transmittance of each part of the subject to form a latent image. By scanning the phosphor layer with the stimulated excitation light, radiation energy accumulated in each part of the radiation image conversion panel is radiated and converted into light, and an optical signal due to the intensity of the light is multiplied by photoelectrons. Radiation image information is obtained by detection with a photoelectric conversion element such as a tube or a photodiode.

As another method, the radiation transmitted through the subject is absorbed in a semiconductor panel having a uniformly charged photoconductive layer such as selenium or silicon to form an electrostatic latent image, and then the semiconductor panel. In some cases, an electrostatic latent image on the panel is electrically detected to form an image by scanning with light (for example, see Patent Document 1).
JP 54-31219 A

  The radiological image information obtained in this way is used as it is or after being subjected to image processing such as spatial frequency processing and gradation processing for the purpose of further improving diagnostic properties by X-ray images. It is output and visualized, or stored in an image storage device such as a semiconductor storage device, a magnetic storage device, an optical disk storage device, etc., and then taken out from these image storage devices as necessary to a silver salt film, CRT, etc. Output and visualized.

As one of the spatial frequency processing methods, an unsharp mask signal Sus for each pixel is obtained, and the original image signal Sorg, the unsharp mask signal Sus and the enhancement factor β are used.
S ′ = Sorg + β (Sorg−Sus)
A processing method for enhancing the sharpness of an image by executing spatial frequency enhancement, so-called edge enhancement, is known (see Japanese Patent Publication No. 62-62373).

  However, the above spatial frequency processing method has a problem that it is difficult to perform image processing in real time because the number of operations is large. Further, in the above method, since the high spatial frequency region is emphasized, there is a problem that image noise is emphasized and image quality is deteriorated due to the image noise, particularly in a portion where the radiation dose is small (signal is small). There was a disadvantage that the image quality deteriorated remarkably.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a new frequency processing apparatus that is simple in arithmetic processing and improved in processing speed. It is another object of the present invention to provide a new frequency processing apparatus that does not cause image quality deterioration due to noise in a small portion of a signal even if the contrast (sharpness) in a high frequency region is improved. Another object of the present invention is to provide a new frequency processing apparatus that improves image quality degradation due to noise in a small portion of a signal while maintaining contrast in a high frequency region.

Therefore, the radiation image signal processing apparatus according to the present invention is configured as shown in FIG.
In FIG. 1, the imaging means detects the amount of radiation that has passed through the subject and converts it into a digital radiographic image signal, and the low frequency component attenuation means determines the low spatial frequency component of the digital radiographic image signal obtained by the imaging means. Reduce the amplitude.

  Here, the low frequency component attenuation means obtains a non-sharp mask signal corresponding to a predetermined low spatial frequency from the digital radiation image signal obtained by the imaging means, and multiplies the non-sharp mask signal by a predetermined attenuation coefficient. It is preferable to reduce the amplitude of the component below the predetermined low spatial frequency by subtracting the value obtained from the digital radiation image signal obtained from the imaging means.

  According to such a configuration, the amplitude in the low spatial frequency region of the digital radiographic image signal is attenuated as compared with the amplitude in the high spatial frequency region, so that it is possible to relatively emphasize the frequency of the signal in the high spatial frequency region. Thereby, the contrast (sharpness) of the signal in the high spatial frequency region is improved, and an image suitable for diagnosis can be reproduced.

  In addition, since the signal in the low spatial frequency region is attenuated, the change width of the radiation transmittance of the subject is reduced in a pseudo manner, and the change width of the shaded portion having a relatively large area in the radiation image is reduced. Can be made. As a result, without reducing the contrast (sharpness) in the high spatial frequency region, it is possible to reproduce the portion that easily transmits the radiation of the subject and the portion that does not easily transmit in a state that allows easy observation. is there. In addition, this improves the S / N ratio of the portion that hardly transmits the radiation of the subject, and improves the image quality degradation due to noise.

  In attenuating the amplitude in the low spatial frequency region of the digital radiographic image signal, a non-sharp mask signal corresponding to a predetermined low spatial frequency is obtained from the digital radiographic image signal, and the non-sharp mask signal is multiplied by a predetermined attenuation coefficient. If the value is subtracted from the original digital radiographic image signal, the amplitude of the spatial frequency region below a predetermined low spatial frequency can be easily attenuated with a desired characteristic.

Examples of the present invention will be described below.
FIG. 2 which shows one embodiment is a basic configuration diagram of a radiographic image capturing / reproducing apparatus including a radiographic image signal processing apparatus according to the present invention, and shows an example when applied to radiography of a human body for medical purposes.

  Here, the radiation generator 1 irradiates a subject (such as a human chest) 2 with radiation such as X-rays. A stimulable fluorescent plate 3 is provided on the side facing the radiation generating apparatus 1 with the subject 2 interposed therebetween. The stimulable fluorescent plate 3 is a radiation of the subject 2 with respect to the radiation dose from the radiation generating apparatus 1. Energy according to the transmittance distribution (passing dose) is accumulated and recorded in the photostimulable phosphor layer, and a latent image of the subject 2 is formed there.

  The photostimulable phosphor plate 3 is provided with a photostimulable phosphor layer on a support by vapor deposition of photostimulable phosphor or application of photostimulable phosphor paint. It is shielded or covered by a protective member to block adverse effects and damages caused by the environment. As the photostimulable phosphor material, for example, a material as disclosed in JP-A-61-72091 or JP-A-59-75200 is used.

On the other hand, reading of the image information from the photostimulable fluorescent plate 3 in which the radiographic image information of the subject is accumulated and recorded as described above is performed as follows.
That is, the excitation light source (gas laser, solid-state laser, semiconductor laser, etc.) 4 generates an excitation light beam whose emission intensity is controlled, and the excitation light beam is the emission light beam in which the radiation image information of the subject is accumulated and recorded. The stimulable fluorescent plate 3 is scanned, and the radiation energy (latent image) accumulated in the stimulable fluorescent plate 3 is emitted as fluorescence (stimulated luminescence).

The photoelectric conversion device 5 receives fluorescence (stimulated light emission) emitted by scanning the stimulable fluorescent plate 3 with an excitation light beam through a filter 6 that allows only the fluorescence (stimulated light emission) to pass through. Then, photoelectric conversion is performed for each scanning point into a current signal corresponding to the incident light to obtain a radiation image signal for each scanning point.
The analog radiation image signal photoelectrically read by the photoelectric conversion device 5 is sequentially A / D converted by an A / D converter (not shown) and output as a digital radiation image signal to an image processing device 7 having a microcomputer. Is done. Therefore, in this embodiment, the stimulating fluorescent plate 3, the stimulating light source 4, and the photoelectric conversion device 5 constitute an imaging means.

  The image processing device 7 performs image processing such as gradation processing on the digital radiation image signal, and performs arithmetic processing for reducing the amplitude of the low spatial frequency component according to the present invention, and outputs from the photoelectric conversion device 5. The digital radiographic image signal is output to the radiographic image reproduction device 8 after being in a form suitable for diagnosis. Therefore, in this embodiment, the image processing apparatus 7 has a function as a low frequency component attenuation means.

The radiographic image reproduction device 8 is a monitor such as a printer or a CRT, and receives a digital radiographic image signal processed by the image processing device 7 and visualizes the captured radiographic image as a hard copy or a reproduction screen.
Note that a storage device (filing system) such as a semiconductor storage device may be provided together with the radiation image reproducing device 8 or instead of the radiation image reproducing device 8 to store the read radiation image signal.

  Further, in this embodiment, the digital radiographic image signal is obtained by reading the radiographic image information accumulated and recorded on the stimulable fluorescent plate 3 as described above, but the passing dose of the subject 2 is directly measured by a CCD line sensor or the like. The configuration may be such that a digital radiation image signal is obtained by detection with a semiconductor detector. The screen film may be irradiated with radiation that has passed through the subject, and an image on the film obtained by developing the film may be read by a reading device to obtain a digital radiation image signal. The configuration of the imaging means for obtaining the image signal is not limited to the above embodiment.

Here, processing for reducing the amplitude of the low spatial frequency component performed in the image processing device 7 will be described.
The image processing device 7 stores the digital radiation image signal input from the photoelectric conversion device 5 and obtains a non-sharp mask signal Su corresponding to a predetermined low frequency from the stored data. When the stored original digital radiographic image signal is So and the attenuation coefficient is K (0 <K ≦ 1),
S = So-K · Su
To obtain a digital radiation image signal S in which the amplitude of the component equal to or lower than the predetermined low spatial frequency is attenuated, and after performing image processing such as gradation processing on the signal S, the radiation image is reproduced. Output to device 8.

  The gradation process is a process necessary for displaying a radiographic image in a diagnostically preferable state, and is used in combination with the spatial frequency process as described above. Here, the gradation processing may be performed after the spatial frequency processing, but may be performed before the spatial frequency processing. The gradation processing itself may be performed before the A / D conversion or after the D / A conversion. The signal may be processed by an analog circuit, or the digital signal may be digitally processed on a computer.

Here, in the process of obtaining the digital radiographic image signal S in which the amplitude of the component equal to or lower than the predetermined low spatial frequency is attenuated, the following process is performed to normalize the value of the modulation transfer function in the vicinity of the zero spatial frequency. May be performed.
S ′ = So / (1-K) −K · Su / (1-K)
As the non-sharp mask signal Su, it is preferable to use a non-sharp mask signal having a modulation transfer function of 0.5 or less at a spatial frequency of 0.5 cycle / mm. It is more preferable to use a non-sharp mask signal that sometimes has a modulation transfer function of 0.5 or more, and yet has a modulation transfer function of 0.5 or less at a spatial frequency of 0.5 cycle / mm.

The non-sharp mask signal Su can be created as follows.
The most basic first method is to store the original digital radiographic image signal So, read the signal of each scanning point together with the peripheral data according to the size of the non-sharp mask, and calculate the average value (simple There is a method of obtaining an unsharp mask signal Su as an average or various weighted averages).

  Secondly, after the original signal So is read from the photostimulable fluorescent plate 3 with a small-sized beam by the photostimulated excitation light source 4, the image information still remains on the photostimulable phosphor plate 3. There is a method in which a signal at each scanning point is averaged together with a signal at its peripheral portion and read out using a light beam having a large size that matches the size of the sharp mask signal.

  Third, the readout light beam uses the fact that its beam diameter gradually expands due to scattering in the phosphor layer, and the original signal So is created from the light emission signal from the incident side of the light beam. There is a method of generating an unsharp mask signal Su by light emission on the side where the light beam is transmitted. In this case, the size of the unsharp mask signal can be controlled by changing the degree of light scattering of the phosphor layer or changing the size of the aperture for receiving the light.

As a fourth method, there is a method in which an analog signal is low-pass filtered in the beam scanning direction (main scanning direction) and an A / D-converted digital radiation signal is averaged in the sub-scanning direction.
Here, in the amplitude attenuation of components below a predetermined low spatial frequency using the unsharp mask signal Su, the value is 0.2 when the modulation transfer function M0 near the zero spatial frequency before the amplitude is attenuated is used as a reference. Attenuation to satisfy ≦ M0 ≦ 0.9 is preferable for providing an image suitable for diagnosis, and 0.3 ≦ M0 ≦ 0.8 is more preferable. Furthermore, when 0.68 ≦ M0 ≦ 0.80, the obtained image can be used for diagnosis without a sense of discomfort compared to a system using a conventional screen film.

Therefore, the attenuation coefficient K is preferably 0.1 ≦ K ≦ 0.8, and more preferably 0.2 ≦ K ≦ 0.7. The most preferable range of the attenuation coefficient K is 0.20 ≦ K ≦ 0.32.
Note that the shape of the unsharp mask signal determined by whether the modulation transfer function is 0.5 at any spatial frequency fc of 0.5 cycles / mm or less (see FIG. 3), or the value of the attenuation coefficient K is determined by processing. Must be specified before doing. These values may be individually input using an external input device, or may be selectively input when an imaging region or the like is designated from among a plurality of preset portions depending on a human body region or case, or For example, a method of analyzing the read image data and automatically setting a preferable value is assumed.

If the value obtained by multiplying the unsharp mask signal Su by the attenuation coefficient K as described above is subtracted from the original digital radiation image signal So (see FIG. 3), as shown in FIG. 4, a predetermined low spatial frequency is obtained. An image signal in which the amplitude of the following signal components is attenuated relative to the original signal can be obtained.
If the amplitude of the signal component below a predetermined low spatial frequency is attenuated, for example, in the case of chest imaging of a human body, there are portions that are likely to transmit radiation and portions that are difficult to transmit, such as lung field and mediastinum. Even if they exist at the same time, the image after the signal processing is an image in which the change width of the radiation transmittance is reduced in a pseudo manner.

  Therefore, it is possible to reproduce a portion that easily transmits radiation and a portion that does not easily transmit radiation in a state that allows easy observation, and the amplitude of the signal component in the spatial frequency region other than the low spatial frequency region is Since it is the original, contrast in a high spatial frequency region is ensured, and diagnostic information such as blood vessels and small lesions is not crushed by the frequency processing, and an image with excellent diagnostic properties can be obtained.

In addition, when the amplitude is attenuated to a relatively high spatial frequency region, the signal amplitude in the high spatial frequency region is relatively large compared to the middle and low frequency regions, and the image edge portion is emphasized. It becomes. Therefore, with such frequency processing, the contrast (sharpness) of the image can be apparently improved.
Next, a more specific state of the frequency processing will be described with reference to FIGS.

  3 shows the spatial frequency response characteristic of the digital radiographic image signal obtained when the radiographic image accumulated on the photostimulable phosphor plate 3 is sampled at 100 μm / 1 pixel. Su represents the spatial frequency response characteristic of the Gaussian unsharp mask signal set so that the modulation transfer function at 0.5 cycle / mm is 0.5 or less. In this example, when the above image sampled at 100 μm / 1 pixel is averaged for 80 pixels × 80 pixels, a non-sharp mask signal is created by weighting in a Gaussian distribution. In the figure, fc is a spatial frequency when the modulation transfer function is 0.5.

The non-sharp mask signal Su may be created by omitting weighting and taking a simple addition average. FIG. 5 shows the spatial frequency response characteristics of the rectangular unsharp mask signal in that case. However, the shape of the non-sharp mask does not need to be a square like 80 pixels × 80 pixels, and may be a rectangle.
FIG. 4 shows the calculation result S of (So−K · Su). In this example, the attenuation coefficient K is set to 0.3. The attenuation coefficient K is normally fixed to a predetermined value. However, if the attenuation coefficient K is changed according to the magnitude of the original digital radiation image signal So or the unsharp mask signal Su, the low signal range and the high signal are set. The degree of amplitude attenuation of the signal in the low spatial frequency region can be changed depending on the region, and an image with improved diagnostic performance can be provided.

  As described above, according to the present invention, by reducing the amplitude of the low spatial frequency component of the radiographic image signal, it is possible to relatively emphasize the frequency of the signal in the high spatial frequency region, and the signal in the high spatial frequency region. The contrast is improved. In addition, reducing the amplitude of the low spatial frequency component artificially reduces the range of change in the radiation transmittance of the subject, and simultaneously observes the portion that is likely to transmit the subject's radiation and the portion that is difficult to transmit. It is possible to reproduce in an easy state. Therefore, particularly in the case of medical radiographic images, it is possible to obtain a reconstructed image suitable for diagnosis, and the diagnostic performance using the radiographic image can be improved.

The block diagram which shows the structure of this invention. BRIEF DESCRIPTION OF THE DRAWINGS The system schematic which shows one Example of this invention. The diagram which shows the spatial frequency response characteristic of the original radiographic image signal So and an unsharp mask signal. The diagram which shows the spatial frequency response characteristic of the result of having attenuated the amplitude of the low spatial frequency component. The diagram which shows the spatial frequency response characteristic at the time of producing by a simple addition average, without performing weighting of a non-sharp mask signal.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Radiation generator 2 Subject 3 Stimulable fluorescent plate 4 Stimulated excitation light source 5 Photoelectric conversion device 7 Image processing device 8 Radiation image reproduction device

Claims (3)

Imaging means for detecting a radiation amount that has passed through the subject and converting it into a digital radiation image signal;
Low frequency component attenuation means for attenuating the amplitude of the low spatial frequency component of the digital radiographic image signal obtained by the imaging means,
The low-frequency component attenuation means transmits a modulation transfer from the digital radiographic image signal when the modulation transfer function is 0.5 or less and a spatial frequency of 0.01 cycle / mm when the spatial frequency is 0.5 cycle / mm. Obtaining a non-sharp mask signal corresponding to a predetermined low spatial frequency having a function of 0.5 or more;
An attenuation coefficient K selected from the range of 0.1 ≦ K ≦ 0.8 is determined by inputting using an external input device,
A radiographic image signal processing apparatus, wherein a value obtained by multiplying the non-sharp mask signal by the attenuation signal K is subtracted from the digital radiographic image signal. Imaging means for detecting a radiation amount that has passed through the subject and converting it into a digital radiation image signal;
Low frequency component attenuation means for attenuating the amplitude of the low spatial frequency component of the digital radiographic image signal obtained by the imaging means,
The low-frequency component attenuation means transmits a modulation transfer from the digital radiographic image signal when the modulation transfer function is 0.5 or less and a spatial frequency of 0.01 cycle / mm when the spatial frequency is 0.5 cycle / mm. Obtaining a non-sharp mask signal corresponding to a predetermined low spatial frequency having a function of 0.5 or more;
Attenuation coefficient K selected from the range of 0.1 ≦ K ≦ 0.8 is determined by selectively inputting an imaged region or the like from among a plurality of portions set in advance depending on the region or case of the subject. ,
A radiographic image signal processing apparatus, wherein a value obtained by multiplying the non-sharp mask signal by the attenuation signal K is subtracted from the digital radiographic image signal. Imaging means for detecting a radiation amount that has passed through the subject and converting it into a digital radiation image signal;
Low frequency component attenuation means for attenuating the amplitude of the low spatial frequency component of the digital radiographic image signal obtained by the imaging means,
The low-frequency component attenuation means transmits a modulation transfer from the digital radiographic image signal when the modulation transfer function is 0.5 or less and a spatial frequency of 0.01 cycle / mm when the spatial frequency is 0.5 cycle / mm. Obtaining a non-sharp mask signal corresponding to a predetermined low spatial frequency having a function of 0.5 or more;
When the value obtained by multiplying the non-sharp mask signal by an attenuation signal K selected from the range of 0.1 ≦ K ≦ 0.8 is subtracted from the digital radiation image signal, the digital radiation image signal or the non-sharp mask signal An apparatus for processing a radiographic image signal, wherein the attenuation signal K is changed in accordance with a sharp mask signal.

Images