JP3196033B2 - Bone radiation image region of interest detection device and image processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for detecting a region of interest of a bone radiographic image and an image processing apparatus. More specifically, the present invention relates to a method of detecting a bone region as a region of interest in an image from radiation image data of a subject including bone. The present invention relates to a device that performs extraction and image processing that makes a bone region more visible based on the extraction result.

[0002]

2. Description of the Related Art Radiation images such as X-ray images are widely used for diagnosing diseases and the like. To obtain such X-ray images, X-rays transmitted through a subject are applied to a phosphor layer (fluorescent screen). A so-called radiograph, which is obtained by irradiating a visible light to thereby generate a visible light and irradiating the visible light to a film using a silver salt in the same manner as a normal photograph and developing the film, has been conventionally used.

However, in recent years, a method has been devised for directly taking out an image from a phosphor layer without using a film coated with a silver salt. This includes:
The radiation transmitted through the subject is absorbed by the phosphor, and then the phosphor is excited by, for example, light or heat energy, thereby radiating the radiation energy accumulated by the phosphor as a result of the absorption. There is a method of obtaining an image signal by photoelectrically converting

Specifically, for example, US Pat. No. 3,859,527 and JP-A-55-12144 disclose a radiation image conversion method using a stimulable phosphor and using visible light or infrared light as stimulating excitation light. ing. This method uses a radiation image conversion panel having a stimulable phosphor layer formed on a support,
Radiation transmitted through the subject is applied to the stimulable phosphor layer of this conversion panel to accumulate radiation energy corresponding to the radiation transmittance of each part of the subject to form a latent image. By scanning with the excitation light, the radiation energy accumulated is emitted, converted into light, and this optical signal is photoelectrically converted to obtain a radiation image signal.

The radiation image signal thus obtained is output as it is or after being subjected to image processing to a silver halide film, a CRT, or the like, and is visualized. Often it is. Further, the digitized radiation image signal is stored in an image storage device such as a semiconductor storage device, a magnetic storage device, an optical disk storage device, and a magneto-optical storage device. In some cases, it is output to a silver halide film, a CRT, or the like and visualized.

A silver halide film on which a radiation image has been recorded is irradiated with light from a light source such as a laser or a fluorescent lamp to obtain light transmitted through the silver halide film. There is also a method of obtaining a signal and further digitizing it. As described above, the configuration of the apparatus for obtaining a digital radiation image signal from a silver halide film on which a radiation image has been recorded is such that a light beam is one-dimensionally scanned on the silver halide film, and the silver halide film is scanned in the scanning direction. Transported in a direction perpendicular to the light source, and configured to detect transmitted light with a photodetector provided on the side opposite to the light source, or a silver halide film with a radiation image recorded on the side of a transparent drum containing a light source. Attachment, rotating the drum, and moving the aperture that guides the transmitted light to the photodetector in parallel with the rotation axis of the drum are also available.

By the way, when reproducing the radiation image signal obtained as described above, the purpose of making the density of the region of interest (the region including the image portion necessary for diagnosis in medical treatment) constant in the reproduced image, and After performing image processing such as gradation processing and spatial frequency processing in order to output shadows (regions of interest) of the human body structure and lesions more easily, the CRT
Etc. for visualization and use for diagnosis.

[0008]

In the above-described image processing, it is desired to output an image signal of a region of interest under good conditions and to reproduce the region of interest in an easily viewable manner. It is necessary to specify a corresponding image signal and determine a processing condition based on the specified result.

For example, when radiographing a subject including bones such as a human leg and diagnosing a bone state, it is necessary to perform image processing so that the bones of interest are easily seen. As a method of specifying a corresponding image signal, a method of obtaining the image signal from a histogram of a radiation image signal as shown in FIG. 14 is known. That is, since the bone has a small amount of radiation transmission, among the image signals proportional to the amount of radiation transmission, the range from the minimum image signal (min) on the histogram to a predetermined value applied to the imaging of the bone portion corresponds to the image corresponding to the bone region. There is a method of determining a signal range and determining image reading conditions or image processing conditions so that the image signal range is easy to see (see Japanese Patent Application Laid-Open No. 61-280163).

However, since the signal distribution area occupied by the bone part varies greatly depending on the photographing position and body type, the signal range from the minimum image signal on the histogram to a constant signal value corresponds to the bone area as described above. In the method defined as the above, the image signal distribution corresponding to the bone portion cannot be obtained with high accuracy, and the optimum image processing conditions cannot always be set.

The present invention has been made in view of the above problems, and accurately extracts a bone region, which is a region of interest, from an image region without being affected by a photographing position or a body shape.
An optimal image processing condition can be determined based on the extracted image data in the bone region, so that the subject including the bone is reproduced in a state where the bone as the region of interest is easily seen. The purpose is to.

[0012]

SUMMARY OF THE INVENTION Therefore, an apparatus for detecting a region of interest in a bone radiation image according to the present invention provides image data of a bone radiation image formed in accordance with the amount of radiation transmitted through a subject including a bone. This is a device for detecting a bone region as a region of interest from the apparatus, and is configured as shown in FIG.

In FIG. 1, the binarizing means binarizes the image data by comparing the image data of the bone part radiation image with a threshold to divide the bone part radiation image into a plurality of image areas. Further, the threshold setting unit gradually increases or decreases the threshold from a predetermined initial value.

Here, the specific threshold value detecting means sets an image area consisting of image data smaller than the threshold value as a detection area when the threshold value is increased and changed by the threshold value setting means, and conversely, an image area larger than the threshold value when the threshold value is reduced and changed. An image area composed of data is defined as a detection area, and a threshold when the detection areas are divided into a plurality of pieces when the threshold value is near the predetermined initial value has a predetermined proximity relationship with a threshold value change is a specific threshold value. Detected as

[0015] The bone region detecting means detects, as a bone region, an image region composed of image data corresponding to the smaller radiation transmission amount when binarizing the image data with the detected specific threshold value. Here, the predetermined proximity relationship at the time of detecting the specific threshold value may be a time when the divided detection regions shift from the non-connection state to the connection state with a change in the threshold value.

On the other hand, the image processing apparatus for bone radiation images according to the present invention is similarly configured as shown in FIG. That is, the processing condition determining means determines the image processing condition based on the image data included in the bone region detected as described above, and the image processing means determines the bone radiation based on the determined image processing condition. Image processing is performed on the image data of the image.

[0017]

In a radiographic image of a subject including a bone, the amount of radiation transmitted through the bone part is significantly smaller than that of other parts. Therefore, if the image data is binarized based on an appropriate threshold value, And the other areas. However, an appropriate threshold value for classifying the above-mentioned bone region is affected by an imaging part, a body shape, and the like, and changes. Therefore, an appropriate threshold value for extracting a bone region by devising as follows is set for each image. Detectable.

That is, as a radiographic image of a subject including a general bone, a part of a continuous bone such as a leg or a hand is photographed by being separated by an image frame, and both sides of the bone are exposed to flesh and direct radiation. It is assumed that an image is a part of the incident light. In such an image, for example, assuming that the threshold value in the binarization processing is changed in the direction of increasing the area of the region not including the bone (the threshold value is changed from a higher radiation transmission amount to a lower radiation amount), The area (detection area) of the pixel group consisting of the data of (1) starts from the blank part according to the threshold value change, absorbs the flesh part and expands, and eventually has a relatively large radiation transmission amount in the bone. The part (for example, a joint part or a thin part) is absorbed so that the regions (detection regions) having relatively large radiation transmission amounts, which are divided on both sides of the bone, are interconnected.

Conversely, if the threshold value is changed in a direction to increase the area of the region including the bone (the threshold value is changed from a lower radiation transmission amount to a higher radiation transmission amount), data larger than the threshold value may be obtained. Area of pixel group consisting of (detection area)
Starts with the part with the lowest radiation penetration in the bone, gradually expands its area, and absorbs all the bone parts last with the part with the largest radiation transmission in the bone That is, for example, in an image including a joint, regions including bones separated on both sides of the joint are connected at the joint by changing the threshold.

As described above, by changing the threshold value in the increasing direction or the decreasing direction, the detection region as the region including the bone or the region not including the bone becomes relatively transparent in the bone from the divided state. The connection is performed via the portion with a large dose, and the threshold value at the time of such connection or near the time of connection can be specified as the optimum threshold value that can distinguish the bone and other regions.

As described above, a bone region in an image is detected in each image. Therefore, if image processing conditions are determined based on image data in the detected bone region, the image processing condition and the physique of the subject are affected. It is possible to perform image processing that makes it easier to see the bones, which are regions of interest, without having to do so.

[0022]

Embodiments of the present invention will be described below. FIG. 2 showing an embodiment is a medical radiation image recording / reading apparatus including a device for detecting a region of interest of a bone radiation image and an image processing device according to the present invention. The following description will be made on the assumption that the image of the leg M is reproduced, and the captured image of the leg is reproduced to diagnose the bone of the lower limb, which is the region of interest.

Here, the radiation source 1 is controlled by the radiation control device 2 to emit radiation (generally X-rays) toward the leg M which is the subject. Record reading device 3
Is provided with a radiation image conversion panel 4 on a surface facing the radiation source 1 with the leg M interposed therebetween, and the conversion panel 4 follows the radiation transmittance distribution of the leg M with respect to the irradiation radiation amount from the radiation source 1. Energy stored in the stimulable phosphor layer,
A latent image of the leg M is formed there.

The conversion panel 4 has a stimulable phosphor layer provided on a support by vapor deposition of a stimulable phosphor or application of a stimulable phosphor paint. The layer is shielded or covered by a protective member to prevent adverse effects and damage due to the environment. As the stimulable phosphor material, for example, materials disclosed in JP-A-61-72091 or JP-A-59-75200 are used.

A light beam generator (gas laser, solid-state laser, semiconductor laser, etc.) 5 generates a light beam whose emission intensity is controlled, and the light beam reaches a scanner 6 via various optical systems. , Where it is deflected,
To deflect the light path, and is guided to the conversion panel 4 as stimulating excitation scanning light. The light collector 8 has a light collecting end made of an optical fiber positioned close to the conversion panel 4 on which the stimulating excitation light is scanned, and is proportional to the latent image energy from the conversion panel 4 scanned by the light beam. The photostimulable light having the light emission intensity is received. Reference numeral 9 denotes a filter that passes only light in the photostimulated emission wavelength region from the light introduced from the light collector 8, and the light that has passed through the filter 9 enters the photomultiplier 10 and corresponds to the incident light. The current signal is photoelectrically converted into a current signal.

The output current from the photomultiplier 10 is converted into a voltage signal by a current / voltage converter 11, amplified by an amplifier 12, and then converted into digital data (digital radiation image signal) by an A / D converter 13. Is done. The digital image signal proportional to the radiation transmission amount of each part of the subject is sequentially image-processed in the image processing device 14, and the image signal after the image processing is transmitted to the printer 17 via the interface 16. I have.

Reference numeral 15 denotes a CPU for controlling image processing in the image processing apparatus 14, and various image processing (for example, spatial frequency) including gradation processing for digital radiation image data output from the A / D converter 13. Processing, enlargement, reduction, movement, rotation, statistical processing, etc.) in the image processing device 14 and output to the printer 17 in a form suitable for diagnosis.
The printer 17 can obtain a hard copy of the leg radiation image.

Note that what is connected via the interface 16 may be a monitor such as a CRT or a storage device (filing system) such as a semiconductor storage device. Reference numeral 18 denotes a read gain adjusting circuit, which adjusts the light beam intensity of the light beam generator 5, the gain of the photomultiplier 10 by adjusting the power supply voltage of the photomultiplier high-voltage power supply 19, and the current / voltage converter. The gain of the amplifier 11 and the amplifier 12 and the adjustment of the input dynamic range of the A / D converter 13 are adjusted, and the reading gain of the radiation image signal is adjusted comprehensively.

The part of the image processing device 14 relating to the bone region (region of interest) detection and gradation processing according to the present invention is specifically configured as shown in FIG. That is, the digital radiation image signal of the leg M photoelectrically read from the stimulable phosphor layer of the conversion panel 4 is converted into a binarizing means for extracting a bone region as a region of interest before the gradation processing. The binarization unit 21 performs binarization by comparing with a predetermined threshold value. The threshold value used for the binarization is set to be gradually increased or decreased from an initial value by a threshold value setting unit 22 as a threshold value setting unit. Is set to each.

The image signal binarized by the binarizing section 21 is divided into an image area into an area composed of a pixel group of a signal larger than a threshold and an area composed of a pixel group of a signal smaller than the threshold. The detection region setting unit 23 sets a detection region as a region of interest to detect a bone region from among the plurality of regions, and performs a specific threshold detection unit and a bone region detection unit according to the setting. The detection area connectivity determination unit 24 determines the optimal threshold for roughly dividing the bone area and other areas by determining the division / connection of the detection area.

That is, for example, if an identification signal of 1 is assigned to a pixel whose image signal (radiation transmission amount) is larger than a threshold and an identification signal of 0 is assigned to a pixel smaller than the threshold, the threshold setting unit 22 If the threshold value is gradually changed, one of the region consisting of the one pixel group and the region consisting of the zero pixel group will have its area enlarged, and the other will have its area relatively reduced. When the threshold value reaches the boundary signal level between the bone region and another region, the region consisting of the pixels (pixels having a smaller radiation transmission amount than the threshold level) to which the identification signal of 0 is attached is substantially a bone region. Will be shown.

Here, before the threshold value reaches the optimum value for distinguishing between the bone region and the other region, the region on the side where the area is increased is divided through other regions. When the threshold value is reached near the optimum threshold value, the bone has a characteristic of being connected via a portion of the bone having the least amount of radiation transmission (for example, the thinnest portion or a knee joint portion). Is determined based on whether or not the areas to which the same identification signal, which has been divided, are connected to each other. As a criterion for determining connectivity, a pixel included in one region may be located at one of four pixel positions adjacent to the pixel included in the other region in the up, down, left, and right directions or in an eight pixel position surrounding the pixel. If it exists in any of them, it can be regarded as a connection.

That is, in the photographing of the leg M, generally, as shown in FIG. 5, a part of the bone is photographed by being divided by an image frame, and both sides of the bone become flesh and omission parts. In such an image, for example, when the threshold value is reduced (the threshold value is changed from a higher radiation transmission amount to a lower one), a region including a pixel group having a signal value equal to or larger than the threshold value is detected in order to detect an optimal threshold value. When the threshold value is near the low initial value of the signal level, this detection region corresponds to a blank portion or a thin portion on both sides of the bone, and is thus divided on both sides of the bone. However, in response to a decrease in the threshold value, a portion having a smaller amount of transmission such as a meat portion is absorbed and expanded, and a portion having a relatively large amount of radiation transmission within the bone (for example, a knee joint portion or a thin portion) Portions, etc.), so that the detection regions that were separated on both sides of the bone will be interconnected.

As described above, when the detection areas are connected in a state where the detection areas are separated, it is considered that the threshold value at that time is set slightly beyond the optimum area that separates the bone area from the other areas. Therefore, the non-connection / connection of the detection area is determined, and the threshold value at the first transition to the connection state is set as the specific threshold value for extracting the bone area.

When the threshold value is increased and changed, a region composed of a group of pixels having a signal value smaller than the threshold value is set as a detection region, and the detection region is divided by a portion of the bone having a relatively large radiation transmission amount. The threshold value at the time of connection from the closed state is set as a specific threshold value for extracting a bone region. When the detection region connectivity determination unit 24 detects that the detection regions are connected to each other, the threshold at that time is held as a specific threshold, and a signal smaller than the threshold when binarized by the specific threshold is used. Is detected as a bone region.

When the bone region is detected on the basis of the binary image based on the specific threshold value in this way, the gradation processing condition determining unit 25 as the processing condition determining means is operated based on the image signal included in the bone region. The gradation processing conditions are determined so that the region is easy to see in the diagnosis, and the gradation processing unit 26 as an image processing means performs gradation processing of the original radiation image signal according to the determination and outputs the signal to the printer 17.

The image signal used in each unit other than the gradation processing unit 26 does not need to be performed by using the total number of pixels read by the recording and reading device 3, but by using an image signal thinned out from the original image signal. The process for determining the gradation processing conditions may be simplified. Also, Japanese Patent Application Laid-Open No. Sho 58-6
The detection of the bone region and the determination of the gradation processing condition are performed based on the image signal obtained by “read ahead” as disclosed in US Pat. No. 7,240, and an image obtained by “main read” according to the determined processing condition. The signal may be processed. The `` read-ahead '' is to read an image signal by stimulating emission with excitation light having a lower energy than the `` read-ahead '' prior to the `` read-ahead '' provided for observation reading, which is obtained by the `` read-ahead '' In response to the determination of the gradation processing condition based on the image signal obtained, the gradation processing is performed on the image signal obtained in the “main reading”.

The printer 17 is constructed, for example, as shown in FIG. In FIG. 4, the digital radiation image signal read via the interface 16 is first subjected to various signal correction processing by a signal correction circuit 31 via a buffer memory 30, and then by a D / A converter 32. It is converted to an analog signal. Then, in order to modulate the laser beam in accordance with the analog signal, the output of the D / A converter 32 is input to a modulator driving circuit 33, and the modulator driving circuit 33
A drive voltage corresponding to the output level of the output 32 is output to the optical modulator.

The light modulator 34 modulates the laser light emitted from the laser light source 35 according to the image signal level based on the drive voltage, and the modulated laser light is rotated by a motor (not shown). Polygon mirror) 36
Are reflected on the polygonal reflecting surface and are distributed in the main scanning direction. Note that a galvanometer mirror may be used as the deflection mirror.

The reflected light from the deflecting mirror 36 is transmitted to an fθ lens
The two-dimensional radiation image is recorded on the recording medium 38 by being adjusted to a constant scanning speed after passing through the recording medium 38 and being received by the recording medium (photosensitive material) 38 conveyed in the sub-scanning direction. After that, by developing the recording medium 38, a hard copy of the digital radiation image can be obtained.

Next, the state of the region of interest detection and image processing according to the present invention performed in the image processing device 14 whose detailed configuration is shown in FIG. 3 will be described in detail. First, the threshold setting unit 22 sets an initial value of a threshold of an image signal used when binarizing a radiation image signal. In the present embodiment, the purpose is to extract the bone region by specifying the optimal threshold that can distinguish between the bone region and the other as described above. It is necessary to gradually change the threshold in the direction of the optimal threshold from the larger or smaller one.

Therefore, the initial setting of the threshold value is performed as follows. In the case where the threshold is gradually increased, in this case, since the initial value needs to be smaller than the optimum threshold, the initial value may be set to the minimum value of the signal range that can be treated as data. The deviation between the value and the initial value increases, and the processing time is increased.

Therefore, for example, as shown in FIGS.
A profile prof1 of the pixel row at the top (or bottom) of the image signal and a profile prof2 of the pixel row at the left end (or right end) are obtained, and the profile prof1 of the pixel row across the bone is selected from the shapes of these profiles. Then, the minimum value in the profile prof1 is set as an initial threshold value. That is, the threshold value is set to be increased with the portion of the bone having the lowest radiation transmittance as the initial level.

It should be noted that the vertical and horizontal profiles are obtained separately as described above in order to ensure that the profile of the pixel row that crosses the bone is obtained even when the direction of the bone cannot be specified in advance. In addition to the method using the profile as described above, a histogram of the image signal is created, and the minimum value (see FIG. 8) of the effective image signal range (the image signal range excluding the blank portion) of the histogram is set to the initial value. Alternatively, a signal value of a predetermined percentage of the cumulative histogram may be set as an initial value.

On the other hand, when the threshold is lowered from the larger one to obtain the optimum threshold, it is necessary to set a value larger than the optimum threshold as an initial value. Therefore, the maximum value of the signal range may be set as the initial value. For the same reason as described above, it is desired to set an initial value according to the characteristics of each image. The threshold value for decreasing the threshold value from the larger value can be obtained from the profile, the histogram, and the cumulative histogram in the same manner as the initial value for the increase setting described above. The initial value may be set as a value smaller than the maximum value of the profile by a predetermined minute value (a value slightly smaller than the signal level of the background), and when a histogram is used, the maximum value of the effective image signal range is set. Is used as the initial value, and when a cumulative histogram is used, a predetermined percentage of the signal value as the initial value may be set in advance.

The threshold setting unit 22 increases or decreases the threshold in a stepwise manner from the initial value set as described above, and the binarization unit 21 uses the threshold set by the threshold setting unit 22 to convert the image signal. Although the binarization is performed, such binarization processing does not necessarily need to be performed on the entire image, and is preferably performed in a limited area in order to simplify the processing. When the leg M is photographed, the leg M is usually photographed so that the leg M is located at the center.
10) may be set as an area for binarization, or a limited area for binarization as shown in FIG. 9 or FIG. 10 may be individually set for each image based on profile information. In particular, 2
In the binarization process, it is useless to binarize the blank part, so that a line segment excluding the vertical blank area based on the profile information in the direction crossing the bone as shown in FIG. 9 is used. It is preferable that L3 and L4 are set so that only the subject portion sandwiched between the line segments L3 and L4 is a binarized area.

The binarizing unit 21 compares the image signal corresponding to each pixel in the limited area with the threshold value, and discriminates a pixel having a value equal to or larger than the threshold value from a pixel having a value smaller than the threshold value.
Specifically, pixels having a threshold value or more are assigned an identification code of 1, for example, and pixels having a value less than the threshold value are assigned an identification code of 0, and the image in the region is composed of “1” and “0”. Let it be a binary image.
The identification code may be on / off.

When “1” or “0” is assigned to each pixel as described above and binarized, the detection area setting unit 23 next performs labeling of the area based on the sign of each pixel. The detection area is divided into a plurality of areas as a set of the same identification code, and a detection area to be noticed for obtaining an optimum threshold value is set from these areas. Here, when the threshold value is increased and changed, a region including a pixel group having a signal value smaller than the threshold value is a detection region to be noted in order to obtain an optimum threshold value. If they are consecutive, the same label is given to all the pixels, and different labels are given to each of the plurality of detection regions whose signal values are less than the threshold value.

Then, the detection area connectivity determination unit 24
As shown in, when two labeled detection regions (regions including bones) in contact with the line segments L1 and L2 crossing the bones are connected for the first time in accordance with an increase in the threshold from the mutually separated state, The threshold at that time is held. That is, when the threshold value is increased and changed, only pixels corresponding to a portion having a low transmitted dose in the bone are initially identified as being less than the threshold value, but pixels having a higher signal value are also increased with the increase in the threshold value. It is now identified as being below the threshold, and the region consisting of pixels below the threshold absorbs and expands a relatively large portion of the transmitted dose in the bone, and finally has the largest transmitted dose in the bone When the joint is identified as being smaller than the threshold value, the joint is vertically connected via the joint (see FIG. 9).

In the connected state, the bone region is included in the pixel group (detection region) identified as being smaller than the threshold value.
The threshold at that time is set to a specific threshold corresponding to an optimum threshold for distinguishing a bone region from another region. In the above embodiment, a region including a pixel group smaller than the threshold value, which is in contact with the two line segments L1 and L2 crossing the bone, is focused on as a region including the bone.
The threshold when the detection areas are connected to each other from the divided state is set as the specific threshold. However, as shown in FIG. 10, pixels A having values close to the initial value of the threshold near the upper and lower ends of the image are respectively provided. , B are considered to be bone-containing regions, and the threshold when these regions are connected for the first time is set to the specific threshold. good.

On the other hand, when the threshold value is decreased and changed, a region (a region not including a bone) composed of a pixel group having a signal value larger than the threshold value is a region to be noted. A different label is given to each of the pixel groups for which. Then, of the labeled regions, two regions that are in contact with the two line segments L3 and L4 sandwiching the bone are determined as detection regions including a flesh portion or a blank portion, and the detection region connectivity determination unit 24 Is expanded as the threshold decreases, and the threshold when connected is set as the specific threshold.

When the threshold value is reduced and changed, a region composed of a pixel group having a signal value larger than the threshold value includes only a blank portion or a relatively thin portion when the threshold value is near the initial value. However, as the threshold value decreases, pixels with a smaller transmitted dose are absorbed, and the first bite into the bone region is a joint portion of the bone that has a relatively large transmitted dose. Therefore, as described above, when two regions consisting of a group of pixels larger than the threshold, which is divided on both sides of the bone, are connected via the joint portion, the pixel whose signal value is smaller than the threshold reduced by the threshold change is used. The region composed of groups includes only a substantially bone region.

As described above, when the threshold value is near the initial value, the two divided detection regions (regions including bones or regions not including bones) are compared with each other in the bones according to the change in the threshold value. The state when connected via a part (joint part) with a large target transmitted dose is a state in which the bone region and other regions are substantially divided, and the threshold at this time is a specific threshold from which the bone region can be extracted. When the image signal is binarized using the specific threshold, a region including a pixel group identified as having a signal value smaller than the specific threshold (a hatched portion shown in FIGS. 9 and 10 and a white portion shown in FIG. 11) Indicates a substantially bone region.

If the bone region is detected as described above, the bone region, which is the region of interest, can be accurately detected in each of the images without being affected by the imaging site, the physique, and the like. The image processing conditions of (1) can be accurately set based on the signal value of the bone region, and appropriate image processing can be stably performed. In the above embodiment, the threshold value at the time when the divided regions are connected to each other is determined as the specific threshold value at which the bone region can be extracted.However, it is not necessary to change the threshold value until the connection is performed. The threshold at the time when the shortest distance between the two detection areas is equal to or less than a predetermined value may be set as the specific threshold, or the threshold at the time when the predetermined proximity state including the connected state is reached may be set as the specific threshold.

When a bone area (a specific threshold from which a bone area can be extracted) is detected in each image as described above, the gradation processing condition determination unit 25 determines whether an image signal in the bone area (pixels below the specific threshold) has been detected. The gradation processing conditions are set based on the group signals). In determining such gradation processing conditions, for example,
As shown in FIGS. 12 and 13, a histogram of the image signal in the extracted bone region is created, and the maximum value Smax and the minimum value Smax of the image signal in the bone region are created from the histogram.
min, and the maximum value Smax is the maximum density Dmax.
And the minimum value Smin is the minimum density Dmin
The gradation conversion table is set so that In this way, by assigning the minimum density Dmin to the maximum density Dmax to the signal range in the bone region, it is possible to finish the bone region with an easy-to-see contrast, and to perform the bone diagnostic performance based on the reproduced image. Can be improved.

The gradation processing conditions are determined in addition to the above method.
A gradation conversion table is set so that the average value of the image signals in the bone region has a predetermined density, or a cumulative histogram of the image signals in the bone region is created. What is necessary is just to set the gradation conversion table so that the value becomes a predetermined density, and the method of determining the gradation processing condition based on the extracted image signal in the bone region is not limited.

In setting the gradation conversion table as described above, as disclosed in JP-A-59-83149, the basic characteristic curve of gradation conversion is rotated on coordinates, By moving, a characteristic curve of gradation conversion matching the characteristics of the image signal in the bone region may be obtained. Further, in the above embodiment, the gradation processing is described as an example of the image processing.
The condition of the spatial frequency processing as disclosed in Japanese Patent Laid-Open No. 62376 may be set based on the image signal in the bone region extracted as described above. That is, the enhancement parameter (emphasis coefficient β in Japanese Patent Publication No. 62-62376) is set such that the enhancement degree is relatively large with respect to the image signal of the bone region obtained as being equal to or less than the specific threshold value.
By setting, it is possible to selectively emphasize a bone part important in diagnosis to make it easy to see.

The bone radiation image signal subjected to the gradation processing according to the present invention is immediately transmitted to the printer 17 as described above.
May be hard copied, but CR
The information may be reproduced on the T or stored once in the filing system, and read out when necessary to be hard-copied or displayed on the CRT. When storing the bone radiation image in the filing system, the processed radiation image signal subjected to the gradation processing according to the present invention may be stored. May be stored as a pair with the gradation processing condition (gradation conversion table) determined based on the above, and the gradation processing may be performed at the time of reading.

Further, in this embodiment, the bone radiation image signal read photoelectrically from the stimulable phosphor layer is subjected to gradation processing, but is limited to image reading using the stimulable phosphor. Instead, a configuration using another two-dimensional radiation detector may be used. For example, a silver halide film on which a bone radiation image is recorded is irradiated with light from a light source such as a laser or a fluorescent lamp to obtain transmitted light of the silver halide film, and the transmitted light is photoelectrically converted to a bone radiation image. A configuration for obtaining a signal may be used.

In the above embodiment, the leg M of the human body is described as an example of a subject including bones. However, the subject may be a hand or the like, and the subject is not limited.

[0061]

As described above, according to the present invention, since a bone region can be accurately extracted from a radiation image signal of a subject including a bone, imaging is performed based on the extracted image data in the bone region. It is possible to optimally set the conditions of image processing such as gradation processing without being affected by the body position, site, and even body shape, and to reproduce the bone portion, which is the region of interest, stably and easily. effective.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a basic configuration of the present invention.

FIG. 2 is an overall system schematic diagram showing an embodiment of the present invention.

FIG. 3 is a block diagram illustrating a configuration of an image processing unit according to the embodiment.

FIG. 4 is a block diagram illustrating a configuration example of a printer.

FIG. 5 is a diagram showing characteristics of vertical and horizontal profiles in an image of a leg.

FIG. 6 is a diagram showing how threshold initial values are set based on a profile crossing a bone.

FIG. 7 is a diagram showing a profile in a vertical direction not including a bone.

FIG. 8 is a diagram showing how threshold initial values are set using a histogram.

FIG. 9 is a state change diagram showing how a specific threshold is detected by increasing the threshold.

FIG. 10 is a state change diagram showing another embodiment of specifying a detection area for monitoring connectivity.

FIG. 11 is a state change diagram showing how a specific threshold is detected by decreasing the threshold.

FIG. 12 is a diagram illustrating a state of a histogram of a bone region.

FIG. 13 is a diagram illustrating setting of gradation conversion characteristics using a histogram of a bone region.

FIG. 14 is a diagram showing a state of setting a signal range of a bone using a conventional histogram.

[Explanation of symbols]

 21 Binarization section 22 Threshold setting section 23 Detection area setting section 24 Detection area connectivity determination section 25 Gradation processing condition determination section 26 Gradation processing section

Claims (3)

(57) [Claims] An apparatus for detecting a region of interest in a bone radiation image which detects a bone region as a region of interest from image data of a bone radiation image formed corresponding to the amount of radiation transmitted through a subject including bone. A binarizing unit that binarizes the image data by comparing the image data with a threshold value and divides the bone radiation image into a plurality of image regions;
A threshold setting unit for gradually increasing or decreasing the threshold from a predetermined initial value; and when increasing and changing the threshold by the threshold setting unit, an image region composed of image data smaller than the threshold is used as a detection region. When decreasing the threshold value, an image area composed of image data larger than a threshold value is set as a detection area. When the threshold value is close to the predetermined initial value, the detection areas divided into a plurality of pieces have a predetermined proximity relationship with the threshold value change. A specific threshold value detecting means for detecting a threshold value when the threshold value has become a specific threshold value, and image data corresponding to the smaller radiation transmission amount when binarizing the image data with the specific threshold value detected by the specific threshold value detecting means And a bone region detecting means for detecting an image region composed of a bone region as a bone region. 2. The apparatus according to claim 1, wherein the predetermined proximity relation in said specific threshold value detecting means is when the divided detection areas shift from a non-connection state to a connection state in accordance with a threshold value change. A region-of-interest detection apparatus for a bone radiation image according to the above description. 3. A process for determining an image processing condition based on image data included in a bone region as a region of interest detected by the region-of-interest detection apparatus for bone radiation images according to claim 1 or 2. A bone radiation comprising: a condition determining means; and an image processing means for performing image processing on image data of a bone radiation image based on the image processing condition determined by the processing condition determining means. Image processing device for images.