JPWO2021140309A5 - - Google Patents

Description

The present invention relates generally to apparatus and methods for producing tomograms, and finds particular, but not limited, utility in pathology and/or non-destructive evaluation.

It is often required to perform high resolution scans of relatively large areas, but as resolution increases and area size increases, this becomes cumbersome. After reconstruction is performed, high-resolution scans are not required over the entire region, but lower resolution may be used in specific sub-regions, which takes some of the burden off both imaging and reconstruction. It may become clear that the

The present invention seeks to alleviate this burden by dynamically determining during the imaging process where high-resolution and low-resolution scans are appropriate. In this way, large structures such as aircraft wings can be scanned quickly and easily. In particular, the system can scan large areas at low resolution while automatically determining sub-areas where higher resolution is required based on the low resolution scan.

According to a first aspect of the present invention, there is provided a method for generating a tomogram, the method comprising: acquiring a plurality of first X-ray attenuation images of a subject; the first X-ray attenuation image is suitable for reconstructing a first density function indicative of the attenuation of X-ray radiation, said first density function having a first resolution; identifying from a first X-ray attenuation image of at least one first region whose entropy and/or gradient exceed a predefined threshold entropy and/or gradient, respectively; acquiring at least one second X-ray attenuation image of one first region; reconstructing a second density function indicative of attenuation, the second density function having a second resolution higher than the first resolution. Ru.

The approach uses an initial scan on a predetermined grid to obtain a minimal set of images and then identifies regions of interest for further scans. Further scan positions are determined by the level of information (e.g. image entropy or gradient) found in previous iterations, and such regions are identified by edges (or edges/edges), cracks (or , cracks) or complex structures. After each iteration, the level of information (eg, image entropy or gradient) decreases relative to the pixel/voxel size. A voxel is defined herein as a 3D pixel, ie a size that determines the resolution of the reconstructed density function.

In this way a more efficient method for scanning is achieved. This means that the number of scan points (i.e. images) can be increased, which means that lower power equipment with shorter standoff distances can be used, which makes it more compact and portable. brings a unit. For example, a typical source produces a cone of x-rays with an opening angle of 30 to 60 degrees. Moving such a source to a stand off distance of several times the thickness of the object allows complete coverage for 2D images, but reduces the X-ray power by several times due to the inverse square law. It requires a much larger detector area or multiple exposures from one location, at the cost of doubling the need. The volume and weight of the entire device will increase compared to moving the source to a small multiple of the object thickness.

The predefined threshold may be adjustable by an operator or may be automatically selected by a processor based on one or more physical constraints. Within the range of possible predefined thresholds, lower thresholds may be chosen that still exceed those typically measured for homogeneous materials, taking into account the background noise levels typical of the setting in which the imaging is performed. (i.e., a low threshold may be chosen to look for subtle defects in an otherwise uniform material). Conversely, a high threshold may be selected such that fractures or sharp edges reach such settings (ie, a high threshold may be selected to simply identify fractures). A lower threshold may be used to determine areas that require further inspection, and a higher threshold may be used to determine when an area has reached sufficient detail that small features can be detected. . The intermediate threshold may be selected by the user depending on the acquisition and/or subject situation.

The method comprises determining from the plurality of first X-ray attenuation images and the at least one second X-ray attenuation image at least one second and obtaining at least one third X-ray attenuation image of the at least one second region, the plurality of X-ray attenuation images, the at least one second X-ray reconstructing a third density function indicative of x-ray radiation attenuation from an attenuation image and the at least one third x-ray attenuation image, the third density function being higher than the second resolution; having a third resolution.

identifying at least one additional (e.g., third, fourth, etc.) region; acquiring at least one additional (e.g., fourth, fifth, etc.) x-ray attenuation image; and the further density function. It is contemplated that further sequences involving reconfiguring (eg, fourth, fifth, etc.) may apply as well. The approach disclosed herein is based iteratively on the acquired image data and thus on the features of the object.

This iterative process may continue until at least one additional region whose entropy and/or gradient exceeds a predefined threshold entropy and/or gradient, respectively, cannot be identified. Alternatively or additionally, this iterative process is repeated until no further improvement in the level of information (e.g. image entropy or gradient) is seen, i.e. with respect to pixel/voxel size, there is no further reduction or at least This can continue until the decrease is below a certain threshold level. The threshold level may be a percentage and/or absolute change to a defined threshold entropy and/or slope.

These boundaries (or limits) define the end of the iterative process and may be required to apply to all voxels.

acquiring a plurality of first x-ray attenuation images of the object determines a minimum number of first x-ray attenuation images needed to reconstruct a first density function indicative of attenuation of x-ray radiation; may include. That is, the acquiring step may include determining the first set of emitter positions required to completely illuminate the subject with an x-ray cone originating from each emitter position. Completely irradiating means at least one beam of X-ray radiation passing through each part of the object, and/or at least two, three or more beams of X-ray radiation passing through each part of the object. be able to.

Reconstructing the first density function indicative of the attenuation of the X-ray radiation may include generating a 2D image of the object, or may include generating a 3D tomogram. The 3D tomogram thus generated may include a predetermined resolution such as 10 microns, 20 microns, 50 microns, 100 microns, 200 microns, 500 microns, 1000 microns, etc. The selection of such resolution may be performed by an operator or may be automatically selected based on the requirements of the resulting tomogram and/or the object under consideration.

Similarly, the 3D tomogram thus generated may include predetermined slice thicknesses such as 10 microns, 20 microns, 50 microns, 100 microns, 200 microns, 500 microns, 1000 microns, etc. Such slice thickness selection may be performed by an operator or may be automatically selected based on the resulting tomogram and/or the requirements of the subject under consideration. The slice thickness may be determined based on the desired resolution (e.g., the slice thickness/resolution may be selected to match the resolution within the slice), or may be independent of the desired resolution. You can. In this way, the term resolution can mean the planar resolution of a slice (e.g., in the XY plane, e.g., one or both of the X and Y axes) and can be related to pixel size; and/or slice resolution (eg, resolution in the Z direction perpendicular to the planar resolution). In particular, the term resolution may relate to voxel size.

Reconstructing a subsequent (e.g., second, third, fourth, fifth, etc.) density function from a preceding (e.g., first, second, third, fourth, etc.) density function The method may include generating a further 3D tomogram having the same or smaller slice thickness as the generated 3D tomogram.

The object may be a person, a body part, a specimen, a manufactured part, or any other item that is being examined (or interrogated). The object may be a part thereof, ie the object may be a region of interest within the above-mentioned article. The region of interest can be any shape (e.g., cubic, cylindrical, or any other geometric shape defined by the user) and includes the total area from which X-rays can be emitted and/or detected. It can be only a part.

Although there may be a single movable X-ray emitter, in preferred embodiments there may be multiple X-ray emitters arranged, for example, in an array (eg, forming a flat panel). The array may be a triangular/hexagonal grid, a square/rectangular grid, and/or any other desired grid, but a single emitter may be moved to replicate the array of such grids. It is possible that The multiple x-ray emitters may be movable or non-moveable relative to the subject. The plurality of x-ray emitters may be movable or non-movable relative to the detector (eg, a detector panel).

identifying from said plurality of first X-ray attenuation images at least one first region whose entropy and/or slope exceeds a predefined threshold entropy and/or slope, respectively, indicative of attenuation of X-ray radiation; reconstructing a first density function; and determining, for each point of a plurality of points in the first density function, the entropy and/or gradient in a region surrounding the point. can. This identifying step may be performed automatically by the processor.

The points can be selected and/or predetermined to correspond to a grid/array that can be applied to each slice of the first density function. Alternatively or additionally, the point may correspond to a pixel/voxel in the first density function.

For example, identifying from a plurality of first X-ray attenuation images at least one first region whose entropy and/or slope exceed a predefined threshold entropy and/or slope, respectively, reconstructing a first density function representing the first density function, the first density function comprising a plurality of pixels/voxels having values corresponding to the determined x-ray attenuation; for each pixel/voxel of the plurality of pixels/voxels within the density function of the plurality of pixels/voxels, determining the entropy and/or gradient within the region surrounding that pixel/voxel.

The area surrounding that point/pixel/voxel is determined by its absolute distance from that pixel/voxel and/or by the center (e.g. radius) of that point/pixel/voxel, and/or by its relative distance from that point/pixel/voxel and/or by its relative distance from that point/pixel/voxel. Or it may be determined based on the center (eg radius) of that point/pixel/voxel. In this context, absolute distance can mean the distance measured relative to the object (e.g. distance in microns) and/or relative distance may mean the distance measured relative to the pixel/voxel (e.g. pixel /voxels and/or slices, for example a number of slices of at least 1, 2, 5, 10, 20, 50 or more pixels/voxels).

Such calculation of entropy (eg, by calculating the total variation within a region) may include using a weighting term based on distance from the scan location under examination.

The plurality of pixels/voxels within the first density function may form all pixels/voxels within the first density function, or only a subset of all pixels/voxels within the first density function. , so that the determining step is performed only on that subset. Thereby, processing can be reduced.

The same principles described above with respect to the first density function can be applied to subsequent (eg, second, third, fourth, etc.) density functions. For example, the second density function can include a second plurality of pixels/voxels having values corresponding to the determined x-ray attenuation, and the plurality of first x-ray attenuation images and the at least one identifying at least one second region from two second X-ray attenuation images whose entropy and/or slope exceed a predefined threshold entropy and/or slope, respectively For each pixel/voxel of the pixels/voxel, the method may include determining the entropy and/or gradient within the region surrounding that pixel/voxel.

The step of identifying from said plurality of first X-ray attenuation images at least one first region whose entropy and/or slope exceeds a predefined threshold entropy and/or slope, respectively, comprises back-projecting for each potential location from the detector plane through the at least one first region; and for each voxel in the at least one first region, establishing whether to acquire the at least one second x-ray attenuation image from the potential location, the at least one second x-ray image of the at least one first region; Acquiring attenuation images includes acquiring attenuation images corresponding to those established potential locations.

establishing a geometric factor that maximizes angular coverage of said voxel; an overlap factor that ensures that said voxel is imaged at least once; and said at least one an entropy factor that minimizes entropy within one first region; and selecting potential locations based on the values of the function. be able to.

Geometric factors tend to drive the scan points as far apart as possible (ie, weighted towards larger angles, but less than the maximum cone angle of the source). The overlap factor tends to maximize the overlap, which tends to drive the scan points as close as possible. The overlap factor may be considered a 'no blind spot' condition. The weightings can be determined by measurements or modeling and will vary depending on the hardware under consideration.

Subsequent scans can be performed in the direction of the highest entropy gradient or at the nearest grid point.

The method may further include determining a substantially optimal path between each of the at least one second X-ray attenuation image to be acquired.

Once a complete set of source location "requests" has been generated, a near-optimal route can be calculated. In general, an optimal path between n points without repeated visits is not solvable in closed form and is NP-hard (thus requiring large computational costs for n>>>1). Techniques from the well-studied "Travelling Salesman Problem" (TSP) are typically used to find near-optimal paths through a set of coordinates.

The source may be moved to each point along the optimal path and an exposure may be made and recorded on the detector at each point. This movement may be real, ie the source may be physically moved between points along the path. Alternatively or additionally, the movement may be virtual, ie a series of X-ray emitters may be activated within an emitter array and the path tracked. In some embodiments, a combination or real and virtual movement can be used, i.e., the emitter array can be physically moved and individual X-ray emitters can be actuated to track the path. . Multiple exposures may be required for a given location if the detector is small compared to X-ray cone projection, in which case the detector is , must be moved to some position. In such cases, the area and position of the detector is treated as a subset of the total required detection area, and the detectors are repeatedly placed to cover the total required detection area.

A predetermined offset (e.g., half the detector pixel size, although other offsets are envisaged) in one or more principal directions for ultra-high resolution tomograms, plus or minus all established positions. A repeated scan may be performed. A final reconstruction may be performed on the final composite data set.

It can be appreciated that it is not necessary to completely scan the object before performing subsequent steps; rather, the identifying step may be performed immediately after the initial acquisition.

According to a second aspect of the invention, the apparatus for generating a tomogram comprises an X-ray emitter, an X-ray detection panel and a processor for carrying out the method according to any of the preceding claims. Equipment is provided.

The device may include an X-ray emitter panel consisting of a plurality of X-ray emitters, of which the described X-ray emitter is one such example. Alternatively, the described X-ray emitter may be the only such X-ray emitter in the device.

The X-ray emitter (and/or the X-ray emitter panel) may be movable. In particular, the X-ray emitter (and/or the X-ray emitter panel) can be mounted on an armature for movement, eg automatic movement by a processor.

These and other features, features and advantages of the invention will become apparent from the following detailed description, taken in conjunction with the accompanying drawings, which illustrate by way of example the principles of the invention. This description is given for illustrative purposes only, without limiting the scope of the invention. The reference figures cited below refer to the attached drawings.

FIG. 1 shows an array of X-ray sources with overlapping regions of interest shown.

FIG. 2 shows the same view as FIG. 1, with an indication of subsequent image capture positions marked for a first of the regions of interest.

FIG. 3 shows the same view as FIG. 2, with approximately optimal paths marked between subsequent image capture locations.

Although the invention will be described with respect to particular drawings, the invention is not limited thereto but only by the scope of the claims. The drawings described are only schematic and are non-limiting. Each drawing does not contain all features of the invention, and therefore should not necessarily be considered an embodiment of the invention. In the drawings, the size of some of the elements may be exaggerated and not drawn on scale for illustrative purposes. The dimensions and relative dimensions do not correspond to any practical reduction to the implementation of the invention.

Furthermore, the terms first, second, third, etc. in the description and claims are used to distinguish between similar elements, and not necessarily temporally, spatially, in a ranking, or in any other way. It is not used to describe a sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances and that the operations may occur in other orders than those described or illustrated herein. Similarly, method steps described or claimed in a particular order may be understood to operate in a different order.

Furthermore, the terms top, bottom, above, below and the like in the specification and claims are used for purposes of description and not necessarily for describing relative positions. It is to be understood that the terms so used are interchangeable under appropriate circumstances and operations may be otherwise capable of orientations than those described or illustrated herein.

It is to be noticed that the word "comprising", used in the claims, should not be interpreted as being restricted to the means listed thereafter, but does not exclude other elements or steps. . Therefore, although the presence of the described feature, integer, step or component as mentioned should be interpreted as specifying the presence of one or more other features, integers, steps or components, or This does not preclude the existence or addition of such groups. Therefore, the scope of the expression "device comprising means A and B" should not be limited to a device consisting only of the components A and B; for the purposes of the present invention, the relevant components of the device are only A and B. It means something.

Throughout this specification, reference to an "embodiment" or "aspect" means that a particular feature, configuration, or characteristic described in connection with an embodiment or aspect is included in at least one embodiment or aspect of the invention. means to be Thus, the appearances of the phrases "in one embodiment," "in an embodiment," or "in an aspect" in various places throughout this specification are not necessarily all referring to the same embodiment or aspect. Instead, reference may be made to different embodiments or aspects. Moreover, the particular features, structures, or characteristics of any one embodiment or aspect of the present invention may be apparent to those skilled in the art from this disclosure. may be combined with any other specific features, structures, or characteristics of the embodiments or aspects in any suitable manner.

Similarly, in the description, various features of the invention may be presented in a single embodiment, figure, or description thereof to streamline the disclosure and aid in understanding one or more of the various aspects of the invention. It should be understood that they may be grouped together. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Moreover, any individual drawing or description of an aspect is not necessarily to be considered an embodiment of the invention. Rather, as the following claims reflect, aspects of the invention lie in less than all features of a single disclosed embodiment described above. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.

Furthermore, while some embodiments described herein include some features included in other embodiments, combinations of features of different embodiments are well understood by those skilled in the art. and are meant to form further embodiments within the scope. For example, in the following claims, any of the claimed embodiments may be used in any combination.

In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In other instances, well-known methods, structures, and techniques have not been shown in detail in order not to obscure the understanding of this description.

Unless stated to the contrary in the discussion of the present invention, disclosure of alternative values for the upper or lower limits of a permissible range for a parameter, coupled with an indication that one of said values is highly preferred over the other, does not include said alternative value. Each intermediate value of the parameter between the more preferred value and the less preferred value is itself more preferable than the less preferred value, and each intermediate value between the less preferred value and the intermediate value should be interpreted as an implicit statement of preference over value.

Use of the term "at least one" may mean only one in certain circumstances. Use of the term "any" may mean "all" and/or "each" in certain circumstances.

The principles of the invention are now explained by a detailed description of at least one drawing that includes illustrative features. It is clear that other configurations may be constructed according to the knowledge of those skilled in the art without departing from the underlying concepts or technical teachings, and the invention is limited only by the terms of the appended claims.

FIG. 1 shows part of a triangular/hexagonal array in which the X-ray sources are located at the intersections of grid lines 10. FIG. Dashed circles 20 indicate where regions of interest have been identified where higher resolution images are desired.

FIG. 2 shows the same view (or view/field of view) as FIG. 1, this time with grid lines 10 shown in dashed lines for clarity. A dashed circle 20 remains to indicate where the region of interest has been identified. A first region of interest 30 is shown in solid lines. Shown within and adjacent to the first region of interest 30 are subsequent image capture locations 40 marked with respective crosses.

Each of the identified image capture locations 40 is selected to establish a high resolution reconstruction of the region of interest. The image capture positions 40 optimize a weighted equation whose positions correspond to the entropy of the image around each voxel in the reconstruction, the angular coverage of the voxel, and the degree of overlap of the x-ray cones at the voxel. It is established on the basis that it is not uniformly distributed over the region of interest.

FIG. 3 shows the same view as FIG. 2, with an approximately optimal path 50 marked between subsequent image capture locations.

Although in this example the near-optimal path passes through all image capture locations 40 within the first region of interest 60 before moving to the second region of interest 60, in the preferred embodiment the near-optimal path 50 passes through all image capture locations 40 in the first region of interest. It is determined by considering all image capture positions across the region of interest 20. Therefore, a near-optimal path can be traveled between regions of interest and back to previous regions of interest to collect additional image capture locations.

In this example, a substantially optimal path 50 has been selected as an illustrative example only, such that the first region of interest 30 passes through all image capture locations 40 before moving to the second region of interest 60. There is. However, it should be understood that this is not the case if a different region of interest 20 is selected for the illustrative example.

Claims (7)

A method of generating a tomogram, the method comprising:
acquiring a plurality of first X-ray attenuation images of the object, the plurality of first X-ray attenuation images being suitable for reconstructing a first density function indicative of attenuation of X-ray radiation; and the first density function has a first resolution;
identifying from the plurality of first X-ray attenuation images at least one first region whose entropy and/or slope exceed a predefined threshold entropy and/or slope, respectively;
acquiring at least one second X-ray attenuation image of the at least one first region;
reconstructing a second density function indicative of X-ray radiation attenuation from the plurality of first X-ray attenuation images and the at least one second X-ray attenuation image, the second density function has a second resolution higher than the first resolution;
method including. from the plurality of first X-ray attenuation images and the at least one second X-ray attenuation image, at least one second region whose entropy and/or slope exceed a predefined threshold entropy and/or slope, respectively; a step of identifying;
acquiring at least one third X-ray attenuation image of the at least one second region;
reconstructing a third density function indicative of attenuation of X-ray radiation from the plurality of X-ray attenuation images, the at least one second X-ray attenuation image and the at least one third X-ray attenuation image; the third density function has a third resolution higher than the second resolution;
The method of generating a tomogram according to claim 1, further comprising: identifying from the plurality of first X-ray attenuation images at least one first region whose entropy and/or slope exceed a predefined threshold entropy and/or slope, respectively;
reconstructing a first density function indicative of the attenuation of the X-ray radiation;
for each point of the plurality of points in the first density function, determining the entropy and/or gradient in the region surrounding the point;
The method for generating a tomogram according to claim 1 or 2, comprising: identifying from the plurality of first X-ray attenuation images at least one first region whose entropy and/or slope exceed a predefined threshold entropy and/or slope, respectively;
For each potential position of the x-ray source,
backprojecting from the detector plane through the at least one first region to each potential location;
establishing for each voxel in the at least one first region whether to obtain the at least one second X-ray attenuation image from its potential location;
including;
Claims 1-1, wherein the step of obtaining the at least one second X-ray attenuation image of the at least one first region comprises obtaining attenuation images corresponding to their established potential locations. 3. The method for generating a tomogram according to any one of 3. The step of establishing includes:
a geometric factor that maximizes angular coverage of the voxel;
an overlap factor ensuring that the voxel is imaged at least once;
an entropy factor that minimizes entropy within the at least one first region;
optimizing a function including weighted constraints including;
selecting potential locations based on the value of the function;
The method of generating a tomogram according to claim 4, comprising: A method for generating a tomogram according to any of claims 1 to 5, further comprising the step of determining a substantially optimal path between each of the at least one second X-ray attenuation image acquired. an X-ray emitter;
an X-ray detection panel;
A processor for carrying out the method according to any one of claims 1 to 6;
An apparatus for generating tomograms, including: