Detection of geometric calibration markers in a spectral tomosynthesis system

Abstract

A method for geometric calibration of a volumetric imaging device includes placing a calibration phantom in an irradiation path that includes an object positioned between an x-ray source and a detector. The phantom has a plurality of radiopaque markers formed of a marker material. In a repeated sequence, 2D projection image data is acquired of the object and the phantom at each of a plurality of positional relationships between the x-ray source and the detector, where the 2D projection image data distinguish at least first and second x-ray energy distributions, and source and detector geometries of the imaging device are calculated corresponding to the 2D projection image data acquired for the first and second x-ray energy distributions. The method reconstructs and displays a 3D volumetric image of the object according to the anatomical image data acquired from the object and the source and detector geometries in the 2D projection images.

Description

[Technical field]

[0001] The present disclosure relates generally to 3D volumetric radiological imaging, and more particularly to a method and apparatus for automated detection of calibration markers used in tomosynthesis imaging. [Background technology]

[0002] Three-dimensional (3D) or volume x-ray images can be of significant value for the diagnosis and treatment of teeth and supporting structures. Volume x-ray images for this purpose are formed by combining image data from two or more individual 2D projection images, acquired within a short time of each other and with well-defined angular and positional geometries between each projection image and the teeth of interest, and between each projection image and another projection image.

[0003] Cone beam computed tomography (CBCT) is one established method for obtaining volumetric images of dental structures from multiple projection images. In CBCT imaging, the image detector and radiation source orbit the object, acquiring a series of x-ray projection images at small angular increments. The information obtained is then used to synthesize a volumetric image that faithfully represents the imaged object within the available resolution of the system, so that the volumetric image formed may be viewed from any angle. Commercially available CBCT devices for dentistry include the CS81003D system from Carestream Dental LLC, Atlanta, Georgia.

[0004] Although CBCT imaging is a powerful diagnostic tool, there may be cases where full-scale CBCT imaging capabilities are not required, even if volume imaging is beneficial. This is acknowledged, for example, in the disclosures of U.S. Pat. No. 5,399,323 and U.S. Pat. No. 5,523,633. For some types of volume imaging, such as for use in guiding implant placement, basic volume imaging capabilities may be useful, for example. Volume imaging may also be useful in avoiding overlay anomalies between adjacent dental structures. In such applications, a large number of x-ray projection images, as provided by a CBCT system, may not be required. Instead, sufficient volume information may be obtained using a small number of x-ray images, provided that spatial coordinate references between the images are maintained.

[0005] As a general rule, it would be advantageous to obtain the minimum number of x-ray exposures necessary to produce volumetric diagnostic data. A full CBCT series of projection images acquired in a 180° trajectory would require a higher cumulative radiation dose than would a partial series, which would use fewer projection images taken at a narrower range of angles or at increased relative angular increments. Thus, the methods taught in the Kalke and Siltanen et al. disclosures can help reduce patient exposure when full CBCT imaging is not required.

[0006] Tomosynthesis, sometimes referred to as "2.5D imaging," appears to offer the dentist many advantages over conventional 2D radiographic and 3D tomographic imaging of intraoral features, such as CBCT imaging. In tomosynthesis, as with other volumetric imaging approaches, a limited number of 2D projection images are acquired in succession, with each image frame shifted at a relative angle from the previously acquired image frame. Reconstruction techniques may then be used to form a volumetric image of sufficient depth and resolution for many diagnostic and evaluation functions. Tomosynthesis thereby offers some of the advantages of full-scale tomographic imaging to provide volumetric data, but at a lower dose than tomography requires.

[0007] Tomosynthesis imaging uses incremental geometric changes in the relative radiation source angle to the detector surface with each image. Conventional tomosynthesis systems, such as those used for mammography, have a mechanical coupling that provides inherent control of the source position relative to the detector. This mechanical coupling essentially provides geometric calibration with alignment applied from one imaging exam to the next.

[0008] A fixed mechanical coupling of the detector to the source is not always possible: in some types of tomosynthesis imaging, the detector and source are separated, requiring some method to provide geometric alignment and calibration.

[0009] It should be noted that alignment and geometric calibration are related but not identical. Alignment relates to beam direction and occurs prior to imaging. Geometric calibration may be provided following imaging, for example by processing the individual acquired images as described herein.

[0010] By properly aligning the source to the detector, the x-ray energy is positioned such that the x-ray beam is incident only within the boundaries of the detector. Once alignment is achieved, correct geometric calibration data derived from the image content then defines the relative spatial positions of the x-ray source and detector for each successive projection image, so that reconstruction techniques can accurately create volumetric 3D image data from the acquired 2D projection images.

[0011] For most dental applications, intraoral detectors are deployed behind the patient's teeth relative to the x-ray source, are largely hidden from view, and need to be flexibly positionable at various locations within the mouth, hindering attempts at uncomplicated source-to-detector alignment and complicating the task of geometric calibration. Not only must the system be able to reliably identify the boundary geometry of the detector contours with proper alignment, but the relative position of the x-ray source with respect to the detector needs to be known using geometric calibration in order to enable accurate 3D volume reconstruction from 2D projection images.

[0012] The use of radiopaque markers placed in the path of the x-ray beam provides a viable solution for geometric calibration, but presents significant drawbacks. The radiopacity of the markers obscures portions of the underlying anatomical structures, which can impair the quality of acquired images for diagnostic purposes. Attempting to alleviate this problem by reducing the size or density of the markers can make it difficult to distinguish the markers from the anatomical structures, complicating or compromising accurate geometric calibration.

[0013] Although many solutions have been proposed to provide intraoral tomosynthesis, there is considerable room for improvement to make tomosynthesis technology suitable for dentists.

[0014] Therefore, it can be seen that there is a need for improved apparatus and methods that provide accurate geometric calibration for intraoral tomosynthesis systems in which the detector and radiation source are not mechanically coupled. [Prior art documents] [Patent documents]

[0015] [Patent Document 1] US Patent Application Publication No. 2007 / 0127801 [Patent Document 2] U.S. Patent No. 7,269,241 [Patent Document 3] US Patent Application Publication No. 2009 / 0060145 Summary of the Invention [Means for solving the problem]

[0016] Broadly described and in accordance with exemplary embodiments described herein, the present invention includes an apparatus and method for creating a volumetric image from a small number of radiographic images acquired by an intraoral imaging detector. In particular, the present invention includes an apparatus and method for geometric calibration of a volumetric imaging device in which the detector and source are not mechanically coupled. Advantageously, the present invention addresses the problems posed by the tradeoff between accurate marker detection and impact on image quality. Notwithstanding anything otherwise provided herein, the present invention is defined by the appended claims.

[0017] According to one aspect of the present disclosure, a method is provided for geometric calibration of a volumetric imaging device, the method comprising: (a) placing a calibration phantom in an irradiation path including an object disposed between an x-ray source and a detector, the phantom having a plurality of radiopaque markers formed of a marker material; (b) In a repeating sequence, at each of several relative positions of the x-ray source and the detector, (i) acquiring 2D projection image data for the subject and the phantom, where the 2D projection image data distinguishes between at least first and second x-ray energy distributions; (ii) calculating source-to-detector geometry of an imaging device corresponding to the acquired 2D projection image data for the first and second x-ray energy distributions; (c) reconstructing and displaying a 3D volumetric image of the object according to the anatomical image data acquired from the object and the calculated source and detector geometries within the plurality of 2D projection images.

[0018] The present invention and its exemplary embodiments described herein address, in whole or at least in part, the problems, deficiencies, and shortcomings of the related art. In addition, the present invention and such exemplary embodiments provide various advantages and benefits, which are described herein. Other advantages and benefits inherently achieved by the disclosed apparatus and methods may occur or become apparent to those skilled in the art. [Brief description of the drawings]

[0019] The foregoing, as well as other objects, features and advantages of the present invention will become apparent from the following more particular description of illustrative embodiments thereof, as illustrated in the accompanying drawings, in which elements are not necessarily to scale relative to each other. [Figure 1] 1 is a schematic diagram illustrating components of a chairside tomosynthesis imaging device according to an exemplary embodiment according to the present application. [Figure 2A] FIG. 2 is a schematic diagram showing a photon counting sequence. [Figure 2B] FIG. 1 is a schematic diagram showing a photon counting arrangement for measurements at two different energy levels. [Figure 3A] 1A-1D are simplified schematic block diagrams illustrating different aspects of the alignment problem; [Figure 3B] 1A-1D are simplified schematic block diagrams illustrating different aspects of the alignment problem; [Figure 4] FIG. 1 is a schematic block diagram showing an imaging device for calculating the lateral position and angular orientation of the intraoral image detector. [Diagram 5] FIG. 1 is a schematic block diagram showing an imaging device that calculates the lateral position and angular orientation of the intraoral image detector and projects a representation onto the patient's cheek. [Figure 6A]FIG. 2 is a schematic diagram illustrating how triangulation is used for position detection in one exemplary embodiment of the present invention. [Figure 6B] FIG. 11 is a schematic diagram illustrating position detection in an alternative exemplary embodiment of the present invention. [Figure 6C] FIG. 13 shows a holder having an arrangement of radiopaque alignment markers around a central opening. [Figure 6D] FIG. 13 illustrates how markers appear in the acquired images along the boundaries of the imaged intra-oral features. [Figure 6E] FIG. 13 illustrates how markers appear in the acquired images along the boundaries of the imaged intra-oral features. [Figure 7A] 1 is a perspective view of an intraoral x-ray imaging device in an incorrect alignment, according to one exemplary embodiment. [Figure 7B] FIG. 1 is a perspective view of an intraoral x-ray imaging device with correct alignment, according to one exemplary embodiment. [Figure 8A] FIG. 13 is a bottom view showing a radiation source assembly with collimation in a substantially square arrangement. [Figure 8B] 13A-13C are bottom views showing collimator plate assemblies with apertures of different aspect ratios. [Figure 8C] FIG. 11 is a bottom view showing an alternative collimator plate arrangement that is translated in a linear direction to place different sets of apertures over different subsets of the radiation source array. [Figure 8D] FIG. 11 is a bottom view showing an alternative collimator plate arrangement that is translated in a linear direction to place different sets of apertures over different subsets of the radiation source array. [Figure 8E] FIG. 13 is a bottom view showing a collimator plate having a paired set of apertures of different aspect ratios. [Figure 8F] FIG. 1 is a diagram of a collimator that is linearly curved and translated along a linearly distributed source array. [Figure 8G]FIG. 13 is a diagram of a collimator that is curved and translated by rotating around a linearly distributed source array. [Figure 9A] FIG. 1 is a schematic block diagram illustrating an imaging pattern used to acquire a volumetric image from a limited number of x-rays. [Figure 9B] FIG. 13 is a perspective view showing how different positions of an x-ray emitter relative to a patient provide individual images for use in forming a volumetric image. [Figure 10A] FIG. 1 is a schematic diagram illustrating an intra-oral imaging device for tomosynthesis imaging with an x-ray source transported along an arcuate trajectory. [Figure 10B] FIG. 1 is a schematic diagram illustrating an intra-oral imaging device for tomosynthesis imaging using an x-ray source transported along a linear track. [Figure 10C] FIG. 1 is a top schematic diagram showing a frame having multiple articulated sections at each adjustable joint for reporting sensed extension and rotation data. [Figure 11A] FIG. 10B is a perspective view showing the radial path arrangement x-ray source configuration shown in FIG. 10A. [Figure 11B] FIG. 10C is a perspective view showing the linear path arrangement x-ray source configuration shown in FIG. 10B. [Figure 12A] FIG. 13 is a schematic diagram illustrating the use of a marker guide in combination with an intra-oral detector. [Figure 12B] FIG. 13 is a schematic diagram illustrating the use of a marker guide that is not directly coupled to an intraoral detector. [Figure 13A] FIG. 1 shows a treatment system including a dental procedure chair and other devices. [Figure 13B] FIG. 1 is a top schematic diagram showing an arrangement with an x-ray source coupled to a collimator, chair, floor, or other fixture. [Figure 14A] 1 illustrates an assembly and components of a marker guide for alignment according to an exemplary embodiment of the present application. [Figure 14B] 1 illustrates an assembly and components of a marker guide for alignment according to an exemplary embodiment of the present application. [Figure 15]FIG. 13 illustrates the position of the frame relative to the patient's face for alignment and collimation support. [Figure 16] 1A and 1B show side and top views of a frame and associated components for collimation and alignment. [Figure 17] 1 is a schematic diagram illustrating an exemplary relative movement path that may be tracked for an imaged object for tomosynthesis imaging. [Figure 18] FIG. 1 is a schematic block diagram illustrating a Spindt-type field emitter-based source or other x-ray source array for image acquisition. [Figure 19] FIG. 11 is a schematic block diagram illustrating an imaging pattern used to acquire a volumetric image from a limited number of x-rays in an alternative exemplary embodiment. [Figure 20] FIG. 2 is a logic flow diagram showing a sequence for capturing x-ray images to create a volumetric image. [Figure 21] FIG. 2 is a block diagram illustrating spatial location and angular orientation information associated with image data. [Figure 22] FIG. 2 is a logic flow diagram showing system activity in preparation for each image capture in a sequence. [Figure 23] FIG. 13 shows a comparative display of projection images and tomosynthesis image slices. [Figure 24] FIG. 13 is a schematic diagram showing an arrangement with markers fixed in position relative to a detector. [Diagram 25] FIG. 13 is a schematic diagram showing an alternative arrangement in which the phantom markers are fixed to the source. [Figure 26A] 1 is a graph illustrating x-ray energy and mass attenuation characteristics that are combined to define a given x-ray energy distribution according to an exemplary embodiment of the present disclosure. [Figure 26B] 1 is a graph illustrating x-ray energy and mass attenuation characteristics that are combined to define a given x-ray energy distribution according to an exemplary embodiment of the present disclosure. [Figure 26C]1 is a graph illustrating x-ray energy and mass attenuation characteristics that are combined to define a given x-ray energy distribution according to an exemplary embodiment of the present disclosure. [Figure 27] FIG. 1 is a logic flow diagram illustrating a sequence for providing marker and anatomical 2D images from a single 2D projection image obtained using a spectral x-ray imaging system. [Figure 28A] 1A-1C illustrate simulated results of image processing for marker detection in 2D projection images for tomosynthesis imaging, according to an exemplary embodiment of the present disclosure. [Figure 28B] 1A-1C illustrate simulated results of image processing for marker detection in 2D projection images for tomosynthesis imaging, according to an exemplary embodiment of the present disclosure. [Figure 28C] 1A-1C illustrate simulated results of image processing for marker detection in 2D projection images for tomosynthesis imaging, according to an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] The following is a detailed description of exemplary embodiments, in which reference is made to the drawings in which the same reference numerals sometimes identify the same element of structure or the same step of a method in each of the several views.

[0021] As used in the context of this disclosure, the terms "first", "second", etc. do not necessarily denote any order, sequential or priority relationship, but are merely used to more clearly distinguish one step, element, or set of elements from another, unless otherwise specified.

[0022] As used herein, the term "energizable" refers to a device or set of components that performs an indicated function upon receiving power and, optionally, upon receiving an enable signal.

[0023] In the context of this disclosure, the terms "observer," "operator," and "user" are considered equivalent and refer to an observing practitioner, technician, or other person viewing and manipulating an image, such as a dental image, on a display monitor. "Operator commands" or "observer commands" result from explicit commands entered by the observer, for example, by clicking a button on a camera, by using a computer mouse, or by touch screen or keyboard entry.

[0024] In the context of this disclosure, the phrase "in signal communication" indicates that two or more devices and / or components can communicate with each other via signals transmitted over some type of signal path. The signal communication may be wired or wireless. The signals may be communication signals, power signals, data signals, or energy signals. The signal path may include physical, electrical, magnetic, electromagnetic, optical, wired, and / or wireless connections between a first device and / or component and a second device and / or component. The signal path may also include additional devices and / or components between the first device and / or component and the second device and / or component.

[0025] The term "subject" refers to the patient's teeth or other part being imaged, and in optical terms may be considered synonymous with the corresponding imaging system "object."

[0026] In this disclosure, the term "detector" refers to an element that is placed in the patient's mouth to receive radiation and provide image content. Such detectors are digital detectors that provide x-ray image data directly to the imaging system.

[0027] In the context of this disclosure, the terms "pixel" and "voxel" may be used interchangeably to describe an individual digital image data element, i.e., a single value that represents a measured image signal intensity. Conventionally, individual digital image data elements are called voxels for three-dimensional volume images and pixels for two-dimensional images. Volume images, such as those from CT or CBCT devices, are formed by acquiring multiple 2D images of a pixel taken at different relative angles and then combining the image data to form a corresponding 3D voxel. For purposes of description herein, the terms voxel and pixel may generally be considered synonymous and describe image element data that may have a range of numerical values. Voxels and pixels have the attributes of both spatial location and image data code value.

[0028] Planes may be considered "parallel" if they are parallel to any direction by no more than 12°.

[0029] Although specifically described with respect to tomosynthesis imaging, an embodiment of the present disclosure may also be used with other types of volumetric imagers, such as computed tomography (CT) imaging systems, which create multiple 2D projection images that are used to reconstruct a 3D volume, such as a CBCT imager.

[0030] FIG. 1 is a schematic diagram illustrating components of an exemplary chairside tomosynthesis imaging device 100, in accordance with certain exemplary method and / or apparatus embodiments of the present disclosure. An x-ray source 10 directs radiant energy through a tooth 14 or other feature of a subject to an intra-oral detector 20 at each of a number of positional relationships of the x-ray source relative to the detector. Each positional relationship can be considered as a pair of spatial coordinates of the source and detector, respectively. A collimator 16 adjusts the angular extent of the radiation of the source 10 so that exposure is limited within the region of interest. An alignment device 70 senses and controls, as necessary, the alignment of the radiation field from the source 10 through the collimator 16 to provide radiation across the region of interest. The intra-oral detector 20 is in signal communication with a control logic processor 26, which acquires and processes image content to provide a tomosynthesis image on a display 28. Tomosynthesis imaging requires varying relative angles, or more generally, positional relationships of the source 10 relative to the detector 20, as described in more detail herein. A control logic processor 26 provides the necessary control for tomosynthesis image acquisition.

[0031] Tomosynthesis imaging requires the components shown in Figure 1 to acquire two or more 2D projection images of a region of interest, such as an image of one or more adjacent teeth. The image content produced includes some contour and depth information, but does not include the more geometrically complete 3D image volume data obtained from tomography such as a CBCT system.

[0032] Tomosynthesis data provides a measure of depth information without the full-volume image content. Tomosynthesis allows for the creation of slices into the imaged object, the slices being at different depths.

[0033] Acquiring reflectance images An optional reflectance imaging device 96 may also be provided as part of some exemplary chairside intraoral imaging method and / or device embodiments, for example to provide more accurate positioning information for the detector 20 placed in the patient's mouth. The imaging device 96 may provide contour imaging, such as by projecting a structured light pattern onto the intraoral features of interest. The contour information is then processed to create a 3D mesh representative of the surface features. To this end, the reflectance imaging camera functions as an optical scanner. Alternatively, the imaging device 96 may be a 2D camera, acquiring one or more monochromatic or color images from and around the region of interest.

[0034] Reflectance imaging may be used, for example, to determine head size and / or orientation. The acquired reflectance images may also serve as an aid to detect patient movement during tomosynthesis and / or other radiological image acquisition. Contour imaging camera images, such as those provided by Carestream Dental LLC's CS3600 intraoral scanner, may provide more information than 2D reflectance images to guide and / or correct the volume reconstruction process used in tomosynthesis and to detect movement during a tomosynthesis examination.

[0035] An optional ultrasound imaging device 98 may likewise be provided as a support system for the chairside oral imaging device 100 .

[0036] According to an alternative exemplary embodiment according to the present application, the full oral scanning device works in conjunction with a radiography system, which allows for simultaneous acquisition of both radiography and reflectance images, which may be useful for subsequent reconstruction processing, for example. Reflectance and tomosynthesis image content may be fused together to display some depth information with reference to very accurate surface contour information.

[0037] The types of imaging devices that acquire depth-resolved image content, e.g., optical coherence tomography (OCT) and ultrasound imaging systems, may not only obtain surface contour information from the captured reflectance signal, but may also provide some additional information for characterization of detected tissues and features to some depth below the surface. This type of depth-resolved image content may be a more useful aid, supporting and validating the positioning of the tomosynthesis acquisition, as well as helping to identify and report or correct patient motion detected during the imaging session. There may be supporting information acquired by the depth-resolved imaging device of features just below the surface, which may be more useful, for example, for positioning guidance and validation, than using only surface contour image content.

[0038] radiation source According to exemplary method and / or apparatus embodiments of the present disclosure, the x-ray source 10 is a Spindt-type field emitter (including carbon nanotube-based field emitters) and provides radiant energy from a number of distributed x-ray sources. The x-ray sources can be, for example, a distributed array of Spindt-type field emitters, which can be peripherally arranged around a central thermionic source. The x-ray sources are stationary or fixed in their positions relative to each other in the array, and the array itself moves as a single unit. This type of x-ray source is capable of rapid on / off switching on the order of microseconds.

[0039] Other suitable x-ray sources may include paired pulsed conventional fluoro-compatible thermionic electron sources in an array, where the sources are spatially separated. These options provide sufficient x-ray fluence at short exposure times while allowing exposure sequences without overheating.

[0040] Spindt field emitter-based x-ray sources have one or more cathodes within a vacuum chamber, each formed from a number of individual Spindt field emitters, which when supplied with an excitation current provide electrons that are then accelerated towards one or more anodes within the chamber.

[0041] Alternatively, the x-ray source may be a more conventional thermionic electron source, coupled to a transport device to provide the energy necessary to move the x-ray source along a linear or non-linear (e.g., curved) travel path, which may be segmented or continuous to direct the radiation at the target.

[0042] According to exemplary embodiments according to the present application, the same x-ray source can be used in either a set of modes for conventional radiography or 3D imaging. Thus, the same radiographic device can be used to acquire single shot radiographic images or to acquire and process projection images for tomography including CBCT, tomosynthesis, or for fluoroscopy or fluoroscopic imaging, as described in more detail herein.

[0043] Generator The radiation generator that is part of the x-ray source may provide pulsed or continuous operation. The generator may provide a single pulse or a series of pulses with varying pulse widths to provide suitable exposure conditions for a particular feature.

[0044] Imaging Detector The imaging detector in an exemplary method and / or apparatus embodiment is a small intraoral digital radiography (DR) detector that acquires image data fast enough for tomosynthesis imaging. The imaging detector may be of any suitable shape and may be rigid or flexible.

[0045] Signal communication with the imaging detector can be wired or wireless. The imaging detector can receive power from a cable or can be equipped with a rechargeable battery.

[0046] To meet the requirements of tomosynthesis imaging, the intraoral detector has a fast response time and provides an image acquisition rate sufficient for tomosynthesis acquisition, acquiring at least about 2 frames per second (fps), at least 5 fps, or at least 10 fps.

[0047] As discussed above in the Background section, both source and detector alignment and geometric calibration data of the source position relative to the detector are used to acquire and process the 2D projection images acquired by a tomosynthesis imaging device. The following discussion outlines some solutions to the alignment problem, as well as apparatus and methods for acquiring the geometric calibration data used for accurate tomosynthesis or tomographic reconstruction.

[0048] Certain embodiments of the present disclosure address the problem of geometric calibration using the spectral sensing capabilities of the tomosynthesis imager 100. For this purpose, the spectral characteristics of the tomosynthesis imager are used, as will be described in more detail below.

[0049] One aspect of tomosynthesis imaging that can be used to obtain spectral information of acquired image content relates to photon-counting x-ray detectors. Some advantages of photon-counting detection for imaging compared to conventional detectors using energy-integrating detection include (i) reduced electrical noise and improved signal-to-noise ratio, and (ii) improved image contrast, e.g., adjusting the weighting factor of the acquired image with energy binning. Thus, photon-counting tomosynthesis can improve diagnostic accuracy. The embodiments of the present disclosure can further utilize photon counting to obtain spectral information related to geometric calibration.

[0050] In photon counting, each incident photon generates a charge and each charge event is recorded. The actual count of photons, or a corresponding calculated value according to the count, is provided as image data for each pixel. Advantageously, photon counting has high immunity to noise when the pulse intensity exceeds the background noise level.

[0051] FIG. 2A shows in schematic form a photon counting sequence. An incident photon generates a pulse 180 at a given energy level corresponding to its frequency (or inversely proportional to its wavelength). The energy of the pulse 180 is compared to a threshold in a comparator 182 and shaped in a pulse shaper 184 to form a shaped pulse 188. A counter 186 then records the pulse event and provides a digital output, a pulse count value 190. A separate pulse count value 190 is obtained for each pixel element of the imaging sensor 170, which is used by the detector 20. The threshold value may be adjustable or selectable from a range of values, depending on the photon energy of interest. Photon counting x-ray detectors provide adequate performance at low signal levels, thereby reducing the x-ray dose given to the patient.

[0052] Photon-counting detectors that may be used in the geometric calibration methods of the present disclosure may be of either of two types. (1) Indirect detection using photon counting. Indirect detection has a scintillator layer that receives the x-ray signal and generates photons at an energy level corresponding to the received x-ray spectrum energy; and (2) Direct detection using photon counting. Direct detection involves a highly sensitive semiconductor material that converts the energy from the x-ray photons into a stream of electrons, the energy levels of which correspond to the spectral energy of the x-rays received.

[0053] A further advantage of pulse counting relates to its ability to count pulses 180 at multiple thresholds, where each value represents a spectral range of x-ray energies. With reference to the schematic diagram of FIG. 2B, two comparators 182a and 182b are shown to measure pulse energy. In this particular configuration, comparator 182a, pulse shaper 184a, and counter 186a provide a value of count 190a for all pulses that exceed a first threshold energy, and similarly, comparator 182b, pulse shaper 184b, and counter 186b consider only pulses that exceed a higher second threshold and provide count 190b accordingly. A simple subtraction then identifies the different power levels achieved with each pulse. It can be understood that three or more threshold levels can be measured using a corresponding arrangement of comparator circuits to allow pulse counting at any of several thresholds. In addition, the thresholds can be selectable, for example adjustable to adjust the response of imaging sensor 170 to various photon energy levels. Thus, for example, an operator may use a set of pre-set thresholds to distinguish between softer and denser tissues in the final image produced.

[0054] In addition to setting a minimum floor threshold (e.g., for noise reduction), embodiments of the present disclosure using multispectral x-ray imaging may also provide the option of using an additional upper or maximum threshold for photon energy. This upper threshold capability may serve many functions, such as reducing the occurrence of excess noise signal from metal artifacts or x-rays passing directly through the detection material.

[0055] The ability to count photons at different energy thresholds allows the intraoral detector to distinguish between energy levels resulting from irradiating an object with different x-ray wavelengths, as explained with reference to FIG. 2B, providing an additional dimension to the image data provided as a result of each exposure. This ability is described as multispectral or “color” x-ray imaging, and allows obtaining information about the material composition of the object's pixels. As is well known, two materials A and B may have different attenuation coefficients μ, which vary with the level of radiant energy, exposure E. At a given exposure, material A attenuates photons at an energy corresponding to material A. Similarly, radiation striking material B attenuates photons at a level corresponding to material B. If these photons of different energy values ​​can be distinguished from one another, it is possible to identify one or both materials within the same pixel or voxel image element of the resulting image. This same basic behavior of responding to radiation also allows some ability to distinguish between types of tissue. Different linear absorption characteristics allow different types of tissue to be distinguished, such as the ability to distinguish between types of bone.

[0056] Color x-ray using photon counting detectors offers low cost and low dose color x-ray imaging. The use of multispectral or "color" x-ray imaging can bring many potential benefits to the utility of intraoral imaging. These include minimization of metal artifacts, separate reconstruction of soft and hard tissues, more efficient segmentation algorithms of dental and bone features, improved pathology detection for cancer and other diseases, and detection of trace substances or contrast agents.

[0057] Among the techniques that can be used to provide tunable resolution and increase acquisition speed is detector binning, which is described in more detail herein. Binning groups uniform sets of adjacent sensor elements to provide a single average value for the individual areas of each set of pixels.

[0058] Source / Detector Alignment Detector alignment can be difficult in dental or intraoral radiography. The detector location is in the patient's mouth and is not visible to the technician. Instead, the technician typically places the detector in some type of holder and then inserts the holder into the mouth at a predetermined location. The holder may have a bite plate or other type of support member that helps to properly position the detector. As is well known, this type of holder can be cumbersome and uncomfortable for the patient. Holders and other positioning devices are not error-proof, and positioning errors that these devices have can mean that the images acquired are not suitable for diagnosis. For example, poor detector alignment can cause cone cuts, apical defects, and elongation and associated angulation or parallax errors. These alignment problems can require retakes, additional image captures, to obtain acceptable images. Retakes are undesirable because they prolong the patient's exposure to additional x-ray radiation and the patient's discomfort with the detector or sensor in the mouth.

[0059] Conventional x-ray sources include an aiming indicator to assist the technician in adjusting the position and angle of the x-ray source. These aiming indicators often use visible light to draw an outline that assists in centering the radiation beam. These work well when the radiation detector is visible, but fall short when the detector is not visible, such as in intraoral imaging. The technician must guess or estimate both the location of the intraoral sensor and the angle of incidence of the x-rays on the sensor.

[0060] The simplified diagrams of Figures 3A and 3B show how misalignment between an x-ray source 10 and a detector 20 can be caused. In these examples, the x-ray source 10 provides a visible light aiming indicator 12 that is used for aim centering. When accurate aim alignment is achieved as shown in example (a), the detector 20 is centered and displayed within the aiming indicator 12. Aiming is inaccurate in examples (b) and (d).

[0061] Proper angular alignment, or angulation, is also required to obtain the best imaging results. The incident radiation from the x-ray source 10 is preferably orthogonal to the detector 20 as shown in example (a). Line N in Figures 3A-B shows a line normal, or perpendicular, to the surface of the detector 20. Examples (c) and (d) show imprecise angular alignment. In example (c), the aiming or centering is accurate, but the angulation or pitch is inaccurate. In example (d), both the aiming (centering) and angulation (pitch) are inaccurate. In example (e), the detector 20 is rotated (rolled) in a plane.

[0062] 3A and 3B assume an orthogonal positioning of the x-ray source 10 relative to the detector 20. In some embodiments, an oblique orientation may be used.

[0063] Alignment and positioning are particularly important for volume imaging applications, where images taken at different angles are combined in some way to form volumetric image data.

[0064] In tomosynthesis, the relative motion between the source and detector introduces further complications into the alignment problem: in general, in the reconstruction process, it is most desirable to arrange the motion lines or arcs such that the spatial locations of the sources are in the same plane with respect to the detector surface or equidistant from the surface, and the motions coincide with pixel locations on the detector surface for each projection image acquired.

[0065] To better understand the parts and operation of the apparatus of the present disclosure, it is helpful to show how proper alignment can be detected by an imaging system. Referring to the schematic block diagram of Figure 4, an intra-oral imaging device 22 is shown that detects alignment of an imaging detector 20 with an x-ray source 10.

[0066] In the arrangement of FIG. 4, the detector 20 is placed at a detector location adjacent to the teeth 14 inside the patient's cheek 18. As part of the detector 20, several detectable elements 30 are incorporated, which are shown as electromagnetic signal emitters, e.g., radio frequency (RF) emitters. The detectable elements 30 are typically spaced apart from one another to provide triangulation information. The sensor 24 is itself aligned with and positionally coupled to the x-ray source 10, and senses the presence of the detectable elements 30 in some manner, e.g., by sensing the emitted RF signal. Methods for activating and sensing RF emitters, e.g., tiny emitters used in RFID tags, are well known to those skilled in the art of signal detection. A control logic processor 26 is in signal communication with one or more sensors 24 and performs conventional triangulation calculations based on received signals from the detectable elements 30 and the sensors 24 at known positions relative to the x-ray source 10, or other detectable features. This is performed to determine the corresponding positional and angular alignment of the detector 20 in the patient's mouth relative to the x-ray source 10. The operator console display 28, a computer display monitor, can then present the alignment information to the operator and recommend any needed adjustment settings. The sensor 24 is energizable to receive electromagnetic signals of one or more predetermined frequencies.

[0067] Certain exemplary method and / or apparatus embodiments of the present disclosure improve upon the basic system of FIG. 4 by providing alignment information to the technician that can be more easily used, particularly when this information is required to acquire the individual images used to form the volumetric image. The exemplary alignment device of the present invention can project an image onto the cheek or other portion of the dental patient as a guide for properly aligning the x-ray tube relative to the detector position and angle. Referring to one embodiment of the imaging device 36 of FIG. 5, the control logic processor 26 obtains the alignment information in a manner similar to that described in FIG. 4. Additionally, as shown in FIG. 5, the control logic processor 26 can be in signal communication of image data with a projector 40 to project an image onto the patient's cheek 18, lips, or face.

[0068] The perspective view of FIG. 6A shows, in schematic form, how triangulation can be used to indicate the position and angle of the detector 20 to determine an alignment offset in one embodiment. In one embodiment, the sensors 24a and 24b, RF transmitters, are in known positions relative to the x-ray source 10, for example mounted near the x-ray source on an x-ray tube. Signal emitters or other types of detectable elements 30 are typically arranged in pairs and positioned at the corners of the detector 20. Each detectable element 30 has a detectable characteristic that can be sensed by the sensors 24a and 24b. In one embodiment, each detectable element 30 is an RF device, for example, generating an electromagnetic field in response to a signal transmitted from its corresponding signal receiver, the sensor 24a or 24b. The phase, intensity, or other characteristics of the emitted electromagnetic field are measured by the corresponding sensor 24a and 24b and used to determine the relative distance between the emitting and receiving components. For example, in the RF detection embodiment of Figure 6A, good alignment is achieved when the signals of each pair of emitters acting as detectable elements 30 are in phase. An out-of-phase condition indicates poor alignment and may display an indication of the need for adjustment. Sensors 24a and 24b are in signal communication with control logic processor 26.

[0069] Similarly, the relative signal strengths can alternatively be used to indicate the position and angle of the detector 20 relative to the x-ray source for measuring alignment offsets. Using this approach in RF embodiments, the closest signal emitter acting as the detectable element 30 has the strongest signal strength at the sensor 24a or 24b correspondingly. When the arrangement of FIG. 6A is used, equal strength of the emitted signals from all four emitters, or other types of detectable elements 30, indicates good alignment. When the signal strengths differ, the pattern for those variations can be used to indicate what adjustments are needed. As an example, U.S. Patent Application Publication No. 2009 / 0060145, entitled "Positioning Adjustment of a Mobile Radiology Facility" by Tranchant et al., describes a position detection system that uses triangulation and sensing of multiple emitted signals to calculate alignment positioning. It will be appreciated that proper alignment may be determined using one or more sensors 24 and detectable elements 30 using any of a number of different configurations, which are well known to those skilled in the art of signal processing and position sensing.

[0070] In one alternative embodiment, as shown in Figure 6B, the emitter and detector arrangement shown in Figure 6A is reversed, such that one or more emitters providing one or more detectable elements 30 are mechanically coupled to the x-ray source 10 and two or more sensors 24 are attached to the detector 20. In the embodiment shown in Figure 6B, for example, the detectable elements 30 are shown in dashed outline and are coils that generate an electromagnetic field that is sensed by the sensors 24. The sensors 24 are in signal communication with the control logic processor 26 via either a direct (e.g., wired) or indirect (e.g., wireless) connection.

[0071] Alternative Alignment Mechanisms In one exemplary embodiment, an intraoral scanner or other reflectance imaging sensor may be used as an aid in source alignment with the detector. The optical scan data obtained from the contour image or conventional reflectance image may be analyzed as a type of "scout" scan to determine the desired trajectory for the tomosynthesis scan.

[0072] In another exemplary embodiment, ultrasound imaging may also be used as an alignment aid for source positioning. Ultrasound may be particularly useful with its ability to image soft tissue structures within the anatomy.

[0073] It should be noted that the alignment mechanism of the CNT source is adjustable, allowing one to control the trajectory of the relative positional change of the radiation source for each subsequent image.

[0074] One exemplary method and / or apparatus embodiment according to the present application provides autofocus and alignment capabilities using an arrangement of markers embedded in a holder used to position the intraoral sensor. FIG. 6C shows a holder 72 in a frame 78 having an arrangement of radiopaque markers 74 around a central opening 76 through which the x-ray source 10 (not shown in FIG. 6C) is directed. FIGS. 6D and 6E show how the markers 74 appear in the acquired images along the boundaries of the imaged intraoral features. The use of the alignment markers allows image processing to correlate the positions of successively acquired images to precisely align the projected images with each other for subsequent reconstruction.

[0075] It can be observed that solutions such as those shown in Figures 6A-6C can be used for alignment with any number of source and detector arrangements, as described in more detail herein.

[0076] Projection of contours or other alignment feedback to the patient With reference to the perspective views of Figures 7A and 7B, further advantages of an exemplary embodiment of the present invention providing optional image projection are shown. A projector 40 is positionally coupled to the x-ray source 10, e.g., toward the end of the x-ray tube, or attached to another part of the x-ray system, to project a two-dimensional image, e.g., onto the patient's cheek, indicating the location 42 (shown in dotted outline) of the hidden detector 20, and also indicating the x-ray source's aiming indicia 12 if not already provided by the x-ray source 10. Figure 7A shows an example where the aiming alignment is inaccurate, with the location 42 not aligned with the aiming indicia 12. Figure 7B shows an example where the aiming alignment is accurate, with the location 42 centered between the aiming indicia 12.

[0077] Projector 40 can be any of several types of imaging projectors that can be mounted above x-ray source 10. In one embodiment, projector 40 is a picoprojector, such as the picoprojector display from Microvision Inc. of Redmond, Washington, USA. Such devices are advantageous for a number of reasons, including small size, light weight, and low power requirements. Such picoprojectors, used in mobile phones and other highly portable electronic devices, scan one or more low-power lasers over the display surface. Picoprojectors require minimal optics for projection at various distances. The lasers themselves are rapidly powered on and off as needed, so that power is consumed only for image pixels being projected. This allows the picoprojector to operate at low power levels, allowing battery power to be used for projector 40. Alternative embodiments use other types of electronic imaging projectors, for example projectors that employ digital micromirror arrays such as Texas Instruments, Inc.'s Digital Light Processor (DLP), arrays of microelectromechanical grating light valves such as Silicon Light Machines, Inc.'s Grating Light Valve (GLV) devices, or liquid crystal devices (LCDs), including liquid crystal on silicon (LCOS) devices.

[0078] When using a laser as the illumination source for projector 40, additional measures can be taken to minimize the incidence of coherent laser light into the patient's or practitioner's eye. Very low power lasers, e.g., solid-state lasers, can be used at scanning speeds that deliver only a very small amount of light intensity to any point. Diffusion elements can be provided in the light path, e.g., to scatter some of the laser light and reduce the intensity with little or no effect on the quality or usefulness of the projected image. Light emitting diodes (LEDs) or other low power solid-state illumination sources can alternatively be used, e.g., organic LED (OLED) devices.

[0079] The image projected by projector 40 (FIGS. 7A and 7B) may have image content that may be in any of a number of forms and may include both the aiming indicia 12 and an indication of the position 42 of detector 20 relative to the x-ray source. Alternatively, if the aiming indicia 12 is already provided by the x-ray system, projector 40 may only provide a projection indicating the position 42. Because projector 40 uses a two-dimensional imaging device, the displayed image may have multiple portions and may include additional text areas, orientation markers, and other elements. The position 42 may be indicated in the form of an outline as shown in FIGS. 7A and 7B, or may be displayed in some other manner. In one exemplary embodiment, the value of the angular offset of detector 20 is shown as a numerical message display on the patient's cheek. Alternatively, the video or other capabilities of projector 40 may be used to provide additional position and angle information as image content.

[0080] The use of color can help indicate the relative amount of alignment offset in various ways. For example, even if the outline of the detector 20 is projected onto the cheek surface, it may be difficult for the technician to know how to adjust the angle alignment. Displaying the indicia 12 and the position 42 in different colors can help guide the technician in adjusting the angle of the x-ray tube until, for example, both the aiming indicia 12 and the position 42 are displayed in the same color. Blinking of different parts of the indicia or of the displayed elements helps to indicate and guide the alignment adjustment. An audible beep may be provided to indicate whether the alignment is acceptable or not. Static indicia such as an arrow or target symbol can be projected onto the patient's cheek as image content. Animation can be provided to guide the adjustment.

[0081] In one exemplary embodiment, the projected image from projector 40 (FIG. 7B) instructs the technician how to re-aim the x-ray source 10 or adjust the position of the procedure chair to set the next image in the sequence. Projected colors, patterns, alphanumeric text, animation, flashing or blinking, or other mechanisms may be used to guide positioning adjustments between image captures.

[0082] A patient head support device is provided to stabilize the head position during the tomosynthesis image acquisition cycle. It should be noted that any type of head rest or other support mechanism cannot be metal or other highly radiopaque material. The patient head support device can be, for example, donut-shaped, expandable, or inflatable.

[0083] Collimation In selected exemplary method and / or apparatus embodiments, collimation is required to restrict the radiation field to a region of interest (ROI) within the patient's mouth.

[0084] One beneficial aspect of collimation relates to the elimination or reduction of cone cuts, where excess radiation from projected x-rays impinges on areas outside the region of interest.

[0085] The difficulty of distributed source configurations such as CNT arrays relates to the need for proper collimation of the radiation. Among its functions, collimation controls the spread of the radiant energy so that the radiation is properly directed to the anatomical structures of interest and the radiation field does not extend beyond the outer edge of the imaging receiver. Collimation also helps to reduce scattering. With CNT and other types of small x-ray sources in an array, collimation presents particular challenges. One set of problems relates to dimensional constraints. Because the spacing between x-ray sources is typically small, it can be difficult to effectively separate the radiant energy from any individual source, crosstalk can occur, and the edges of the radiation field are difficult to clearly define. Yet another complication relates to identifying the radiation field for imaging from each source. With conventional radiography sources, the problem is easily solved, using a light source coupled to the radiation source to delineate or highlight the radiation field by using the collimator edges themselves to delineate the extent of the radiation field. However, it may be impractical or impossible to provide a corresponding dual-use configuration using the collimator openings provided for CNT and other types of distributed array sources.

[0086] The simplified schematic diagram of FIG. 8A illustrates some of the geometrical considerations and relationships that are generally relevant to the x-ray collimation of a single x-ray source 210 and establishes some prescriptions that are useful in the subsequent description of the collimation of an array of x-ray sources. The x-ray source 210 is idealized to a first approximation as a point source. Radiant energy from the source 210 is directed along an illumination path that extends through a first aperture 122, which is typically close to the source 210 and may be optional under some conditions for very small x-ray sources. The radiant energy then continues along the illumination path through a second aperture 124, which forms an x-ray field 130 on the detector 20. The shape and dimensions of the illumination path, which determine the aspect ratio of the x-ray field 130, are then determined by geometrical constraints, such as the size and mutual position of the apertures 122, 124 relative to the source 210, as well as the source-to-image distance (SID). The shape of the x-ray field 130 is typically bounded by the dimensions of the detector 20, but may be smaller and of different shapes depending on the anatomy being imaged. Figure 8A shows the geometric relationships of a single source 210, while subsequently described embodiments have multiple x-ray sources 210, each with collimation along its illumination path in a manner similar to that shown in Figure 8A.

[0087] According to an exemplary embodiment according to the present application shown in FIG. 8B, an array of sources 20 can be used with a rotatable collimator plate assembly 220 to create radiation fields of various shapes and aspect ratios depending on the dimensions of the apertures 222a, 222b, the rotation angle of the collimator plate assembly 220, and the arrangement of the corresponding sources 20 energized in the energization sequence. Thus, for example, with reference to FIG. 8B, the radiation source assembly 200 has a generally square shape with the sources 20 distributed along the sides of the square. The collimator plate assembly 220 of FIG. 8B has apertures 222a and 222b of multiple aspect ratios. One set of apertures 222a are square and another set of apertures 222b are rectangular. By rotating the collimator plate assembly 220 to different positions and energizing the corresponding open-hole sources 20 for a given radiation field shape, the sources 20 can be used in succession to provide the radiation field shape and angle variations required for tomosynthesis.

[0088] 8C and 8D show another alternative configuration, where the collimator plate assembly 220 is translatable back and forth in one direction along the direction indicated by line L1. In this example, the radiation source assembly 200 is a linear array of sources 20. A set of square apertures 222a are arranged such that a half subset of the apertures is aligned with the plate assembly 220 and the source 20 in a first position (FIG. 8C) and another half subset of the apertures is aligned with the plate assembly 220 and the source 20 in a second position (FIG. 8D). FIG. 8E shows another arrangement, where three different types of apertures are provided, where the set of square apertures 222a has two subsets arranged in a similar manner as shown in FIGS. 8C and 8D, a set of rectangular apertures 222b, and a set of elliptical apertures 222c are also used in a similar manner. The shapes of the apertures 222 can be mixed (not shown).

[0089] 8F and 8G show another alternative embodiment, where the collimator plate assembly 220 has a generally curved or tubular shape and the apertures 222a are arranged for the x-ray sources 20 of the linear radiation source assembly 200. In the arrangement of FIG. 8F, the collimator plate assembly 220 is moved in a linear manner to shift the apertures 222a between the subsets of the x-ray sources 20. In the arrangement of FIG. 8G, the collimator plate assembly 220 is rotated about the linear array to shift the apertures 222a between the subsets of the x-ray sources 20.

[0090] The collimator plate assembly 220 may be formed from a pair of metal plates spaced apart to form apertures 122 and 124 (FIG. 8A), with aperture 124 sized and positioned for proper beam shaping. The aperture is aligned with the source 20 based on the required beam profile and angle. For collimation control, the aperture 122 near the source may be in a fixed position, with only the distal aperture 124 adjustable.

[0091] Scanning Sequences for Tomosynthesis Imaging The alignment device provided by the triangulation sensing device of FIG. 6A or FIG. 6B can be used to assist in capturing a series of images of the same tooth or other structure, taken in rapid succession, each with a slightly different positional relationship between the x-ray source and detector, to form a limited depth volumetric image. As noted in the Background section above, this type of volumetric imaging has diagnostic value and advantages over a single x-ray image, but without the expense and dose requirements of full-scale CBCT imaging. Additionally, unlike CBCT imaging, limited depth volumetric images from tomosynthesis can be acquired with the patient seated in the treatment chair.

[0092] 9A, from a top view, a schematic block diagram of an imaging pattern is shown that is used to obtain a volumetric image of a limited depth from a patient 32 using a limited number of x-rays from a single source 10 and digital detector. Using the x-ray source 10, exposures are directed to the detector 20 from several angular directions, which are shown as capture angles, or exposure angles, in FIG. 9A, along a non-linear, curved, or arcuate path A. At each of two or more x-ray source and detector configurations, two angles Q1 and Q2 are presented as examples in FIG. 9A, radiant energy is directed to the detector 20, and corresponding image data from the digital detector is acquired by the control logic processor 26 and stored as component or projection images 44, labeled according to the relative acquisition geometry of the images, such as by the orientation of the exposure angles. In this way, one component image 44 is acquired and stored for each x-ray source and detector pair configuration, alternately considered for each exposure angle. Control logic processor 26 may then use the combined data from the individual component 2D projection images 44 to create a 3D volumetric image as a composite image.

[0093] It should be noted that the pattern tracked by the change in relative position of the x-ray emitter with respect to the detector can be linear or curved, for example as shown in the top view of FIG. 9A.

[0094] Additional sensing components and associated logic are used to provide source and detector position and angle information for each image acquired. In one embodiment, for example, fixed position and angle coordinates are assigned to the initial spatial position and relative angular orientation of the x-ray source 10. The system logic then records the modified relative position and angle, which corresponds to each imaging position of the series of 2D projection images acquired. This data then provides the reference geometry required for the reconstruction of a 3D volumetric image from the series of 2D image captures. The spatial position data can be acquired in a number of ways, for example, using an angle sensor 48, which is coupled to a gantry or other transport device used to move the x-ray source 10.

[0095] For the limited angle volumetric imaging of tomosynthesis to function correctly, the angular orientation and spatial location of the x-ray source 10 relative to the detector 20 must be known for each projection image acquired throughout the imaging cycle, so that the resulting component data can be properly aligned and correlated between projection images. For the embodiment shown in FIG. 9A, in the perspective view of FIG. 9B, the spatial positions of the head of the patient 32 and the detector 20 (shown in dashed outline in FIG. 9B) are rigidly fixed in position while the x-ray source 10 is orbitally moved from one relative angular orientation to the next. It may be necessary to mechanically fix the spatial position of the detector 20 relative to the object being imaged. With respect to FIGS. 9A and 9B, for example, one or more bite blocks or clip-on devices may be useful to rigidly fix the detector 20 in position within the mouth of the patient 32.

[0096] FIG. 10A is a schematic diagram showing an exemplary intra-oral imaging apparatus for tomosynthesis imaging with an x-ray source 10 conveyed along a curved or arc-shaped trajectory 132. The arc-shaped trajectory 132 is curved to approximate an arc substantially centered on the detector position. The detector 20 is attached to a holder 72 and held in the patient's mouth. The holder 72 provides a type of positioning device and correlates the detector position to the collimator. A frame 78 is suspended outside the mouth and provides a aiming and alignment device for the x-ray source 10 and a holder for positioning the secondary collimator 56. The schematic diagram of FIG. 10B shows a similar exemplary arrangement of the apparatus 100 using a linear trajectory 134. In the embodiment of FIG. 10B, the x-ray source 10 pivots to different angles as it is translated along a linear path, thereby emulating the radial arc translation of FIG. 10A.

[0097] In the embodiment of Figures 10A and 10B, the detector 20 is rigidly coupled to the frame 78 as previously shown in the example of Figure 6C. The holder 72 acts as a positioning device 194 to correlate the detector and collimator positioning and fixes the relative position of the detector 20 and the frame 78. Different sizes of holders or different holder configurations can be used for different patient head sizes. Alternative exemplary embodiments can use different arrangements of sensors and encoders to provide a mechanical or sensed positioning device that positions the detector 20 relative to the collimator 56 and frame 78 using signals obtained from one or more sensor and encoder devices.

[0098] 10C shows an embodiment of an exemplary frame 278 having multiple articulating sections 280a, 280b, 280c with an encoder 282 at each adjustable joint to report sensed expansion and rotation data. This arrangement provides a positioning device 194 that allows for patient size changes, provides repositioning of the detector 20 relative to the collimator 56, and provides sensed data that can be used to correlate component positions to a reference position and to each other. Alternatively, accelerometers or electromagnetic, magnetic, or radio frequency (RF) sensing may be provided and used as the positioning device 194 to correlate the position of the detector 20 to the secondary collimator 56 and relate these positions to the position of the x-ray source at any acquisition angle in the tomosynthesis sequence.

[0099] FIG. 11A is a perspective view showing the configuration of the x-ray source 10 for the radial path arrangement shown in FIG. 10A, showing an example of an isocentric signal-to-detector distance (SDD). The x-ray source 10 moves along rails 146, following a curved or arcuate trajectory 132 (FIG. 10A), and is driven by a transport 136. The source 10 has a primary collimator 150 that is integral to the source hardware. In the context of this disclosure, the designation "primary collimator" applies to any collimator(s) that are integral to the x-ray source and cannot be separated from the x-ray source. A secondary collimator is provided using a frame 78, as described next. A camera 140 coupled to a guide 144 may be used to assist with source / detector alignment. The perspective view of FIG. 11B shows an alternative configuration of an exemplary linear transport, using a non-isocentric SDD.

[0100] The schematic diagram of Figure 12A illustrates the use of an exemplary marker guide embodiment 160 for markers coupled to the intraoral detector 20 via a frame 78. The marker guide 160, described in more detail herein, provides several functions to aid in the alignment and collimation of the x-ray source 10 and relate the spatial position of the intraoral x-ray detector to the position of the x-ray source 10. Figure 12B illustrates a configuration in which the marker guide 160 is decoupled from the detector 20.

[0101] When using radiopaque markers, the spatial position of the source relative to the detector can be determined from the image content, preferably around the edges of the FOV. The collimator and detector can be mechanically decoupled with these markers being provided to the radiation field and displayed in the image content. When using radiolucent optical markers, the collimator and sensor must be mechanically coupled or have some kind of sensed positioning. Alternatively, the optical markers can have a 3D orientation that allows the position of the source relative to the detector to be determined from camera, reflectance, or optical image content, which can be acquired sequentially or simultaneously with the x-ray image content.

[0102] FIG. 13A shows an embodiment of an exemplary treatment system 164 with a dental procedure chair and other equipment. A frame 78 houses the marker guide 160 and provides the secondary collimator 56, which is supported from a mount 162 on the system 164, such as from a base 198 or from the dental chair. Alternatively, the collimator 56 or its support frame can be attached by supports extending from the ceiling. This arrangement can help to provide inherent and sensed alignment of the x-ray source 10 to the patient and the intra-oral detector 20 as well as support the weight of the frame 78 providing the secondary collimator, with attachments to stabilize the position of the collimator 56 without the need for the patient to hold the device. As shown by the arrow, the x-ray source 10 can be moved towards the frame 78 so that it is at the proper distance and alignment for the image acquisition sequence. The frame 78 can be part of a headrest, such as an adjustable headrest, for patient positioning to enable tomosynthesis imaging.

[0103] 13B is a top schematic view showing the positioning of the frame 78 with the secondary collimator, which is attached to an exemplary support embodiment extending from the dental chair, the ground or ceiling, or another nearby support structure external to the patient. An arm 192, or other linkage or coupling device, may also be attached to the frame 78 to guide the positioning and movement of the source 10 during tomosynthesis imaging. With this arrangement, the source 10 is effectively coupled to the secondary collimator 56 of the frame 78. This eliminates the need for any type of alignment mechanism to align the source 10 to the secondary collimator. This arrangement may be used with radiopaque or visible markers to indicate the relative position of the detector.

[0104] One or more sensors 152, for example electromagnetic devices such as accelerometers or Hall sensors, may be provided to detect movement and positioning of the frame 78 on the mount 162 and help determine the precise alignment of the marker guide 160 relative to the detector 20 and x-ray source 10.

[0105] Marker Guide Configuration 14A and 14B show the assembly and components of a marker guide 160 according to an exemplary method and / or apparatus embodiment of the present disclosure. FIG. 14A shows the exemplary marker guide 160 assembled. FIG. 14B shows the layered components for forming the marker guide 160. The collimation layer 166 acts as a secondary collimator of the incident x-ray beam and is held substantially against the patient's face and includes a window 172 bordered with a radiopaque shield for local collimation. The shield may be provided, for example, by a lead-laced material. Layer 2 174 and Layer 3 176 provide optical markers 178 to aid in the extraction of the scanning geometry for alignment. Separating layers 174, 176 by a distance d helps facilitate alignment measurements. The markers 178 have a stacked arrangement as shown in FIG. 14B and may be sensed by the camera 140 and the resulting image processed to obtain alignment data. The markers 178 may be of different shapes (e.g., asymmetric), layers, 3D configurations, or colors to aid in alignment detection. One or more of the markers 178 may alternately be radiopaque.

[0106] In Figure 15, frame 78 is shown in position relative to the patient's face, with support structure for proper alignment of collimation layer 166 of marker guide 160 relative to frame 78 (Figures 14A, 14B, 15). Figure 15 also shows the position of intra-oral detector 20 properly aligned relative to frame 78.

[0107] 16 shows a side view and a top view of the frame 78 and associated components. The bite block 154 helps to stabilize the position of the detector 20 in the mouth. An adjustable rod 156 allows for positioning of the bite block 154 and detector 20 to suit the patient's comfort.

[0108] It should be noted that the frame 78 and its associated marker components may be used with a single source x-ray source 10, or an array of x-ray sources, such as those provided using a Spindt type field emitter based x-ray source.

[0109] The schematic diagram of Fig. 17 shows exemplary relative motion paths that can be tracked relative to an imaged object for tomosynthesis imaging. Linear motion path 110 or radial motion path 112 can be provided without rotating the detector. In motion path 114, the source remains in a fixed position while the detector rotates. In motion path 116, a planar or spherical source motion is provided, and the detector does not rotate. For any of motion paths 110, 112, and 116, the relative motion can be provided by sequentially energizing individual sources of an array, such as, for example, a CNT source array.

[0110] It can be appreciated that control logic processor 26 acquires and stores both image data and position information when performing tomosynthesis imaging. As each image is acquired, control logic processor 26 stores the image data and corresponding information regarding the relative spatial positions of the energized x-ray source and detector 20. The position data and image data may be stored as part of the same data structure, such as an image data file, or may be stored in separate data structures, such as separate files or database locations. In one embodiment, control logic processor 26 then optionally provides information indicating a recommended position adjustment of the x-ray source for acquiring the next x-ray image at the next spatial position and next angular orientation. This information regarding the recommended position adjustment may be provided in a number of ways, including information displayed on display 28 (FIG. 5), using an audible cue, or by providing graphical guidance to the operator for setting the next exposure, which may take the form of projected image content and format, for example, by projecting instructions or targeting information onto the patient's cheek. The relative position information associated with each image is stored in some form and is used by image processing logic on control logic processor 26 to create the volumetric image.

[0111] Use of Spindt-type field emitter-based or other radiation source arrays The schematic diagram of FIG. 18 illustrates the use of a radiation source array as the x-ray source 10. Each x-ray source 38 is from a cathode utilizing a Spindt field emitter. With a Spindt field emitter cathode, the x-ray sources are stationary or relatively fixed in position relative to each other in the array, and the array itself moves as a single unit. This type of x-ray source is capable of rapid on / off switching, on the order of microseconds. Other suitable x-ray sources may include spatially separated pairs of pulsed conventional fluoro-compatible thermionic electron sources. These options provide sufficient x-ray fluence at short exposure times while allowing exposure sequences without overheating.

[0112] According to an exemplary embodiment, each individual source 38 has its own collimator 16, such as the exemplary embodiment shown in FIG.

[0113] Alternative concepts of relative movement Data must be acquired to identify the spatial position of the detector 20 and the relative spatial position of the x-ray source 10 for each image.

[0114] 13B, the x-ray source 10 is coupled to the frame 78 by a pivot arm 192, which serves as a support and guide for movement of the source 10 in a curved or arcuate path. This arrangement allows movement of the source 10 on a well-defined angular track, simplifying the design of the transport apparatus for movement of the source 10 and further simplifying the collimation design.

[0115] 19, the x-ray source 10 is fixed in position and the patient 32 is rotated, for example by incrementally rotating the treatment chair, to shift from one exposure angle orientation to the next. Again, relative position information of both the detector 20 and the x-ray source 10 must be established and stored for each component image by the control logic processor 26 or associated processing device.

[0116] Image acquisition process The logic flow diagram of FIG. 20 illustrates a series of steps for acquiring a series of component images in one exemplary embodiment. An initialization step 80 starts the sequence and acquires data regarding the initial starting position and angle. In one embodiment, the initialization step 80 also sets or calculates the number of images to be acquired and the corresponding exposure angle for each image. This information may be fixed or variable and may be calculated using the control logic processor 26 or may be input by the dentist or technician using set-up software in communication with the control logic processor 26. The detector 20 is securely placed in the patient's mouth and the necessary source and detector alignment is at least roughly performed by the technician. In a set-up step 82, the imaging device 22 provides the necessary image representation that is projected onto the patient's face or head to help guide the alignment and aiming of the x-ray source 10 as described above with reference to FIGS. 7A and 7B. In one embodiment, the image content projected onto a portion of the patient changes depending on the relative accuracy of the angular orientation. This may be a change in color, intensity, flashing, or other attributes of the projected content. It should be noted that the display may give hints or suggestions as to optimal positions for each subsequent radiographic image capture, but it is important that the actual spatial positions are accurately measured and recorded in order for the limited angle volumetric imaging algorithms to run properly.

[0117] 20, each component image is acquired in an image capture step 84 and the image is saved along with information about the actual measured spatial location and angular orientation at which the exposure was acquired. A decision step 88 checks to determine if all component images required according to the initialization step 80 have been acquired, and loops back to the setup step 82 when a subsequent image is required. At the end of this process for image capture, a volume image creation step 90 is performed to create a composite volume image of the results obtained from this sequence. A display step 92 then displays the created volume image.

[0118] 21 is a block diagram illustrating the spatial location and angular orientation associated with the image data of each component image 44 in the set of acquired images. In the exemplary embodiment shown, the spatial location data field 50 and the angular orientation data field 52 are stored with the x-ray image data 54, for example, by storing the measured location and angular geometry in the header portion of the x-ray image data file. Alternatively, the spatial location and angular orientation data may be stored separately and linked or associated with the image data. This information is necessary to properly reconstruct the volumetric image.

[0119] FIG. 22 is a logic flow diagram illustrating optional system activities within the image setup step 82 of FIG. 20 in preparation for each image capture in the sequence. A calculation step 60 uses position coordinates and angular orientation data from the x-ray system or stored with the previous image to calculate the next position and angular orientation for the relative movement of the x-ray source 10 and / or detector 20. An optional target projection step 62 then projects an image onto the patient, where the image indicates the position and angle adjustments required between the x-ray source 10 and detector 20 to acquire the next x-ray image at the next spatial position and angular orientation. As previously mentioned, the optional projected display may indicate the required adjustments using colors, blinking or other effects, numbers, icons such as directional indicators or arrows, or other visual effects. Then, in a looping operation, a reevaluation step 64 periodically re-adjusts the projected display according to the measured changes in position adjustments and angles made by the technician. If the adjustments are correct within a predetermined tolerance, then a correct adjustment indication step 66 is performed to indicate that the adjustments are acceptable for acquiring the next image.

[0120] Given that information is available on relative positions when using the arrangement of components shown in Figures 9A-11, an exemplary embodiment of the present invention uses continuous recalculation and iterative checking of sensors and other position-sensing components to correct and accommodate small position changes and patient motion. In this arrangement, the detector 20 and x-ray source 10 do not have to have fixed, predetermined positions relative to each other or to achieve their calculated positions exactly for the next image. In either case, however, the detector 20 must have a fixed spatial position relative to the tooth or other object being imaged. The programmed image processing logic may accommodate position changes that are within a reasonable angular range, for example. In one embodiment, one or more additional position sensors at fixed spatial positions are used to establish reference points for angular and positional orientation. In addition, automatic detection and correction of patient motion artifacts may also be performed using image processing techniques well known to those skilled in the art of image acquisition.

[0121] A limited angle volumetric image is formed from two or more component x-ray images and provides some volume-related information of the tooth or other imaged structure. Advantageously, this is provided without the higher levels of exposure required for full CBCT imaging and without the need for specialized CBCT gantry and associated equipment. Position information is obtained using the sensor 24 and detectable elements 30 and used by a 3D image reconstruction algorithm to create a corresponding volumetric image including the tooth or other feature and populate voxels within the volumetric image with appropriate data values. The volumetric image may be formed without the need for complex filtered backprojection algorithms typically used for CBCT reconstruction, for example. The resulting image may be viewed on a conventional display monitor or may be observed, for example, using a stereoscopic viewing device. The required volumetric image may be dynamically created, for example, according to a preferred viewing angle indicated by the practitioner.

[0122] Image acquisition sequence variations According to an exemplary embodiment according to the present application, the image acquisition sequence may be altered to acquire one or more images under different conditions. For example, across a series of images taken at tomosynthesis conditions and angles, one or more images may be captured under different conditions, for example using settings typically applied to conventional 2D radiography. This may be, for example, the central image in the sequence, for example the 10th or 11th image in a series of 20 tomosynthesis image captures. Different capture conditions, including exposure settings, binning, dual energy, and other parameters, may be used for images acquired in this manner.

[0123] When one or more images are acquired under different conditions, features such as higher fidelity and sharpness of these images may be used to improve the image content of other images in the series.

[0124] Detector Binning Binning methods can be used to aid in high speed image acquisition. Binning groups sets of adjacent pixels together to speed up access to image data and data refresh cycles. Binning is usually done in a symmetric pattern, such as 2x2 binning, 3x3 binning, etc. However, binning can also be performed in one direction, for example 2x1 binning. Asymmetric binning is useful for volumetric imaging and has binning in a direction parallel to the relative motion of the x-ray focal spot, which differs from binning in an orthogonal direction.

[0125] Perspective For example, fluoroscopic imaging methods, including x-ray fluoroscopy, acquire images of an object in rapid succession to provide a continuous view of the object that may have a video appearance. In fluoroscopy, volume reconstruction is not provided, instead a sequence of acquired radiographic images is displayed. In fluoroscopy, there is no relative motion between the x-ray source and detector, and the same source and detector geometry is applied to each image acquired.

[0126] Fluoroscopy can be a useful tool for the practitioner, providing a progressive or "real-time" view of an area of ​​interest, such as one or more teeth or a portion of a dental arch. Fluoroscopy can be combined with visualization software to show drill angles or other features that are useful during the procedure. Fluoroscopy acquisition utilizes high speed digital detectors with excellent resolution.

[0127] In general, fluoroscopy has acquisition times similar to those required for tomosynthesis, but with somewhat longer x-ray exposure times. There can be a trade-off between spatial resolution and dose with fluoroscopy acquisitions.

[0128] Dual Energy Spectral Imaging Dual energy imaging embodiments of certain exemplary methods and / or apparatus allow for the advantage of improved ability to analyze different types of tissue found in the imaged anatomy. Dual energy or multispectral imaging can be obtained using conventional detectors, as well as using photon counting detectors with multiple thresholds, as previously described with reference to FIG. 2B. Alternatively, other detector arrangements can be used with dual energy x-ray sources.

[0129] Dual energy (DE) imaging has been used as an alternative method to reduce noise content and differentiate between various types of imaged anatomical structures and materials. In traditional DE imaging, low and high kVp exposures of the same anatomical structures are consecutively performed, so that their results can be easily combined without the need for extensive registration techniques. This can be useful for subsequent segmentation of bone features, e.g., allowing for more accurate interpretation of x-ray image content. In the case of tomosynthesis and full 3D volume imaging, such as that provided by CBCT and CT devices, providing dual energy image content for reconstruction and subsequent analysis can be of great advantage.

[0130] Dual-energy tomosynthesis involves imaging content acquired using two different energy (or wavelength) bands that generate two different radiation spectral energy distributions, allowing different structures to be reconstructed from the same imaged tissue.

[0131] The exposure technique settings may be changed from one projection image to the next during image acquisition.

[0132] Reconstruction The control logic processor 26 or associated processors, or other computers used for image processing, may implement any of several known techniques for limited angle tomosynthesis reconstruction well known to those skilled in the art of 3D imaging. For example, several have described many reconstruction algorithms used to solve similar types of problems in limited angle mammography imaging, including backprojection, algebraic reconstruction, and probabilistic techniques. The aforementioned Siltanen et al. patent describes a method of 3D reconstruction from sparse 2D image data using modeling data of tooth structure. The aforementioned Kalke application describes another method of tooth image reconstruction using a frequency transform. Other reconstruction methods for 3D imaging may be used instead.

[0133] Among its advantages, a volumetric image can be formed depending on how much component image data is available to display image slices from different angles. If a sufficient number of component 2D projection images are acquired at different relative angles, a resulting volumetric image can be formed and displayed from multiple viewing angles, thereby assisting the dentist in making a more accurate diagnostic evaluation of a tooth or other structure.

[0134] Presentation / GUI The operator may have the ability to set the operational mode of the imaging system using operator interface commands.

[0135] According to an exemplary embodiment according to the present application, the operator interface on the display 28 (FIG. 1) can show any subset of the projection images acquired as part of a tomosynthesis series, as well as any suitable tomosynthesis slices from the reconstructed image data. FIG. 23 shows the display 28 with a comparative display, showing a 2D projection image 68 along with a corresponding tomosynthesis slice 94.

[0136] Geometric calibration using spectral X-ray imaging As mentioned above, geometric calibration used in combination with alignment methods and apparatus helps provide improved accuracy and resolution for tomosynthesis reconstruction. The embodiments of the present disclosure provide a number of solutions for geometric calibration that take advantage of the properties of spectral x-ray imaging. The methods of geometric calibration described herein may be suitable for chairside dental tomosynthesis and other radiological 2D and 3D volumetric imaging applications, especially where the detector is not mechanically coupled to the source and may not be visible during setup and imaging.

[0137] Geometric calibration calculations for volumetric imaging devices are well known to those skilled in the art of volumetric image reconstruction. Embodiments of the present disclosure are directed to identifying the relative spatial coordinates of the x-ray source and detector, and given this baseline data, the required spatial coordinate calculations can be performed with sufficient accuracy for tomosynthesis or tomographic reconstruction.

[0138] Although existing methods for geometric calibration may provide accurate data regarding the spatial positioning of the source relative to the detector, there is still some need for improvement. The use of coupled mechanical devices for intraoral imaging may be cumbersome and uncomfortable for patients. When markers are used for geometric calibration, a conflict may arise between the need for an unobstructed view of the image data and the need to clearly and unambiguously detect the markers and distinguish them from other image content to derive the necessary geometric positioning or calibration data. In practice, some of the image content used for diagnosis may become unclear, reducing the diagnostic quality, or one or more position markers may be difficult to detect, leading to erroneous or ambiguous measurements.

[0139] According to an alternative exemplary embodiment of the present disclosure, geometric calibration is provided using spectral x-ray detection. As previously mentioned, two types of x-ray imaging architectures provide spectral information content, and spectral x-ray systems, i.e. (i) one or more dual energy source x-ray machines; and (ii) An x-ray device having a photon counting detector, either direct or indirect.

[0140] For geometric calibration using spectral x-ray imaging, markers of materials having characteristic attenuation responses to x-rays of different wavelengths can be placed in the phantom at known positions relative to each other and positioned between the x-ray source and the detector. The spectral imaging content can be acquired from a conventional detector with a dual energy source as in (i) above, or from a photon counting detector using a polychromatic x-ray source as in (ii) above. This spectral imaging content can be processed to generate two image types: (a) an anatomical image that can be used for diagnostic / clinical purposes, either a 3D tomographic image, a tomosynthesis (partial 3D) volume image, or a (2D) projection image, in which the phantom markers are not discernible or are barely visible; and (b) a marker image, i.e., a marker image that clearly and unambiguously highlights or emphasizes the location of the phantom markers, but has little or no anatomical image content.

[0141] Geometric calibration using spectral x-ray imaging starts with some initially known positional or spatial relationships between markers in a phantom. Markers from the phantom seen in the processed marker image ((b) above) can be used to calculate the remaining unknown spatial relationship data between the source and detector, and the overall coordinate system used for volume reconstruction.

[0142] Thus, for a calibration setup, the phantom markers are typically fixed in place relative to either the detector or the source. The schematic in Fig. 24 shows an arrangement with the markers fixed in place relative to the detector. The markers 232 in the phantom 230 are placed in a frame 78 of a holder 72 with a bite block 254. The holder 72 is used to align the source with the detector 20. Alternative arrangements could include the phantom being bonded to a tooth or other imaged object.

[0143] FIG. 25 is a schematic diagram showing an alternative arrangement in which the markers 232 of the phantom 230 are fixed to the source, for example located within the x-ray hood 250.

[0144] It should be noted that the markers have well-defined coordinate positions relative to each other, regardless of whether their positions relative to the source or detector are known.

[0145] The disclosed embodiments address the need for geometric calibration of the x-ray energy distribution using markers with corresponding processing that allows for reduced impact on image quality of imaged anatomy and the ability to clearly indicate the location of the markers. The exemplary graphs in Figures 26A-26C show the spectral relationships used to create the x-ray energy distribution, which can be combined and processed in different ways to produce the following: (i) marker content can be isolated and enhanced to be more clearly defined for geometric calibration purposes; (ii) This in turn can be achieved by making the marker content less visible or imperceptible to the observer, thereby enabling diagnostic use of the acquired image content.

[0146] To achieve these opposing effects, embodiments of the present disclosure create two different x-ray energy distributions, where each distribution is a product of the number of photons or x-rays generated and the mass attenuation properties of the marker and anatomical material. Subsequent processing of these two different x-ray energy distributions enhances or suppresses the marker content using the sequences described below.

[0147] FIG. 26A shows the energy profile of a dual energy x-ray delivery, where two overlapping spectral wavelength bands of x-ray energy are generated sequentially and directed at the phantom 230 and the anatomical structure of the subject. Two x-ray images made in rapid succession capture the subject in essentially the same position. FIG. 26A shows a lower energy profile curve 240 and a higher energy profile curve 242, for example, a dual energy delivery. Each profile curve can be thought of as displaying the amount of photons emitted across different energy levels (keV) or spectral values, using an x-ray tube at the two different kVp technical settings shown.

[0148] FIG. 26B shows, in somewhat exaggerated form for clarity, the characteristic mass attenuation coefficient μ for different tissue types, bone and soft tissue materials, as well as for one possible type of marker material, e.g., an iodine marker element. The x-ray absorption characteristics of a material provide a kind of "signature" indicative of that material in the imaged subject. It is instructive to observe that there can be a pronounced discontinuity in the slope of the attenuation curve, known as the k-absorption edge 260. The k-absorption edge 260 of iodine is prominent in the range of 30-40 keV (wavelength range of 41-31 nm). Other materials with k-absorption edges in the keV range used for most dental applications include gadolinium, which has a k-absorption edge at 50.2 kV, platinum, which has a k-absorption edge at 78.4 kV, lead, which has a k-absorption edge at 88.0 kV, gold, which has a k-absorption edge at 80.7 kV, and silver, which has a k-absorption edge at 25.5 kV. The difference in attenuation between the two materials is most pronounced at or near the k-absorption edge 260 of one of the materials, as shown.

[0149] To a first approximation, the x-ray energy distribution is determined by the effective mass attenuation μ of the material, e.g. bone in this example. eff The effective mass damping of the material is expressed as μ eff may be viewed as the sum or integral over a portion of the energy spectrum E of the product of the mass attenuation μ(E) curve of FIG. 26B and the spectral probability distribution p(E) of x-ray energies as shown in the curve of FIG. 26A.

number

[0150] For each material type (fat tissue, bone, marker), different x-ray energy distributions are determined by the effective mass attenuation μ eff and is applicable for lower and higher energy emissions. For example, for soft tissue, the values ​​of the low energy (L) and high energy (H) emitted energy distributions can be related as follows:

number

number

number

[0151] Thus, by appropriate selection of the spectral content and by comparison of the difference in effective attenuation coefficients with the spectral characteristics, material components of the target image can be distinguished from each other according to their respective x-ray energy distributions under given conditions. The embodiments of the present disclosure use these differences on a pixel basis to selectively suppress or enhance the marker content of the acquired images. For example, when the difference in energy distribution is measured by subtraction or other combinations such as using weighting, the spectral characteristics can be used to enhance the detection of materials. Prior knowledge of the marker material allows suppression of the marker image, for example in areas where the marker obscures anatomical structures.

[0152] The mass attenuation coefficient relationship shown in Figure 26B can also be used for monoenergetic emissions when using photon counting detectors. Figure 26C shows a characteristic distribution of spectral energies with a threshold T used as a boundary value to distinguish between low energy photons (values ​​below the threshold T) and high energy photons (values ​​above the threshold T).

[0153] As previously mentioned, the ability to count photons above and below a defined energy threshold T allows the detector 20 to distinguish between energy distributions resulting from irradiating an object depending on the material, providing an additional dimension to the image data provided as a result of each exposure. This capability has been described as multispectral or "color" x-ray imaging, and allows for the retrieval of information regarding the material composition of each object pixel.

[0154] As the attenuation curves of Figure 26 show, there may be certain wavelengths (and correspondingly certain keV values) where the attenuation coefficients of two different materials are sufficiently different from each other that the materials can be identified from the x-ray image according to the difference in distribution due to the measured attenuation coefficients. Furthermore, if these different energy distributions of photons can be distinguished from each other, it becomes possible to perform material decomposition, thereby identifying one or both materials within the same pixel or voxel image element of the resulting image. This same basic attenuation behavior in response to radiation also allows some ability to distinguish between tissue types, for example.

[0155] As mentioned in the Background paragraph above, the use of markers facilitates accurate 3D reconstruction. Markers placed in the path of the x-ray beam may be imaged along with the imaged anatomy, providing positional information suitable for geometric calibration purposes. However, calibration markers are to some extent radiopaque and may easily obscure parts of the imaged anatomy, compromising the diagnostic value of the acquired images. This aspect of reducing image quality may be particularly troublesome for practitioners who wish to use both 3D volumetric tomosynthesis reconstructions and 2D projection images acquired as part of a tomosynthesis series. 3D reconstruction algorithms may be able to compensate to some extent for the effect of the markers, reducing their effect on the volumetric reconstruction. On the other hand, the 2D projection images may clearly show the radiopaque markers, blocking the display of parts of the imaged anatomy that are the region of interest. The use of markers with low attenuation coefficients may be an insufficient solution, as the markers themselves may not be visible in parts of the imaged anatomy.

[0156] Exemplary embodiments of the present disclosure use spectral x-ray imaging to address the problem of geometric calibration for tomosynthesis imaging using markers. Exemplary embodiments of the present disclosure may use a phantom having a set of markers 232, as previously described in Figures 24 and 25.

[0157] Preferably, the marker is positioned within the imaging area of ​​the detector, at or near the center of the imaging area, or symmetrically positioned about the center of the imaging area, in which the marker is placed within the imaging area of ​​each 2D projection image.

[0158] The marker 232 may be formed from any of a number of materials suitable for intra-oral use. The marker material may have a lower attenuation coefficient μ than conventional markers. mThe markers 232 may have a diameter of 0.5 mm. Non-toxic materials that may be suitable for intraoral imaging may include, for example, chrome steel, ceramic, tungsten carbide, and gold. The markers 232 may be spherical. A suitable size range for tomosynthesis imaging is 0.5 mm in diameter. Other sizes and shapes may be used as desired.

[0159] Other materials may be used that are radiopaque, particularly those with atomic numbers below 45, according to exemplary embodiments of the present disclosure. Examples of materials of this type include chrome steel (atomic number 26) and ZrO2 ceramic material (atomic number 40). Materials with opacity in this range tend to produce imaging artifacts of correspondingly lower intensity than materials that exhibit higher attenuation to radiation.

[0160] The marker 232 may be iodine or other material having a characteristic k-absorption edge 260 in the x-ray energy range, as shown in the simplified graph of FIG. 26B. As previously mentioned, there may be a noticeable discontinuity in the slope of the attenuation curve of the k-absorption edge 260. The k-absorption edge 260 of iodine is prominent in the range of 30-40 keV (wavelength range of 41-31 nm). Other materials with k-absorption edges in the keV range that are commonly used for dental applications may include gadolinium, platinum, lead, gold, and silver.

[0161] Compared to conventional marker approaches using spherical BB or particles of other highly attenuating materials, the marker material used in Applicant's process may have reduced attenuation, resulting in less impact on the anatomical structures imaged in the 2D projection images. In addition, the exemplary methods described herein may process the acquired images to enhance the content of the markers 232, reduce ambiguous or confusing image content, and more clearly indicate the marker locations.

[0162] The markers appear in each 2D projection image of the tomosynthesis series. Given a well-defined marker placement on the detector and the known source path of the incident radiation for tomosynthesis, the acquired 2D projection images may make it possible to geometrically calibrate the imaging system for accurate reconstruction.

[0163] Image processing techniques can be applied to compensate for the appearance of and disturbances caused by the markers 232. Well-known techniques such as inpainting and interpolation can be used to suppress or remove the marker(s) from one or more projected images, thereby helping to minimize or eliminate, for example, metal artifacts.

[0164] Thus, according to an exemplary embodiment of the present disclosure, a radiation volumetric imaging device is provided, comprising an x-ray source and a detector arranged to define an irradiation path through an object and a calibration phantom, and configured to generate 2D projection image data having at least first and second x-ray spectral energy distributions, where the calibration phantom has a plurality of radiopaque markers formed of a marker material. The imaging device further comprises a control logic processor configured to execute programmed instructions to (i) acquire 2D projection image data from the detector, (ii) calculate source and detector geometries of the imaging device corresponding to the acquired 2D projection image data of the calibration phantom of the first and second x-ray energy distributions, and (iii) reconstruct a 3D volumetric image of the object according to the acquired anatomical image data from the object and the source and detector geometries in the plurality of 2D projection images. A display is in signal communication with the control logic processor and configured to display one or more portions of the reconstructed 3D volumetric image.

[0165] An exemplary method herein employs the principles described with reference to Figures 26A-C to use spectral imaging for geometric calibration of a tomosynthesis or other volumetric imaging system. The logic flow diagram of Figure 27 illustrates a sequence for providing both marker and anatomical content from 2D projection image content obtained using a spectral x-ray imaging system with either a dual energy source or a photon counting detector, as described above.

[0166] In the acquisition step S100, a spectral x-ray projection image is acquired. As shown with reference to Figures 26A, 26B, and 26C, the spectral x-ray projection image can be acquired from either a dual energy source using a standard x-ray detector, or a polychromatic x-ray source using a photon counting detector. Thus, the spectral image content is divided according to the spectral content (photon energy, in keV).

[0167] At least for photon counting detectors, the spectral image data is collected for a low spectral band or subset SS whose keV energies are below a predetermined threshold T. L or a high spectral band or subset SS whose keV energy exceeds a threshold T H Various combinations of low and high spectral responses may accentuate the content of the marker 232 or reduce or eliminate the content of the marker.

[0168] The threshold x-ray energy value T is determined as the boundary between a lower and a higher spectral band. In practice, the threshold is obtained in keV values ​​(or alternatively considered in wavelength nm values) and is related to the attenuation coefficient value μ of the material of the marker 232. m and the attenuation coefficient value of the imaged anatomical structure can be easily measured. A threshold T shown in FIG. 26C is taken along or near the k-edge 260 and may be appropriate to split between the low and high bands, since the value T or values ​​near it may differ significantly for the markers and the anatomical structures.

[0169] Continuing with the sequence of FIG. 27, the material decomposition step S110 then decomposes the subset SS of the spectral image data. L , S.S. H to create two different kinds of image content: (i) Marker material image content 300 that can be processed to clearly and unambiguously show marker placement. The marker material image content 300 utilizes the attenuation properties of the marker material to separate the marker content 300 from background anatomical structures. The marker material image content 300 may contain little or no useful anatomical content, but is primarily concerned with the spatial location of the marker material(s) within the image content. With further processing, unambiguous marker locations can be detected, allowing easy identification of each marker within the image frame in the subsequent marker detection step S120. Given a marker image that clearly shows the relative spatial location of the phantom markers 232 with known phantom locations relative to either the source or detector, and their location in the image plane, there is sufficient information for calculations to provide a geometric calibration, using geometric methods well known to those skilled in the art. (ii) A background image 320 showing the imaged anatomical structures with sufficient contrast and resolution for diagnostic functions.

[0170] 28A-28C show in simulated form the results of image processing for marker 232 detection in 2D projection images for tomosynthesis imaging according to an exemplary embodiment of the present disclosure. FIG. 28A shows an acquired 2D projection image with a set of conventional markers 232 for geometric calibration. In this example, the markers are spherical or BB-type markers 232 and are arranged along two planes extending perpendicular to the beam direction. The markers 232 in each plane may be differentiated by shape, density, or other factors. The relative shift between the positions of the markers 232 in the image corresponds to the relative spatial positions of the source and detector. It can be easily seen that this conventional marker image content blocks a portion of the diagnostic image content.

[0171] FIG. 28B shows an exemplary marker image 310. This image may or may not be created, however, FIG. 28B shows how the methods described herein may be used to highlight the markers 232. Optional labels may be provided to identify individual markers 232, such as labels 1A1-1C1, 2A1-2A4 in the example shown. The marker image 310 is obtained by a process such as the marker detection step S120 of FIG. 27. When using spectral x-ray data, the difference between the low and high spectral data, with appropriate threshold selection, is useful for calculating the location of the marker 232 in the acquired data. Thus, referring to the sequence of FIG. 27, the processing method for forming the marker image 310 typically includes, in a material decomposition step S110, the low spectral projection SS L and high spectrum projection SS H Take the weighted difference between

[0172] Figure 28C shows an example background image 320 corresponding to the marker image 310 given in Figure 28B. To an observer, the markers may be imperceptible or barely perceptible against the anatomical content shown in the background image. A material decomposition step S110 may provide additional processing, which subtracts or otherwise removes the marker image 310 from the complete image obtained from the combined low and high spectral projections. This may effectively remove the marker content from the background material image 320.

[0173] Following the sequence outlined in Figure 27, geometric calibration data is created for each of the 2D projection images of the tomosynthesis series. Subsequent reconstruction processing may use a set of background material images along with associated calibration marker data to create 3D volumetric image content.

[0174] The invention has been described in detail with particular reference to exemplary embodiments thereof as currently understood, but it should be understood that changes and modifications may be effected within the spirit and scope of the invention.

[0175] For example, control logic processor 26 may be any of several types of logic processing devices, including a computer or computer workstation, a dedicated host processor, a microprocessor, a logic array, or other device that executes stored program logic instructions.

[0176] The presently disclosed exemplary embodiments are therefore considered in all respects to be illustrative and not restrictive The scope of the invention is indicated by the appended claims, and all changes that come within the meaning and range of equivalents thereof are intended to be embraced therein.

[0177] Consistent with at least one exemplary embodiment, the exemplary method / apparatus may employ a computer program with stored instructions that executes on image data accessed from an electronic memory. As can be appreciated by those skilled in the image processing art, the computer programs of the exemplary embodiments of the present invention may be utilized by any suitable general-purpose computer system, such as a personal computer or a workstation. However, many other types of computer systems may be used to execute the computer programs of the disclosed exemplary embodiments, including, for example, single or network processor arrangements.

[0178] The computer program for carrying out the method of the specific exemplary embodiment disclosed herein may be stored in a computer readable storage medium. This medium may include, for example, a magnetic storage medium such as a magnetic disk or magnetic tape such as a hard drive or a removable device, an optical storage medium such as an optical disk, optical tape, or a machine-readable optical encoding, a solid-state electronic storage device such as a random access memory (RAM) or a read-only memory (ROM), or any other physical device or medium used to store a computer program. The computer program for carrying out the method of the described exemplary embodiment may also be stored on a computer readable storage medium, which is connected to the image processor by means of the Internet, other networks, or communication media. Those skilled in the art will further readily recognize that the equivalent of such a computer program product may also be constructed in hardware.

[0179] It should be noted that the term "memory" is equivalent to "computer-accessible memory" within the context of the application and can refer to any type of temporary or more permanent data storage workspace that is used to store and operate on image data and is accessible to a computer system, including, for example, a database. Memory can be non-volatile, for example, using long-term storage media such as magnetic or optical storage. Alternatively, memory can be more volatile, using random access memory (RAM) used as a temporary buffer or workspace by electronic circuits, for example, a microprocessor or other control logic processor device. Display data is typically stored, for example, in a temporary storage buffer directly associated with the display device, and is periodically updated as needed to provide the displayed data. This temporary storage buffer can also be considered a memory as the term is used in this application. Memory can also be used as a data workspace for performing and storing intermediate and final results of calculations and other processing. Computer-accessible memory can be volatile, non-volatile, or a hybrid combination of volatile and non-volatile.

[0180] It will be understood that the computer program products of the exemplary embodiments herein may utilize various image manipulation algorithms and / or processes that are well known. It will be further understood that the exemplary computer program product embodiments herein may embody algorithms and / or processes that are useful for implementation, but not specifically shown or described herein. Such algorithms and processes may include conventional utilities that are within the ordinary skill of the image processing art. Additional aspects of such algorithms and systems, and hardware and / or software for creating and otherwise processing images, or for interfacing with the computer program products of the present application, are not specifically shown or described herein, and may be selected from algorithms, systems, hardware, components, and elements well known in the art.

[0181] Exemplary embodiments according to the present application may include various features described herein (either individually or in any combination).

[0182] While the present invention has been illustrated with respect to one or more implementations, substitutions and / or modifications may be made to the illustrated examples without departing from the spirit and scope of the appended claims. In addition, while a specific feature of the present invention may be disclosed with respect to only one of some implementations / exemplary embodiments, such feature may be combined with one or more other features of other implementations / exemplary embodiments as may be desirable and advantageous for any given or specific function. The term "a" or "at least one" is used to mean that one or more of the listed items may be selected. The term "about" indicates that the listed value may be somewhat varied without causing incompatibility in processing or construction with respect to the exemplary embodiment shown. Other embodiments of the present invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. It is intended that the specification and illustrated embodiments be considered as exemplary only, with the true scope and spirit of the invention being indicated by the following claims.

Claims

1. 1. A method for geometric calibration of a volumetric imaging device, performed by said volumetric imaging device, comprising: (a) placing a calibration phantom in an irradiation path including an object disposed between an x-ray source and a detector, the calibration phantom having a plurality of radiopaque markers formed of a marker material; (b) in a repeating sequence, at each of a plurality of relative positions of the x-ray source and the detector, (i) acquiring 2D projection image data for the subject and the calibration phantom, the 2D projection image data distinguishing at least first and second x-ray energy distributions; (ii) calculating an x-ray source-detector geometry of the volumetric imaging device corresponding to the 2D projection image data acquired for the first and second x-ray energy distributions; and (c) reconstructing a 3D volumetric image of the object according to anatomical image data acquired from the object and the calculated x-ray source-detector geometry within the plurality of 2D projection images; (iii) the first and second x-ray energy distributions have different spectral bands. method.

2. The method of claim 1 , further comprising forming at least one 2D anatomical image of the object according to the combined data of the first and second x-ray energy distributions.

3. The method of claim 2 , wherein forming the at least one 2D anatomical image of the object further comprises suppressing radiopaque marker content from the 2D anatomical image.

4. The method of claim 1 , wherein reconstructing the 3D volumetric image further comprises suppressing radiopaque marker content from the 2D projection images.

5. The method of claim 1 , further comprising forming a marker content augmented image according to a material decomposition of the 2D projection image data.

6. The method of claim 1 , wherein acquiring the 2D projection data comprises exposing the object using a photon counting detector.

7. 1. A method for geometric calibration of radiological imaging, performed by a volumetric imaging device, comprising: (a) placing a calibration phantom in an irradiation path including an object disposed between an x-ray source and a detector, the calibration phantom having a plurality of radiopaque markers formed of a marker material; (b) acquiring 2D projection image data of the object and the calibration phantom, where the 2D projection image data distinguishes between at least first and second x-ray energy distributions; (c) identifying a relative spatial location of the radiopaque marker within the 2D projection image data based on at least the first and second x-ray energy distributions; and (d) performing a geometric calibration using the identified spatial locations of the radiopaque markers; and (e) forming a 2D projection image showing the subject's anatomical structure and suppressing the radiopaque markers; method.

8. The method of claim 7 , wherein forming the 2D projection image further comprises using weighted subtraction.

9. The method of claim 7 , wherein acquiring the 2D projection image data comprises acquiring the 2D projection image data of the object using a photon counting detector.

10. The method of claim 1 or 7, wherein acquiring the 2D projection data comprises exposing the object using a dual energy x-ray source.

11. The method of claim 1 or 7, wherein the marker material is taken from the group including iodine, gadolinium, platinum, lead, gold, and silver.

12. The method of claim 1 or 7, wherein the calibration phantom is coupled to the x-ray source or the detector.

13. The method of claim 1 , wherein the volumetric imager is a tomosynthesis imager.

14. (a) an x-ray source and a detector arranged to define an irradiation path through an object and a calibration phantom, wherein the x-ray source and the detector are configured to produce 2D projection image data having at least first and second x-ray spectral energy distributions, wherein the calibration phantom has a plurality of radiopaque markers formed of a marker material; (b) a control logic processor configured to execute programmed instructions, (i) acquiring the 2D projection image data from the detector; (ii) calculating an x-ray source-detector geometry of the volumetric imaging device corresponding to the acquired 2D projection image data of the calibration phantom for the first and second x-ray energy distributions; (iii) reconstructing a 3D volumetric image of the object according to anatomical image data acquired from the object and the calculated x-ray source-detector geometry; a control logic processor; Equipped with the first and second x-ray energy distributions have different spectral bands; Radiography volume imaging device.

15. 15. The method of claim 1 or the radiographic volume imaging apparatus of claim 14, wherein the marker material has a k-absorption edge within an energy range of either the first or second x-ray spectral energy distributions.

16. The method of claim 1, wherein the x-ray energy distribution is determined from the product of the spectral probability distribution of x-ray energy and the mass attenuation characteristics of the radiopaque marker and anatomical material.

17. The radiography volume imaging device of claim 14, wherein the x-ray energy distribution is determined from the product of the spectral probability distribution of x-ray energy and the mass attenuation characteristics of the radiopaque marker and anatomical material.

Images