WO2017079654A1 - Rotating-slit gamma-ray imager and associated imaging method - Google Patents

Rotating-slit gamma-ray imager and associated imaging method Download PDF

Info

Publication number
WO2017079654A1
WO2017079654A1 PCT/US2016/060675 US2016060675W WO2017079654A1 WO 2017079654 A1 WO2017079654 A1 WO 2017079654A1 US 2016060675 W US2016060675 W US 2016060675W WO 2017079654 A1 WO2017079654 A1 WO 2017079654A1
Authority
WO
WIPO (PCT)
Prior art keywords
slit
trial
images
sinogram
detector
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/US2016/060675
Other languages
French (fr)
Inventor
Lars R. Furenlid
Xin Li
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Arizona
Arizona's Public Universities
Original Assignee
University of Arizona
Arizona's Public Universities
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of Arizona, Arizona's Public Universities filed Critical University of Arizona
Priority to US15/773,691 priority Critical patent/US10548544B2/en
Publication of WO2017079654A1 publication Critical patent/WO2017079654A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/42—Arrangements for detecting radiation specially adapted for radiation diagnosis
    • A61B6/4208—Arrangements for detecting radiation specially adapted for radiation diagnosis characterised by using a particular type of detector
    • A61B6/4258—Arrangements for detecting radiation specially adapted for radiation diagnosis characterised by using a particular type of detector for detecting non x-ray radiation, e.g. gamma radiation
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
    • A61B6/03—Computed tomography [CT]
    • A61B6/037—Emission tomography
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/06—Diaphragms
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/42—Arrangements for detecting radiation specially adapted for radiation diagnosis
    • A61B6/4275—Arrangements for detecting radiation specially adapted for radiation diagnosis using a detector unit almost surrounding the patient, e.g. more than 180°
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/42—Arrangements for detecting radiation specially adapted for radiation diagnosis
    • A61B6/4291—Arrangements for detecting radiation specially adapted for radiation diagnosis the detector being combined with a grid or grating
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
    • A61B6/5205—Devices using data or image processing specially adapted for radiation diagnosis involving processing of raw data to produce diagnostic data
    • G—PHYSICS
    • G02—OPTICS
    • G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
    • G02B6/00—Light guides; Structural details of arrangements comprising light guides and other optical elements, e.g. couplings
    • G—PHYSICS
    • G02—OPTICS
    • G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
    • G02B6/00—Light guides; Structural details of arrangements comprising light guides and other optical elements, e.g. couplings
    • G02B6/24—Coupling light guides
    • G02B6/42—Coupling light guides with opto-electronic elements
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T12/00—Tomographic reconstruction from projections
    • G06T12/20—Inverse problem, i.e. transformations from projection space into object space
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T12/00—Tomographic reconstruction from projections
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2211/00—Image generation
    • G06T2211/40—Computed tomography
    • G06T2211/424—Iterative

Definitions

  • SPECT Single photon emission computed tomography
  • a rotating slit is proposed that addresses at least the
  • a method for forming an optimized image of a subject includes steps of acquiring a plurality of one-dimensional images, generating a measured sinogram from the plurality of one-dimensional images, and determining a plurality of trial images.
  • the method acquires a plurality of one-dimensional images of the subject captured by a rotating-slit imager having (a) a detector, and (b) a slit collimator having a slit oriented at one of a respective plurality of slit-rotation angles, relative to the subject, about a longitudinal axis substantially perpendicular to a front surface of the detector.
  • the method iteratively determines a plurality of trial images each having a respective trial sinogram.
  • the optimized image is one of the plurality of trial images and its corresponding trial sinogram differs from the measured sinogram by less than a predetermined tolerance.
  • a rotating-slit gamma-ray imager includes (i) a detector having an array of pixels, (ii) a slit collimator, (iii) a memory storing non-transitory computer-readable instructions, and (iv) a microprocessor.
  • the slit collimator has a slit at least partially aligned with the detector and configured to be oriented at any of a plurality of slit-rotation angles in a plane substantially parallel to the array of pixels.
  • the microprocessor is adapted to execute the instructions to form an optimized image of a subject by implementing the method of the first embodiment.
  • a data processor includes a memory storing non-transitory computer-readable instructions and a microprocessor.
  • microprocessor is configured to execute the instructions to (i) generate a measured sinogram from a plurality of one-dimensional images, of a subject, captured by a rotating-slit imager having (a) a detector, and (b) a slit collimator having a slit oriented at one of a respective plurality of slit-rotation angles, relative to the subject, about a longitudinal axis substantially perpendicular to a front surface of the detector, and (ii) generate an optimized image by iteratively determining a plurality of trial images each having a respective trial sinogram, the optimized image being one of the trial images, its corresponding trial sinogram differing from the measured sinogram by less than a predetermined tolerance.
  • FIG. 1 is a schematic diagram of a rotating-slit gamma-ray imager, in an embodiment.
  • FIG. 2 illustrates an example of the imaging assembly of the rotating-slit gamma-ray imager of FIG. 1.
  • FIG. 3 is a transverse cross-sectional view of a digital phantom used to simulate operation of the rotating-slit gamma-ray imager of FIG. 1, in an embodiment.
  • FIG. 4 illustrates a selected plurality of one-dimensional images of the digital phantom of FIG. 3 captured by the imaging assembly of FIG. 2.
  • FIG. 5 shows an exemplary sinogram (left), its inverse-Radon- transform image (middle), and a Maximum-Likelihood Expectation Maximization (MLEM) reconstructed image (right).
  • MLEM Maximum-Likelihood Expectation Maximization
  • FIG. 6 is a comparison between physical pinhole aperture images and exemplary synthesized virtual pinhole images, the latter formed by an
  • FIG. 7 is a perspective view of a second example of the imaging assembly of the rotating-slit gamma-ray imager of FIG. 1.
  • FIG. 8 is a perspective view of twelve imaging assemblies of FIG. 7 arranged to have a common overlapping field of view, in an embodiment.
  • FIG. 9 is a flowchart illustrating a method for forming an optimized image, in an embodiment.
  • FIG. 10 shows an experimental setup for a demonstration of the rotating-slit gamma-ray imager of FIG. 1, in an embodiment.
  • FIG. 11 is a photograph of a phantom that consists of six iodine- 125 seeds.
  • FIG. 12 shows exemplary projection images from physical-pinhole and pinhole-synthesis collimation, the latter formed by an embodiment of the rotating- slit gamma-ray imager of FIG. 1.
  • FIG. 13 illustrates a comparison of image acquisition via an embodiment of the rotating-slit gamma-ray imager of FIG. 1 and a pinhole gamma- ray imager.
  • FIG. 14 is a perspective view of a third example of the imaging assembly of the rotating-slit gamma-ray imager of FIG. 1.
  • FIG. 15 is a perspective view of a fourth example of the imaging assembly of the rotating-slit gamma-ray imager of FIG. 1.
  • Rotating-slit pinhole-aperture synthesis yields a much higher photon count image than a physical pinhole, given the same data acquisition time.
  • Slit imaging with a collimator only requires a one-dimensional detector if it is made to rotate with the collimator.
  • the sensitivity advantage becomes very significant as pinhole radius R decreases, because the slit sensitivity scales as 2R, whereas the pinhole sensitivity scales as R 2 .
  • FIG. 1 is a schematic diagram of an exemplary rotating-slit gamma-ray imager 100.
  • Rotating-slit gamma-ray imager 100 includes one or more imaging assemblies 120.
  • Each imaging assembly 120 includes a slit collimator 122 and a detector 124.
  • Slit collimator 122 includes a slit 121 oriented at a rotation angle 127.
  • Imager 100 also includes a data processor 130 communicatively coupled to each detector 124.
  • an element in the figures denoted by a reference numeral suffixed by a parenthetical numeral is an example of the element indicated by the reference numeral.
  • imaging assembly 120(2) is an example (2) of imaging assembly 120.
  • Rotating-slit gamma-ray imager 100 may also include additional imaging assemblies 120(2) to 120(N), where integer N ⁇ 2.
  • Each imaging assembly 120 may have a respective viewing angle 128 of a common imaging volume in the field of view of each imaging assembly 120.
  • Detector 124 is configured to capture a plurality of one- dimensional images 129.
  • one-dimensional images 129 from a single imaging assembly 120 include M images indexed by an integer m.
  • Index m E ⁇ 1, 2, ... , M) denotes rotation angle 127(m) of slit 121 about an axis substantially perpendicular to at least a portion of detector 124.
  • Integer M is, for example, between sixty and one-hundred eighty, which corresponds to three degrees per angular step and one degree per angular step, respectively.
  • Detector 124 may corotate with slit collimator 122.
  • Rotation angle 127 may be a relative angle between slit collimator 122 and an object being imaged. Accordingly, rotation angle 127 may be changed by rotating the object being imaged with respect to slit collimator 122, or vice versa.
  • reference to a changing rotation angle 127 corresponds to rotation of at least one of (a) slit collimator 122 and (b) the object being imaged, where rotations occur about an axis substantially perpendicular to at least a portion of detector 124.
  • Data processor 130 includes a memory 132, microprocessor 134, and software 140.
  • Memory 132 may be transitory and/or non-transitory and may represent one or both of volatile memory (e.g., SRAM, DRAM, other volatile memory, or any combination thereof) and non-volatile memory (e.g., FLASH, ROM, magnetic media, optical media, other non-volatile memory, or any combination thereof).
  • Software 140 includes machine-readable instructions.
  • Microprocessor 134 is adapted to execute the instructions to perform functions of rotating-slit gamma-ray imager 100 as described herein.
  • Microprocessor 134 may be a digital signal processor such as an image processor.
  • Software 140 includes a sinogram generator 141, an image reconstructor 142, and optionally a volume-image reconstructor 143.
  • Sinogram generator 141 generates a sinogram 151(1) from one-dimensional images 129(1-M) and optionally generates sinograms 151 (2-N) from respective one-dimensional images 129(1-M, 2-N).
  • Image reconstructor 142 generates a reconstructed image 152(1) from sinogram 151 (1) and optionally generates respective reconstructed images 152(2-N) from sinograms 151 (2-N).
  • Image reconstructor 142 may implement an iterative algorithm.
  • the iterative algorithm may include at least one of an expectation-maximization algorithm, a maximum likelihood estimation method, maximum a posteriori reconstruction, an algebraic reconstruction technique, a Landweber reconstruction algorithm, and a stochastic optimization method.
  • stochastic optimization method include simulated annealing, evolutionary algorithms, and swarm algorithms.
  • a sinogram of reconstructed image 152(1) differs from sinogram 151(1) by less than a predetermined tolerance.
  • the predetermined tolerance functions as a convergence criterion for the iterative algorithm.
  • image reconstructor 142 implements an iterative algorithm that minimizes a merit function, such as an Z ⁇ -norm or an L 2 -norm, or maximizes a merit function such as likelihood function.
  • Optional volume-image reconstructor 143 generates a
  • Volume-image reconstructor 143 implements a tomographic construction algorithm, for example.
  • the tomographic construction algorithm may include at least one of a filtered back projection algorithm, MLEM, an algebraic reconstruction technique, a Landweber reconstruction algorithm, and a maximum a posteriori reconstruction technique.
  • Data processor 130 may also include a hardware controller 136 that controls the angular orientation of slit collimator 122.
  • rotating-slit gamma-ray imager 100 does not include data processor 130, but is communicatively coupled with data processor 130 remotely located from rotating-slit gamma-ray imager 100.
  • rotating-slit gamma-ray imager 100 and data processor 130 may be communicatively coupled via wireless or wired communication protocols known in the art.
  • data processor 130 may operate as a post-processing unit compatible for processing images generated by one or more types of rotating-slit gamma-ray imagers.
  • FIG. 2 illustrates an imaging assembly 220 imaging a subject 290.
  • Imaging assembly 220 is an example of imaging assembly 120 of FIG. 1.
  • Imaging assembly 220 includes a rotating-slit collimator 222 and a detector 224, which are examples of collimator 122 and detector 124, respectively.
  • Collimator 222 may be formed of a material that blocks gamma rays and includes a slit 221.
  • Slit 221 has a slit width 221W and a slit height 221H in the x and y directions respectively in a coordinate system 298.
  • directions x, y, and z are with reference to coordinate system 298 unless otherwise specified.
  • Detector 224 has a plurality of oblong pixels 226(p) in a plane parallel to the x-y plane. Integer p is a pixel index in the x direction, each pixel 226 has a width 226W. The long dimension of the pixels 226 are aligned with slit 221. For clarity of illustration, not all pixels 226 are labeled in FIG. 2. Pixels 226 may include a semiconductor material suitable for detection of gamma rays, such as cadmium zinc telluride (CdZnTe or CZT).
  • CdZnTe cadmium zinc telluride
  • imaging assembly 220 rotates while subject 290 remains stationary.
  • imaging assembly 220 remains stationary while subject 290 rotates.
  • imaging assembly 220 may be mechanically configured such that each pixel 226 remains parallel to slit 221 during rotation.
  • slit 221 being parallel to pixels 226 is equivalent to slit-collimator 222 being parallel to pixels 226.
  • Rotation axis 202 is substantially perpendicular to a plane that includes pixels 226, for example, a front surface of detector 224.
  • Collimator 222 has a rotation angle 227, which corresponds to the orientation of slit 221 with respect to the x direction.
  • Rotation angle 227 is an example of rotation angle 127 (FIG. 1).
  • slit 221 is oriented parallel to t ey- direction for illustrative purposes only.
  • Both slit width 221W and slit height 221H of slit 221 may be adjustable.
  • An adjustable width enables adaptive imaging, e.g., for multiple resolution requirements.
  • An adjustable height enables control of image artifacts when forming reconstructed volume-image 153.
  • One source of artifacts is stray light, such as electromagnetic radiation reaching detector 224 outside the field of view of imaging assembly 220.
  • rotation of imaging assembly 220 or subject 290 may either have continuous smooth motion or discrete motion, where rotation occurs in steps with momentary pauses therebetween.
  • One-dimensional images 129 may be acquired with uniform sampling, for example, corresponding to uniformly-spaced rotation angles between 0° and 360°.
  • angular sampling may be non-uniform, which enables adaptive angular sampling.
  • Rotating-slit collimator 222 and detector 224 are each in planes substantially parallel to the x-y plane and separated by a distance 215 along the z- direction. Distance 215 may also be adjustable, which also allows for adaptive imaging.
  • Rotation axis 202 intersects the center of slit collimator 222, and is substantially perpendicular to the x-y plane.
  • substantially parallel and substantially perpendicular may include, but is not limited to, within ⁇ 5° of being parallel and perpendicular, respectively.
  • Rotating-slit collimator 222 has a width 222W in the x direction.
  • Detector 224 has a width 224W in the x direction and height 224H in the y direction.
  • widths 222W, 224W and heights 221H, 224H are each 100 mm, slit width 221W is 0.3 mm, width 226W is 0.1 mm, and distance 215 is 100 mm.
  • Detector 224 may be replaced with a two-dimensional array of pixels without departing from the scope hereof. Each pixel of such a two-dimensional array has an aspect ratio of less than two, for example.
  • the two-dimensional pixel array may corotate with rotating-slit collimator 222 such that each row of pixels or column of pixels is parallel to slit 221 for any rotational orientation of slit 221.
  • the 2-D detector may be stationary with respect to a rotational orientation of slit 221 such that pixels of the 2-D detector are grouped according to the angle of slit 221.
  • a line parallel to slit 221 intersects a plurality of pixels, the plurality of pixels forming the group.
  • the contribution of each pixel of the group, to a one-dimensional image 129, depends in part on where the line intersects the pixel, e.g., the pixel's center or corner.
  • detector 224 is also translatable in the x-y plane, for example, in a direction perpendicular to the angular orientation of slit 221.
  • Such lateral movement enables dynamic tracking of a region of interest of subject 290, for example, during rotation of one or both of slit-collimator 222 and detector 224.
  • the amplitude of lateral movement in the x-y plane may be in part determined by a trigonometric function of rotation angle 227. For example, the magnitude is proportional to the sine or cosine of rotation angle 227.
  • FIG. 3 is a transverse cross-sectional view of a digital phantom 300 used to simulate operation of rotating-slit gamma-ray imager 100.
  • Digital phantom 300 includes nine hot (radioactive) rods of three different diameters.
  • Images 429(0, 30, 60, 90, 120, 150) are part of a full-set of ninety of images 429 captured at ninety evenly-spaced rotation angles 6 m E ⁇ 0°, 2°, 176°, 178° ⁇ .
  • a two-dimensional x-y coordinate system 498 is superimposed on each one-dimensional image 429 such that the x axis and y axis are oriented, respectively, perpendicular and parallel to the respective slit rotation angle of each image 429.
  • Each image 429(6 m ) has respective pixel values Pe m ( x > y) shown in FIG. 4 as a gray-scale map.
  • pixel values ⁇ ⁇ ; ⁇ ( ⁇ , y) decrease as a function of distance from slit 221 according to an illumination falloff function that varies with x.
  • imaging assembly 220 is rotated through a set of rotation angles 227 spanning a 180° range, the data from detector 224 can be arranged into a sinogram, with one axis representing the pixel index of 1 -D detector 124, and the other axis representing the rotation angle.
  • forming a sinogram can be accomplished with suitable combinations of pixels.
  • FIG. 5 shows an exemplary sinogram 551 and its corresponding inverse-Radon-transform image 520, and a reconstructed image 552.
  • Sinogram 551 is an example of sinogram 151 and is a sinogram of the ninety of images 429.
  • Inverse- Radon-transform image 520 is generated via filtered back projection.
  • Reconstructed image 552 is an example of reconstructed image 152.
  • Image 552 was generated via sixty iterations of a MLEM reconstruction method. Images 520 and 552 were both generated in-part from the ninety one-dimensional images 429 discussed above regarding FIG. 4.
  • Image 552 has a trial sinogram that differs from sinogram by less than a predetermined tolerance of the MLEM reconstruction method.
  • the trial sinogram is, in part, the result of processing image 552 with a Radon transform and operations that remove effects of the aforementioned illumination falloff function E(x) .
  • FIG. 6 is a comparison between physical pinhole aperture images (left column) and synthesized virtual pinhole images (right column). For each row, the data acquisition time is the same. The second row's acquisition time is ten times that of the first row. The phantom in the third row is immersed in a cube-shape background (20% of the peak radioactivity of the hot rods). The phantom in the fourth row contains cold rods immersed in the background activity.
  • the synthesized virtual pinhole images of FIG. 6 are examples of a two-dimensional reconstructed image 152.
  • FIG. 7 is a perspective view of an imaging assembly 720, which is an example of imaging assembly 120 shown in FIG. 1.
  • Imaging assembly 720 includes a slit collimator 722 and a detector 724, which are examples of slit collimator 122 and detector 124, respectively.
  • Slit collimator 722 includes a slit 721, which is an example of slit 121.
  • Imaging assembly 720 also includes a housing 730 and a base 740. Slit collimator 722 and detector 724 are mechanically attached to housing 730, which is configured to rotate in base 740 such that slit collimator 722 and detector 724 corotate.
  • FIG. 9 is a flowchart illustrating a method 900 for forming an optimized image of a subject.
  • Method 900 may be implemented within one or more aspects of rotating-slit gamma-ray imager 100.
  • method 900 is implemented by microprocessor 134 executing computer-readable instructions of software 140, shown in FIG. 1.
  • step 910 method 900 acquires a plurality of one-dimensional images of a subject with a rotating-slit gamma-ray imager having a detector and a slit collimator.
  • Each one-dimensional image corresponds to a different relative angular orientation of the slit collimator and the subject about a longitudinal axis substantially perpendicular to a front surface of the detector.
  • rotating-slit gamma-ray imager 100 acquires one-dimensional images 429.
  • step 920 method 900 generates a measured sinogram from the plurality of one-dimensional images.
  • sinogram generator 141 of rotating-slit gamma-ray imager 100 generates sinogram 551 of FIG. 5.
  • step 930 method 900 iteratively determines a plurality of trial images each having a respective trial sinogram.
  • the optimized image is one of the trial images and has a corresponding trial sinogram differing from the measured sinogram by less than a predetermined tolerance.
  • image reconstructor 142 of data processor 130 generates a two-dimensional reconstructed image 552.
  • Step 940 is optional.
  • method 900 repeats steps 910, 920, and 930 for each of a plurality of additional angular orientations of the subject relative to either the rotating-slit imager or an additional rotating-slit imager.
  • Step 940 results in a plurality of optimized images.
  • each imaging assembly 720 in the figuration of FIG. 8 generates a respective optimized image of subject 890.
  • Step 950 is optional.
  • method 900 generates a reconstructed volume-image by combining the plurality of optimized images using a three-dimensional tomographic construction algorithm.
  • volume-image reconstructor 143 of data processor 130 generates reconstructed volume-image 153.
  • FIG. 10 shows an experimental setup 1095 for a demonstration of a rotating-slit gamma-ray imager 1000, which is an embodiment of rotating-slit gamma- ray imager 100.
  • Rotating-slit gamma-ray imager 1000 includes an intensified quantum-imaging detector (iQID) camera 1020 that is downward-facing above an adjustable-slit collimator 1022 having a slit 1021.
  • Slit 1021 is an example of slit 121.
  • iQID camera 1020 includes a detector 1024 (not shown) with dimensions 30 ⁇ 40 mm and a resolution of approximately 0.14 mm. Camera 1020 is similar to the iQID camera discussed in "The iQID camera: An ionizing-radiation quantum imaging detector," Nucl. Inst. Meth. A, 767, 146-152, 2014 by Miller et al, which is incorporated herein by reference. Other cameras may be used as camera 1020 without departing from the scope hereof.
  • Adjustable-slit collimator 1022 and detector 1024 are embodiments of adjustable-slit collimator 122 and detector 124, respectively.
  • iQID camera 1020 is 88 mm above adjustable-slit collimator 1022 and a rotation stage 1092 is located 68-mm beneath slit collimator 1022. This imaging configuration results in a magnification of 1.3.
  • Rotation stage 1092 has a rotation axis, not labeled, that is example of rotation axis 202 of FIG. 2.
  • Slit 1021 of adjustable slit collimator 1022 may be set to have a 0.5 -mm slit width.
  • Slit 1021 has a length that is one hundred times its width.
  • a corresponding physical pinhole 1014 (with dimensions 0.5 ⁇ 0.5 mm), may be created by adding a second orthogonal slit with the same width.
  • FIG. 1 1 is a photograph of a phantom 1100 in a holder 1 120.
  • Phantom 1100 includes six iodine-125 ( 125 I) seeds 1 110(1-6) each having a maximum gamma-ray photon energy of approximately 36 keV.
  • FIG. 12 shows projection images 1210, 1220, 1230, and 1240 from physical-pinhole and pinhole-synthesis collimation corresponding to experimental setup 1095 with phantom 1100 on rotation stage 1092.
  • Proj ection images 1220 and 1240 are synthesized virtual pinhole proj ection images corresponding to physical pinhole projection images 1210 and 1230 respectively.
  • Projection images 1210, 1220, 1230, and 1240 result from acquisition times of 100 seconds, 90 seconds, 1000 seconds, and 900 seconds, respectively.
  • Projection images 1210 and 1230 are projection images from physical pinhole 1014 having dimensions 0.5 ⁇ 0.5 mm.
  • Projection images 1220 and 1240 are reconstructed projection images from a synthesized virtual pinhole, that is, a pinhole synthesized from slit-aperture imager 1000, with slit width 1021W equal to 0.5 mm, having captured multiple one-dimensional images 129 of phantom 1100 at different slit rotation angles.
  • Projection images 1220 and 1240 are reconstructed by data processor 130, using an iterative MLEM method that enforces positivity, from a plurality of one-dimensional images 129 corresponding to different angular orientations of phantom 1 100, via rotation of rotation stage 1092.
  • Projection images 1220 and 1240 are each examples of a two-dimensional reconstructed image 152.
  • Projection images 1220 and 1240 have the same resolution as images 1210 and 1230 respectively. That is, the center of rotation of slit 1021, in the generation of projection images 1220 and 1240, corresponds to the location of physical pinhole 1014 used to generate projection images 1210 and 1230, respectively.
  • the total sensitivity of projection images 1220 and 1240 is enhanced— the additional sensitivity being afforded by slit 1021 having a slit width 1021W (0.5- mm in this example) approximately equal to the effective diameter of pinhole 1014— but with a much longer length.
  • pinhole 1014 is square, its equivalent diameter is close to 0.5 mm.
  • Rotating-slit gamma-ray imager 1300 includes imaging assembly 720 of FIG. 7.
  • Traditional pinhole gamma-ray imager 1390 includes a pinhole aperture 1392 and a two-dimensional pixel array 1394.
  • Output image 1340 is acquired directly from traditional pinhole gamma-ray imager 1390.
  • rotating-slit gamma-ray imager 100 generates sinogram 551 , discussed above with respect to FIG. 5.
  • Image 552 is reconstructed from sinogram 551 using sixty iterations of a MLEM
  • FIG. 14 illustrates an imaging assembly 1420 that includes a slit collimator 1410 and an elongated detector 1422.
  • Slit collimator 1410 includes a slit 1412.
  • Elongated detector 1422 is for example two adjacent detectors 1422(1 ,2).
  • Imaging assembly 1420 is an example of imaging assembly 120 of FIG. 1.
  • the relatively increased length of elongated detector 1422 enables an embodiment of rotating-slit gamma-ray imager 100 that includes imaging assembly 1420 to detect more gamma-ray producing events than with a conventional smaller detector, which result in enhanced detection sensitivity.
  • FIG. 15 illustrates an imaging assembly 1520 that includes slit collimator 1410 and an extra-wide detector 1522.
  • Extra-wide detector 1522 is for example two detectors side-by-side.
  • Imaging assembly 1520 is an example of imaging assembly 120 of FIG. 1.
  • the relatively increased width of extra- wide detector 1522 enables an embodiment of rotating-slit gamma-ray imager 100 that includes imaging assembly 1520 to obtain larger images by increasing axial distance between slit collimator 1410 and detector 1522.
  • the increased width of detector 1522 enables creation of large images with enhanced field-of-view without sacrificing spatial resolution.
  • a method for forming an optimized image of a subject includes steps of acquiring a plurality of one-dimensional images, generating a measured sinogram from the plurality of one-dimensional images, and determining a plurality of trial images.
  • the method acquires a plurality of one-dimensional images of the subject captured by a rotating-slit imager having (a) a detector, and (b) a slit collimator having a slit oriented at one of a respective plurality of slit-rotation angles, relative to the subject, about a longitudinal axis substantially perpendicular to a front surface of the detector.
  • the method iteratively determines a plurality of trial images each having a respective trial sinogram.
  • the optimized image is one of the plurality of trial images and its corresponding trial sinogram differs from the measured sinogram by less than a predetermined tolerance.
  • each of the plurality of slit- rotation angles being denoted by a respective angle 6> m
  • the step of iteratively determining may include generating each respective trial sinogram at least in part by computing a Radon transform on the respective trial image.
  • the iteratively determining may be performed, in part, by imposing a positivity constraint such that each of the plurality of trial images has a respective plurality of pixel values representing a positive photon count.
  • the step of iteratively determining may be performed, in part, according to at least one of an expectation-maximization algorithm and a maximum likelihood estimation method.
  • the step of iteratively determining may be performed, in part, according to at least one of a maximum a posteriori reconstruction, an algebraic reconstruction technique, a Landweber reconstruction algorithm, and a stochastic optimization method.
  • any method denoted by one of (Al) through (A6), in which the subject is at a first angular orientation relative to the rotating-slit imager may further include, for each of a plurality of additional angular orientations of the subject relative to either the rotating-slit imager or an additional rotating-slit imager, steps of (i) repeating the step of acquiring, (ii) repeating the step of generating, and (iii) repeating the step of iteratively determining to yield a plurality of additional optimized images corresponding to the plurality of additional angular orientations.
  • the above method denoted by (A7) may further include generating a reconstructed volume-image by combining the plurality of optimized images using a three-dimensional tomographic construction algorithm.
  • a rotating-slit gamma-ray imager includes (i) a detector having an array of pixels, (ii) a slit collimator, (iii) a memory storing non-transitory computer-readable instructions, and (iv) a microprocessor.
  • the slit collimator has a slit at least partially aligned with the detector and configured to be oriented at any of a plurality of slit-rotation angles in a plane substantially parallel to the array of pixels.
  • the microprocessor is adapted to execute the instructions to form an optimized image of a subject by implementing any of the methods denoted above as (Al) through (A8).
  • the slit collimator may include a slit orientated parallel to the plurality of columns, and the slit collimator and the detector may be configured to rotate together such that the slit remains parallel to the plurality of pixel columns.
  • Any rotating-slit gamma-ray imager denoted by one of (Bl) through (B3) may further include a second detector and a second slit collimator.
  • the second detector has a second array of pixels and having a second field of view that at least partially overlaps a first field of view of the first detector.
  • the second slit collimator having a second slit at least partially aligned with the second detector and in a plane substantially parallel to the second array of pixels.
  • the microprocessor may be further adapted to execute the instructions to form a second optimized image of the subject by: (i) for each of a plurality of second slit-rotation angles of the second slit, acquiring a respective one of a plurality of second one-dimensional images of the subject captured by a the second detector, (ii) generating a second measured sinogram from the plurality of second one-dimensional images, and (iii) iteratively determining a plurality of second trial images each having a respective second trial sinogram, the second optimized image being one of the second trial images, with its corresponding second trial sinogram differing from the second measured sinogram by less than a second predetermined tolerance.
  • the microprocessor may be further adapted to execute the instructions to generate a reconstructed volume-image by combining the first and second optimized images using a three-dimensional tomographic construction algorithm.
  • a data processor includes (a) a memory storing non-transitory computer-readable instructions and (b) a microprocessor.
  • the microprocessor is configured to execute the instructions to: (i) generate a measured sinogram from a plurality of one-dimensional images captured through a slit oriented at one of a respective plurality of slit-rotation angles, relative to a subject, about a longitudinal axis substantially perpendicular to a front surface of a detector, and (ii) generate an optimized image of the subject by iteratively determining a plurality of trial images each having a respective trial sinogram, the optimized image being one of the trial images, its corresponding trial sinogram differing from the measured sinogram by less than a predetermined tolerance.
  • the microprocessor may be further configured to execute the instructions to:

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • Physics & Mathematics (AREA)
  • Optics & Photonics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Biophysics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Veterinary Medicine (AREA)
  • Pathology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Biomedical Technology (AREA)
  • Public Health (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • High Energy & Nuclear Physics (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Nuclear Medicine (AREA)
  • Algebra (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)

Abstract

A method for forming an optimized image of a subject includes steps of acquiring a plurality of one-dimensional images, generating a measured sinogram from the plurality of one-dimensional images, and determining a plurality of trial images. In the step of acquiring, the method acquires a plurality of one-dimensional images of the subject captured by a rotating-slit imager having (a) a detector, and (b) a slit collimator having a slit oriented at one of a respective plurality of slit-rotation angles, relative to the subject, about a longitudinal axis substantially perpendicular to a front surface of the detector. In the step of determining, the method iteratively determines a plurality of trial images each having a respective trial sinogram. The optimized image is one of the plurality of trial images and its corresponding trial sinogram differs from the measured sinogram by less than a predetermined tolerance.

Description

ROTATING-SLIT GAMMA-RAY IMAGER AND ASSOCIATED
IMAGING METHOD
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims priority to U.S. Provisional Application No. 62/250,994 filed November 4, 2015, which is incorporated by reference in its entirety.
BACKGROUND
[0002] Single photon emission computed tomography (SPECT) plays an important role in cancer diagnoses, cardiac-perfusion testing, and drug-delivery studies. Due to the low photon-collection efficiency of current parallel-hole collimators and pinhole apertures, SPECT projection images have high Poisson noise.
SUMMARY
[0003] A rotating slit is proposed that addresses at least the
aforementioned noise-related limitations.
[0004] In a first embodiment, a method for forming an optimized image of a subject includes steps of acquiring a plurality of one-dimensional images, generating a measured sinogram from the plurality of one-dimensional images, and determining a plurality of trial images. In the step of acquiring, the method acquires a plurality of one-dimensional images of the subject captured by a rotating-slit imager having (a) a detector, and (b) a slit collimator having a slit oriented at one of a respective plurality of slit-rotation angles, relative to the subject, about a longitudinal axis substantially perpendicular to a front surface of the detector. In the step of determining, the method iteratively determines a plurality of trial images each having a respective trial sinogram. The optimized image is one of the plurality of trial images and its corresponding trial sinogram differs from the measured sinogram by less than a predetermined tolerance.
[0005] In a second embodiment, a rotating-slit gamma-ray imager includes (i) a detector having an array of pixels, (ii) a slit collimator, (iii) a memory storing non-transitory computer-readable instructions, and (iv) a microprocessor. The slit collimator has a slit at least partially aligned with the detector and configured to be oriented at any of a plurality of slit-rotation angles in a plane substantially parallel to the array of pixels. The microprocessor is adapted to execute the instructions to form an optimized image of a subject by implementing the method of the first embodiment.
[0006] In a third embodiment, a data processor includes a memory storing non-transitory computer-readable instructions and a microprocessor. The
microprocessor is configured to execute the instructions to (i) generate a measured sinogram from a plurality of one-dimensional images, of a subject, captured by a rotating-slit imager having (a) a detector, and (b) a slit collimator having a slit oriented at one of a respective plurality of slit-rotation angles, relative to the subject, about a longitudinal axis substantially perpendicular to a front surface of the detector, and (ii) generate an optimized image by iteratively determining a plurality of trial images each having a respective trial sinogram, the optimized image being one of the trial images, its corresponding trial sinogram differing from the measured sinogram by less than a predetermined tolerance.
BRIEF DESCRIPTION OF THE FIGURES
[0007] FIG. 1 is a schematic diagram of a rotating-slit gamma-ray imager, in an embodiment.
[0008] FIG. 2 illustrates an example of the imaging assembly of the rotating-slit gamma-ray imager of FIG. 1.
[0009] FIG. 3 is a transverse cross-sectional view of a digital phantom used to simulate operation of the rotating-slit gamma-ray imager of FIG. 1, in an embodiment.
[0010] FIG. 4 illustrates a selected plurality of one-dimensional images of the digital phantom of FIG. 3 captured by the imaging assembly of FIG. 2.
[0011] FIG. 5 shows an exemplary sinogram (left), its inverse-Radon- transform image (middle), and a Maximum-Likelihood Expectation Maximization (MLEM) reconstructed image (right).
[0012] FIG. 6 is a comparison between physical pinhole aperture images and exemplary synthesized virtual pinhole images, the latter formed by an
embodiment of the rotating-slit gamma-ray imager of FIG. 1.
[0013] FIG. 7 is a perspective view of a second example of the imaging assembly of the rotating-slit gamma-ray imager of FIG. 1. [0014] FIG. 8 is a perspective view of twelve imaging assemblies of FIG. 7 arranged to have a common overlapping field of view, in an embodiment.
[0015] FIG. 9 is a flowchart illustrating a method for forming an optimized image, in an embodiment.
[0016] FIG. 10 shows an experimental setup for a demonstration of the rotating-slit gamma-ray imager of FIG. 1, in an embodiment.
[0017] FIG. 11 is a photograph of a phantom that consists of six iodine- 125 seeds.
[0018] FIG. 12 shows exemplary projection images from physical-pinhole and pinhole-synthesis collimation, the latter formed by an embodiment of the rotating- slit gamma-ray imager of FIG. 1.
[0019] FIG. 13 illustrates a comparison of image acquisition via an embodiment of the rotating-slit gamma-ray imager of FIG. 1 and a pinhole gamma- ray imager.
[0020] FIG. 14 is a perspective view of a third example of the imaging assembly of the rotating-slit gamma-ray imager of FIG. 1.
[0021] FIG. 15 is a perspective view of a fourth example of the imaging assembly of the rotating-slit gamma-ray imager of FIG. 1.
DETAILED DESCRIPTION OF THE EMBODIMENTS
[0022] Rotating-slit pinhole-aperture synthesis yields a much higher photon count image than a physical pinhole, given the same data acquisition time. Slit imaging with a collimator only requires a one-dimensional detector if it is made to rotate with the collimator. The sensitivity advantage becomes very significant as pinhole radius R decreases, because the slit sensitivity scales as 2R, whereas the pinhole sensitivity scales as R2.
[0023] FIG. 1 is a schematic diagram of an exemplary rotating-slit gamma-ray imager 100. Rotating-slit gamma-ray imager 100 includes one or more imaging assemblies 120. Each imaging assembly 120 includes a slit collimator 122 and a detector 124. Slit collimator 122 includes a slit 121 oriented at a rotation angle 127. Imager 100 also includes a data processor 130 communicatively coupled to each detector 124. [0024] Herein, an element in the figures denoted by a reference numeral suffixed by a parenthetical numeral is an example of the element indicated by the reference numeral. For example, imaging assembly 120(2) is an example (2) of imaging assembly 120. Rotating-slit gamma-ray imager 100 may also include additional imaging assemblies 120(2) to 120(N), where integer N≥ 2. Each imaging assembly 120 may have a respective viewing angle 128 of a common imaging volume in the field of view of each imaging assembly 120.
[0025] Detector 124 is configured to capture a plurality of one- dimensional images 129. In FIG. 1, one-dimensional images 129 from a single imaging assembly 120 include M images indexed by an integer m. Index m E { 1, 2, ... , M) denotes rotation angle 127(m) of slit 121 about an axis substantially perpendicular to at least a portion of detector 124. Integer M is, for example, between sixty and one-hundred eighty, which corresponds to three degrees per angular step and one degree per angular step, respectively. Detector 124 may corotate with slit collimator 122.
[0026] Rotation angle 127 may be a relative angle between slit collimator 122 and an object being imaged. Accordingly, rotation angle 127 may be changed by rotating the object being imaged with respect to slit collimator 122, or vice versa. Herein, reference to a changing rotation angle 127 corresponds to rotation of at least one of (a) slit collimator 122 and (b) the object being imaged, where rotations occur about an axis substantially perpendicular to at least a portion of detector 124.
[0027] Data processor 130 includes a memory 132, microprocessor 134, and software 140. Memory 132 may be transitory and/or non-transitory and may represent one or both of volatile memory (e.g., SRAM, DRAM, other volatile memory, or any combination thereof) and non-volatile memory (e.g., FLASH, ROM, magnetic media, optical media, other non-volatile memory, or any combination thereof). Software 140 includes machine-readable instructions. Microprocessor 134 is adapted to execute the instructions to perform functions of rotating-slit gamma-ray imager 100 as described herein. Microprocessor 134 may be a digital signal processor such as an image processor.
[0028] Software 140 includes a sinogram generator 141, an image reconstructor 142, and optionally a volume-image reconstructor 143. Sinogram generator 141 generates a sinogram 151(1) from one-dimensional images 129(1-M) and optionally generates sinograms 151 (2-N) from respective one-dimensional images 129(1-M, 2-N). Image reconstructor 142 generates a reconstructed image 152(1) from sinogram 151 (1) and optionally generates respective reconstructed images 152(2-N) from sinograms 151 (2-N).
[0029] Image reconstructor 142 may implement an iterative algorithm. The iterative algorithm may include at least one of an expectation-maximization algorithm, a maximum likelihood estimation method, maximum a posteriori reconstruction, an algebraic reconstruction technique, a Landweber reconstruction algorithm, and a stochastic optimization method. Examples of stochastic optimization method include simulated annealing, evolutionary algorithms, and swarm algorithms.
[0030] In an embodiment, a sinogram of reconstructed image 152(1), generated at least in part by computing a Radon transform of image 152, differs from sinogram 151(1) by less than a predetermined tolerance. The predetermined tolerance, for example, functions as a convergence criterion for the iterative algorithm. In an embodiment, image reconstructor 142 implements an iterative algorithm that minimizes a merit function, such as an Z^-norm or an L2-norm, or maximizes a merit function such as likelihood function.
[0031] Optional volume-image reconstructor 143 generates a
reconstructed volume-image 153 from reconstructed images 152. Volume-image reconstructor 143 implements a tomographic construction algorithm, for example. The tomographic construction algorithm may include at least one of a filtered back projection algorithm, MLEM, an algebraic reconstruction technique, a Landweber reconstruction algorithm, and a maximum a posteriori reconstruction technique.
[0032] Data processor 130 may also include a hardware controller 136 that controls the angular orientation of slit collimator 122. In an embodiment, rotating-slit gamma-ray imager 100 does not include data processor 130, but is communicatively coupled with data processor 130 remotely located from rotating-slit gamma-ray imager 100. In such an embodiment, rotating-slit gamma-ray imager 100 and data processor 130 may be communicatively coupled via wireless or wired communication protocols known in the art. In such an instance, data processor 130 may operate as a post-processing unit compatible for processing images generated by one or more types of rotating-slit gamma-ray imagers. [0033] FIG. 2 illustrates an imaging assembly 220 imaging a subject 290. Imaging assembly 220 is an example of imaging assembly 120 of FIG. 1. Imaging assembly 220 includes a rotating-slit collimator 222 and a detector 224, which are examples of collimator 122 and detector 124, respectively.
[0034] Collimator 222 may be formed of a material that blocks gamma rays and includes a slit 221. Slit 221 has a slit width 221W and a slit height 221H in the x and y directions respectively in a coordinate system 298. Herein, directions x, y, and z are with reference to coordinate system 298 unless otherwise specified.
[0035] Detector 224 has a plurality of oblong pixels 226(p) in a plane parallel to the x-y plane. Integer p is a pixel index in the x direction, each pixel 226 has a width 226W. The long dimension of the pixels 226 are aligned with slit 221. For clarity of illustration, not all pixels 226 are labeled in FIG. 2. Pixels 226 may include a semiconductor material suitable for detection of gamma rays, such as cadmium zinc telluride (CdZnTe or CZT).
[0036] In an exemplary mode of operation of imager 100 that includes at least one imaging assembly 220, at least one of imaging assembly 220 and subject 290 rotate about a rotation axis 202. For example, imaging assembly 220 rotates while subject 290 remains stationary. Alternatively, imaging assembly 220 remains stationary while subject 290 rotates. In either case, imaging assembly 220 may be mechanically configured such that each pixel 226 remains parallel to slit 221 during rotation. Herein, slit 221 being parallel to pixels 226 is equivalent to slit-collimator 222 being parallel to pixels 226. Rotation axis 202 is substantially perpendicular to a plane that includes pixels 226, for example, a front surface of detector 224.
[0037] Collimator 222 has a rotation angle 227, which corresponds to the orientation of slit 221 with respect to the x direction. Rotation angle 227 is an example of rotation angle 127 (FIG. 1). In FIG. 2, slit 221 is oriented parallel to t ey- direction for illustrative purposes only.
[0038] Both slit width 221W and slit height 221H of slit 221 may be adjustable. An adjustable width enables adaptive imaging, e.g., for multiple resolution requirements. An adjustable height enables control of image artifacts when forming reconstructed volume-image 153. One source of artifacts is stray light, such as electromagnetic radiation reaching detector 224 outside the field of view of imaging assembly 220. [0039] During acquisition of one-dimensional images 129, rotation of imaging assembly 220 or subject 290 may either have continuous smooth motion or discrete motion, where rotation occurs in steps with momentary pauses therebetween. One-dimensional images 129 may be acquired with uniform sampling, for example, corresponding to uniformly-spaced rotation angles between 0° and 360°.
Alternatively, angular sampling may be non-uniform, which enables adaptive angular sampling.
[0040] Rotating-slit collimator 222 and detector 224 are each in planes substantially parallel to the x-y plane and separated by a distance 215 along the z- direction. Distance 215 may also be adjustable, which also allows for adaptive imaging. Rotation axis 202 intersects the center of slit collimator 222, and is substantially perpendicular to the x-y plane. Herein, substantially parallel and substantially perpendicular may include, but is not limited to, within ±5° of being parallel and perpendicular, respectively.
[0041] Rotating-slit collimator 222 has a width 222W in the x direction. Detector 224 has a width 224W in the x direction and height 224H in the y direction. In an embodiment, widths 222W, 224W and heights 221H, 224H are each 100 mm, slit width 221W is 0.3 mm, width 226W is 0.1 mm, and distance 215 is 100 mm.
[0042] Detector 224 may be replaced with a two-dimensional array of pixels without departing from the scope hereof. Each pixel of such a two-dimensional array has an aspect ratio of less than two, for example. The two-dimensional pixel array may corotate with rotating-slit collimator 222 such that each row of pixels or column of pixels is parallel to slit 221 for any rotational orientation of slit 221.
Alternatively, the 2-D detector may be stationary with respect to a rotational orientation of slit 221 such that pixels of the 2-D detector are grouped according to the angle of slit 221. For example, a line parallel to slit 221 intersects a plurality of pixels, the plurality of pixels forming the group. The contribution of each pixel of the group, to a one-dimensional image 129, depends in part on where the line intersects the pixel, e.g., the pixel's center or corner.
[0043] In an embodiment, detector 224 is also translatable in the x-y plane, for example, in a direction perpendicular to the angular orientation of slit 221. Such lateral movement enables dynamic tracking of a region of interest of subject 290, for example, during rotation of one or both of slit-collimator 222 and detector 224. The amplitude of lateral movement in the x-y plane may be in part determined by a trigonometric function of rotation angle 227. For example, the magnitude is proportional to the sine or cosine of rotation angle 227.
[0044] FIG. 3 is a transverse cross-sectional view of a digital phantom 300 used to simulate operation of rotating-slit gamma-ray imager 100. Digital phantom 300 includes nine hot (radioactive) rods of three different diameters.
[0045] FIG. 4 illustrates a selected plurality of one-dimensional images 429 of digital phantom 300 captured by imaging assembly 220. Images 429 were captured with imaging assembly 220 oriented such that rotation axis 202 intersects the axes of the rods of phantom 300 at a 45° angle. One-dimensional images 429 are each an example of one-dimensional image 129, and correspond to a different rotation angle 227. FIG. 4 illustrates one-dimensional images 429(0, 30, 60, 90, 120, 150), which correspond to respective rotation angles 9m = {0°, 30°, 60°, 90°, 120°, 150°} .
Images 429(0, 30, 60, 90, 120, 150) are part of a full-set of ninety of images 429 captured at ninety evenly-spaced rotation angles 6m E {0°, 2°, 176°, 178°} .
[0046] In FIG. 4, a two-dimensional x-y coordinate system 498 is superimposed on each one-dimensional image 429 such that the x axis and y axis are oriented, respectively, perpendicular and parallel to the respective slit rotation angle of each image 429. As in coordinate system 298 of FIG. 2, x = 0 of coordinate system 498 corresponds to the region of an image 429 aligned with slit 221. Each image 429(6m) has respective pixel values Pem (x > y) shown in FIG. 4 as a gray-scale map.
[0047] In response to uniform illumination of slit 221 , pixel values ρθ;η (χ, y) decrease as a function of distance from slit 221 according to an illumination falloff function that varies with x. An exemplary illumination falloff function is E(x) = cos(< >), where tan(< >) = x/L and L is distance 215. Hence,
E(x) = L/Vx2 + L2.
[0048] If imaging assembly 220 is rotated through a set of rotation angles 227 spanning a 180° range, the data from detector 224 can be arranged into a sinogram, with one axis representing the pixel index of 1 -D detector 124, and the other axis representing the rotation angle. In embodiments of rotating-slit gamma-ray imager 100 that include a stationary 2-D detector instead of detector 124, forming a sinogram can be accomplished with suitable combinations of pixels. [0049] FIG. 5 shows an exemplary sinogram 551 and its corresponding inverse-Radon-transform image 520, and a reconstructed image 552. Sinogram 551 is an example of sinogram 151 and is a sinogram of the ninety of images 429. Inverse- Radon-transform image 520 is generated via filtered back projection. Reconstructed image 552 is an example of reconstructed image 152. Image 552 was generated via sixty iterations of a MLEM reconstruction method. Images 520 and 552 were both generated in-part from the ninety one-dimensional images 429 discussed above regarding FIG. 4.
[0050] Image 552 has a trial sinogram that differs from sinogram by less than a predetermined tolerance of the MLEM reconstruction method. The trial sinogram is, in part, the result of processing image 552 with a Radon transform and operations that remove effects of the aforementioned illumination falloff function E(x) .
[0051] Sinogram 551 has a plurality of intensity values S551 (6m, x), where 6m and x are defined in FIG. 4. For a given rotation angle 6m sinogram 551 has intensity values S551 (6m, x) = / Pem (x > y)dy. Without departing from the scope hereof, intensity values S551 (6m, x) may be proportional to / Pem (x > y dy or a discrete sum∑y Pem (x, y).
[0052] When the Poisson noise is high due to a shortened data acquisition duration at each angle, the inverse Radon transform (e.g., via filtered back projection) performs poorly. A comparison of images 520 and 552 illustrates this, as image 520, resulting from an inverse Radon transform, has less contrast and more noise than image 552. However, if an iterative reconstruction method (such as MLEM) that enforces positivity is applied, the reconstructed image is improved relative to both image 520 but also a physical pinhole image acquired in the same total acquisition time, as FIG. 6 illustrates.
[0053] FIG. 6 is a comparison between physical pinhole aperture images (left column) and synthesized virtual pinhole images (right column). For each row, the data acquisition time is the same. The second row's acquisition time is ten times that of the first row. The phantom in the third row is immersed in a cube-shape background (20% of the peak radioactivity of the hot rods). The phantom in the fourth row contains cold rods immersed in the background activity. The synthesized virtual pinhole images of FIG. 6 are examples of a two-dimensional reconstructed image 152. [0054] FIG. 7 is a perspective view of an imaging assembly 720, which is an example of imaging assembly 120 shown in FIG. 1. Imaging assembly 720 includes a slit collimator 722 and a detector 724, which are examples of slit collimator 122 and detector 124, respectively. Slit collimator 722 includes a slit 721, which is an example of slit 121. Imaging assembly 720 also includes a housing 730 and a base 740. Slit collimator 722 and detector 724 are mechanically attached to housing 730, which is configured to rotate in base 740 such that slit collimator 722 and detector 724 corotate.
[0055] FIG. 8 is a perspective view of twelve imaging assemblies 720, which are collectively an example of imaging assemblies 120(1— TV) of rotating-slit gamma-ray imager 100, where N = 12. Imaging assemblies 720 have respective viewing angles of a subject 890. The viewing angles are examples of viewing angles 128.
[0056] FIG. 9 is a flowchart illustrating a method 900 for forming an optimized image of a subject. Method 900 may be implemented within one or more aspects of rotating-slit gamma-ray imager 100. For example, method 900 is implemented by microprocessor 134 executing computer-readable instructions of software 140, shown in FIG. 1.
[0057] In step 910, method 900 acquires a plurality of one-dimensional images of a subject with a rotating-slit gamma-ray imager having a detector and a slit collimator. Each one-dimensional image corresponds to a different relative angular orientation of the slit collimator and the subject about a longitudinal axis substantially perpendicular to a front surface of the detector. In an example of step 910, rotating-slit gamma-ray imager 100 acquires one-dimensional images 429.
[0058] In step 920, method 900 generates a measured sinogram from the plurality of one-dimensional images. In an example of step 920, sinogram generator 141 of rotating-slit gamma-ray imager 100 generates sinogram 551 of FIG. 5.
[0059] In step 930, method 900 iteratively determines a plurality of trial images each having a respective trial sinogram. The optimized image is one of the trial images and has a corresponding trial sinogram differing from the measured sinogram by less than a predetermined tolerance. In an example of step 930, image reconstructor 142 of data processor 130 generates a two-dimensional reconstructed image 552. [0060] Step 940 is optional. In step 940, method 900 repeats steps 910, 920, and 930 for each of a plurality of additional angular orientations of the subject relative to either the rotating-slit imager or an additional rotating-slit imager. Step 940 results in a plurality of optimized images. In an example of step 940, each imaging assembly 720 in the figuration of FIG. 8 generates a respective optimized image of subject 890.
[0061] Step 950 is optional. In step 950, method 900 generates a reconstructed volume-image by combining the plurality of optimized images using a three-dimensional tomographic construction algorithm. In an example of step 950, volume-image reconstructor 143 of data processor 130 generates reconstructed volume-image 153.
[0062] FIG. 10 shows an experimental setup 1095 for a demonstration of a rotating-slit gamma-ray imager 1000, which is an embodiment of rotating-slit gamma- ray imager 100. Rotating-slit gamma-ray imager 1000 includes an intensified quantum-imaging detector (iQID) camera 1020 that is downward-facing above an adjustable-slit collimator 1022 having a slit 1021. Slit 1021 is an example of slit 121.
[0063] iQID camera 1020 includes a detector 1024 (not shown) with dimensions 30 χ 40 mm and a resolution of approximately 0.14 mm. Camera 1020 is similar to the iQID camera discussed in "The iQID camera: An ionizing-radiation quantum imaging detector," Nucl. Inst. Meth. A, 767, 146-152, 2014 by Miller et al, which is incorporated herein by reference. Other cameras may be used as camera 1020 without departing from the scope hereof. Adjustable-slit collimator 1022 and detector 1024 are embodiments of adjustable-slit collimator 122 and detector 124, respectively.
[0064] In experimental setup 1095, iQID camera 1020 is 88 mm above adjustable-slit collimator 1022 and a rotation stage 1092 is located 68-mm beneath slit collimator 1022. This imaging configuration results in a magnification of 1.3.
Rotation stage 1092 has a rotation axis, not labeled, that is example of rotation axis 202 of FIG. 2.
[0065] For simplicity, rotation stage 1092 rather than the slit collimator 1022 rotates to generate images. Slit 1021 of adjustable slit collimator 1022 may be set to have a 0.5 -mm slit width. Slit 1021 has a length that is one hundred times its width. A corresponding physical pinhole 1014 (with dimensions 0.5 χ 0.5 mm), may be created by adding a second orthogonal slit with the same width.
[0066] FIG. 1 1 is a photograph of a phantom 1100 in a holder 1 120. Phantom 1100 includes six iodine-125 (125I) seeds 1 110(1-6) each having a maximum gamma-ray photon energy of approximately 36 keV.
[0067] FIG. 12 shows projection images 1210, 1220, 1230, and 1240 from physical-pinhole and pinhole-synthesis collimation corresponding to experimental setup 1095 with phantom 1100 on rotation stage 1092. Proj ection images 1220 and 1240 are synthesized virtual pinhole proj ection images corresponding to physical pinhole projection images 1210 and 1230 respectively. Projection images 1210, 1220, 1230, and 1240 result from acquisition times of 100 seconds, 90 seconds, 1000 seconds, and 900 seconds, respectively.
[0068] Projection images 1210 and 1230 are projection images from physical pinhole 1014 having dimensions 0.5 χ 0.5 mm. Projection images 1220 and 1240 are reconstructed projection images from a synthesized virtual pinhole, that is, a pinhole synthesized from slit-aperture imager 1000, with slit width 1021W equal to 0.5 mm, having captured multiple one-dimensional images 129 of phantom 1100 at different slit rotation angles.
[0069] Projection images 1220 and 1240 are reconstructed by data processor 130, using an iterative MLEM method that enforces positivity, from a plurality of one-dimensional images 129 corresponding to different angular orientations of phantom 1 100, via rotation of rotation stage 1092. Projection images 1220 and 1240 are each examples of a two-dimensional reconstructed image 152.
[0070] Projection images 1220 and 1240 have the same resolution as images 1210 and 1230 respectively. That is, the center of rotation of slit 1021, in the generation of projection images 1220 and 1240, corresponds to the location of physical pinhole 1014 used to generate projection images 1210 and 1230, respectively. The total sensitivity of projection images 1220 and 1240 is enhanced— the additional sensitivity being afforded by slit 1021 having a slit width 1021W (0.5- mm in this example) approximately equal to the effective diameter of pinhole 1014— but with a much longer length. Although pinhole 1014 is square, its equivalent diameter is close to 0.5 mm. [0071] FIG. 13 illustrates a comparison of image acquisition via a rotating-slit gamma-ray imager 1300 and a traditional pinhole gamma-ray imager 1390. Rotating-slit gamma-ray imager 1300 includes imaging assembly 720 of FIG. 7. Traditional pinhole gamma-ray imager 1390 includes a pinhole aperture 1392 and a two-dimensional pixel array 1394. Output image 1340 is acquired directly from traditional pinhole gamma-ray imager 1390. By contrast, rotating-slit gamma-ray imager 100 generates sinogram 551 , discussed above with respect to FIG. 5. Image 552 is reconstructed from sinogram 551 using sixty iterations of a MLEM
reconstruction algorithm.
[0072] FIG. 14 illustrates an imaging assembly 1420 that includes a slit collimator 1410 and an elongated detector 1422. Slit collimator 1410 includes a slit 1412. Elongated detector 1422 is for example two adjacent detectors 1422(1 ,2).
Imaging assembly 1420 is an example of imaging assembly 120 of FIG. 1. The relatively increased length of elongated detector 1422 enables an embodiment of rotating-slit gamma-ray imager 100 that includes imaging assembly 1420 to detect more gamma-ray producing events than with a conventional smaller detector, which result in enhanced detection sensitivity.
[0073] FIG. 15 illustrates an imaging assembly 1520 that includes slit collimator 1410 and an extra-wide detector 1522. Extra-wide detector 1522 is for example two detectors side-by-side. Imaging assembly 1520 is an example of imaging assembly 120 of FIG. 1. The relatively increased width of extra- wide detector 1522 enables an embodiment of rotating-slit gamma-ray imager 100 that includes imaging assembly 1520 to obtain larger images by increasing axial distance between slit collimator 1410 and detector 1522. The increased width of detector 1522 enables creation of large images with enhanced field-of-view without sacrificing spatial resolution.
Combinations of features:
[0074] Features described above as well as those claimed below may be combined in various ways without departing from the scope hereof. The following examples illustrate some possible, non-limiting combinations:
[0075] (Al) A method for forming an optimized image of a subject includes steps of acquiring a plurality of one-dimensional images, generating a measured sinogram from the plurality of one-dimensional images, and determining a plurality of trial images. In the step of acquiring, the method acquires a plurality of one-dimensional images of the subject captured by a rotating-slit imager having (a) a detector, and (b) a slit collimator having a slit oriented at one of a respective plurality of slit-rotation angles, relative to the subject, about a longitudinal axis substantially perpendicular to a front surface of the detector. In the step of determining, the method iteratively determines a plurality of trial images each having a respective trial sinogram. The optimized image is one of the plurality of trial images and its corresponding trial sinogram differs from the measured sinogram by less than a predetermined tolerance.
[0076] (A2) In the method denoted by (Al), each of the plurality of slit- rotation angles being denoted by a respective angle 6>m, and the one-dimensional image corresponding to 6m having pixel values Pem(x > y), the measured sinogram may have pixel values
Figure imgf000016_0001
x) = J Pem(x > y)dy, where x and y are locations along transverse axes that are (a) perpendicular to and parallel to the slit, respectively, and (b) substantially orthogonal to the longitudinal axis.
[0077] (A3) In any method denoted by one of (Al) and (A2), the step of iteratively determining may include generating each respective trial sinogram at least in part by computing a Radon transform on the respective trial image.
[0078] (A4) In any method denoted by one of (Al) through (A3), the iteratively determining may be performed, in part, by imposing a positivity constraint such that each of the plurality of trial images has a respective plurality of pixel values representing a positive photon count.
[0079] (A5) In any method denoted by one of (Al) through (A4), the step of iteratively determining may be performed, in part, according to at least one of an expectation-maximization algorithm and a maximum likelihood estimation method.
[0080] (A6) In any method denoted by one of (Al) through (A5), the step of iteratively determining may be performed, in part, according to at least one of a maximum a posteriori reconstruction, an algebraic reconstruction technique, a Landweber reconstruction algorithm, and a stochastic optimization method.
[0081] (A7) Any method denoted by one of (Al) through (A6), in which the subject is at a first angular orientation relative to the rotating-slit imager, may further include, for each of a plurality of additional angular orientations of the subject relative to either the rotating-slit imager or an additional rotating-slit imager, steps of (i) repeating the step of acquiring, (ii) repeating the step of generating, and (iii) repeating the step of iteratively determining to yield a plurality of additional optimized images corresponding to the plurality of additional angular orientations.
[0082] (A8) The above method denoted by (A7) may further include generating a reconstructed volume-image by combining the plurality of optimized images using a three-dimensional tomographic construction algorithm.
[0083] (Bl) A rotating-slit gamma-ray imager includes (i) a detector having an array of pixels, (ii) a slit collimator, (iii) a memory storing non-transitory computer-readable instructions, and (iv) a microprocessor. The slit collimator has a slit at least partially aligned with the detector and configured to be oriented at any of a plurality of slit-rotation angles in a plane substantially parallel to the array of pixels. The microprocessor is adapted to execute the instructions to form an optimized image of a subject by implementing any of the methods denoted above as (Al) through (A8).
[0084] (B2) In the rotating-slit gamma-ray imager denoted by (Bl), in which the array of pixels includes a plurality of pixel columns, the slit collimator may include a slit orientated parallel to the plurality of columns, and the slit collimator and the detector may be configured to rotate together such that the slit remains parallel to the plurality of pixel columns.
[0085] (B3) In any rotating-slit gamma-ray imager denoted by one of (Bl) and (B2), each of the plurality of slit-rotation angles being denoted by a respective angle 6m, and the one-dimensional image corresponding to 6m having pixel values Pem(x, y), the measured sinogram may have pixel values Si(0m, x) = / Pem(x > y)dy, where x and y are locations along transverse axes that are (a) perpendicular to and parallel to the slit, respectively, and (b) substantially orthogonal to the longitudinal axis.
[0086] (B4) Any rotating-slit gamma-ray imager denoted by one of (Bl) through (B3) may further include a second detector and a second slit collimator. The second detector has a second array of pixels and having a second field of view that at least partially overlaps a first field of view of the first detector. The second slit collimator having a second slit at least partially aligned with the second detector and in a plane substantially parallel to the second array of pixels. The microprocessor may be further adapted to execute the instructions to form a second optimized image of the subject by: (i) for each of a plurality of second slit-rotation angles of the second slit, acquiring a respective one of a plurality of second one-dimensional images of the subject captured by a the second detector, (ii) generating a second measured sinogram from the plurality of second one-dimensional images, and (iii) iteratively determining a plurality of second trial images each having a respective second trial sinogram, the second optimized image being one of the second trial images, with its corresponding second trial sinogram differing from the second measured sinogram by less than a second predetermined tolerance. The microprocessor may be further adapted to execute the instructions to generate a reconstructed volume-image by combining the first and second optimized images using a three-dimensional tomographic construction algorithm.
[0087] (CI) A data processor includes (a) a memory storing non-transitory computer-readable instructions and (b) a microprocessor. The microprocessor is configured to execute the instructions to: (i) generate a measured sinogram from a plurality of one-dimensional images captured through a slit oriented at one of a respective plurality of slit-rotation angles, relative to a subject, about a longitudinal axis substantially perpendicular to a front surface of a detector, and (ii) generate an optimized image of the subject by iteratively determining a plurality of trial images each having a respective trial sinogram, the optimized image being one of the trial images, its corresponding trial sinogram differing from the measured sinogram by less than a predetermined tolerance.
[0088] (C2) In the data processor denoted by (CI), in which each of the plurality of slit-rotation angles being denoted by a respective angle 6m, and the one- dimensional image corresponding to 6m having pixel values Pem(x > y), the measured sinogram may have pixel values Si(0m, x) = / Pem(x > y)dy, where x and y are locations along transverse axes that are (a) perpendicular to and parallel to the slit, respectively, and (b) substantially orthogonal to the longitudinal axis.
[0089] (C3) In any data processor denoted by one of (CI) and (C2), the microprocessor may be further configured to execute the instructions to:
(i) generate a second measured sinogram from a plurality of second one-dimensional images, captured through a second detector and a second slit collimator having a slit oriented at one of a respective plurality of slit-rotation angles, relative to a subject, about a longitudinal axis substantially perpendicular to a front surface of a detector,
(ii) generate a second optimized image of the subject by iteratively determining a plurality of second trial images each having a respective second trial sinogram, the optimized image being one of the second trial images, its corresponding second trial sinogram differing from the second measured sinogram by less than a predetermined tolerance, and
(iii) generate a reconstructed volume-image by combining the first and second optimized images using a three-dimensional tomographic construction algorithm.
[0090] Changes may be made in the above methods and systems without departing from the scope hereof. It should thus be noted that the matter contained in the above description or shown in the accompanying drawings should be interpreted as illustrative and not in a limiting sense. The following claims are intended to cover all generic and specific features described herein, as well as all statements of the scope of the present method and system, which, as a matter of language, might be said to fall there between.

Claims

CLAIMS What is claimed is:
1. A method for forming an optimized image of a subject comprising:
acquiring a plurality of one-dimensional images of the subject captured by a rotating-slit imager having (a) a detector, and (b) a slit collimator having a slit oriented at one of a respective plurality of slit-rotation angles, relative to the subject, about a longitudinal axis substantially perpendicular to a front surface of the detector;
generating a measured sinogram from the plurality of one-dimensional
images; and
iteratively determining a plurality of trial images each having a respective trial sinogram, the optimized image being one of the plurality of trial images, its corresponding trial sinogram differing from the measured sinogram by less than a predetermined tolerance.
2. The method of claim 1, each of the plurality of slit-rotation angles being denoted by a respective angle 6m, the one-dimensional image corresponding to 6m having pixel values Pem(x > y), the measured sinogram having pixel values Si(0m, x) = / Pem(x > y)dy, where x and y are locations along transverse axes that are (a) perpendicular to and parallel to the slit, respectively, and (b) substantially orthogonal to the longitudinal axis.
3. The method claim 1, the iteratively determining including generating each respective trial sinogram at least in part by computing a Radon transform on the respective trial image.
4. The method of claim 1 , the iteratively determining being performed, in part, by imposing a positivity constraint such that each of the plurality of trial images has a respective plurality of pixel values representing a positive photon count.
5. The method of claim 1 , the iteratively determining being performed, in part, according to at least one of an expectation-maximization algorithm and a maximum likelihood estimation method.
6. The method of claim 1 , the iteratively determining being performed, in part, according to at least one of a maximum a posteriori reconstruction, an algebraic reconstruction technique, a Landweber reconstruction algorithm, and a stochastic optimization method.
7. The method of claim 1 , the subject being at a first angular orientation relative to the rotating-slit imager, and further comprising, for each of a plurality of additional angular orientations of the subject relative to either the rotating-slit imager or an additional rotating-slit imager:
repeating the step of acquiring;
repeating the step of generating; and
repeating the step of iteratively determining to yield a plurality of additional optimized images corresponding to the plurality of additional angular orientations.
8. The method of claim 7, further comprising:
generating a reconstructed volume-image by combining the plurality of
optimized images using a three-dimensional tomographic construction algorithm.
9. A rotating-slit gamma-ray imager comprising:
a detector having an array of pixels; a slit collimator having a slit at least partially aligned with the detector and configured to be oriented at any of a plurality of slit-rotation angles in a plane substantially parallel to the array of pixels;
a memory storing non-transitory computer-readable instructions;
a microprocessor adapted to execute the instructions to form an optimized image of a subject by: (i) acquiring, for each of the plurality of slit-rotation angles of the slit, a respective one of a plurality of one-dimensional images of the subject captured by the detector,
(ii) generating a measured sinogram from the plurality of one-dimensional images, and
(iii) iteratively determining a plurality of trial images each having a
respective trial sinogram, the optimized image being one of the trial images, its corresponding trial sinogram differing from the measured sinogram by less than a predetermined tolerance.
The rotating-slit gamma-ray imager of claim 9, the array of pixels including a plurality of pixel columns, the slit collimator including a slit orientated parallel to the plurality of columns, and the slit collimator and the detector being configured to rotate together such that the slit remains parallel to the plurality of pixel columns.
The rotating-slit gamma-ray imager of claim 9, each of the plurality of slit- rotation angles being denoted by a respective angle 6m, the one- dimensional image corresponding to 6m having pixel values Pem( y), the measured sinogram having pixel values Si(0m, x) = / Pem(x > y)dy, where x and y are locations along transverse axes that are (a) perpendicular to and parallel to the slit, respectively, and (b) substantially orthogonal to the longitudinal axis.
The rotating-slit gamma-ray imager of claim 9, the microprocessor further adapted to execute the instructions to, when iteratively determining, generate each respective trial sinogram at least in part by computing a Radon transform on the respective trial image.
The rotating-slit gamma-ray imager of claim 9, the iteratively determining being performed, in part, by imposing a positivity constraint such that each of the plurality of trial images has a respective plurality of pixel values representing a positive photon count.
14. The rotating-slit gamma-ray imager of claim 9, the iteratively determining being performed, in part, according to at least one of an expectation- maximization algorithm and a maximum likelihood estimation method.
15. The rotating-slit gamma-ray imager of claim 9, the iteratively determining being performed, in part according to at least one of a maximum a posteriori reconstruction, an algebraic reconstruction technique, a Landweber reconstruction algorithm, and a stochastic optimization method.
16. The rotating-slit gamma-ray imager of claim 9, further comprising: a second detector having a second array of pixels and having a second field of view that at least partially overlaps a first field of view of the first detector; and a second slit collimator having a second slit at least partially aligned with the second detector and in a plane substantially parallel to the second array of pixels,
the microprocessor being further adapted to execute the instructions to form a second optimized image of the subject by:
(i) for each of a plurality of second slit-rotation angles of the second slit, acquiring a respective one of a plurality of second one- dimensional images of the subject captured by the second detector,
(ii) generating a second measured sinogram from the plurality of
second one-dimensional images, and
(iii) iteratively determining a plurality of second trial images each having a respective second trial sinogram, the second optimized image being one of the second trial images, its corresponding second trial sinogram differing from the second measured sinogram by less than a second predetermined tolerance, the microprocessor being further adapted to execute the instructions to
generate a reconstructed volume-image by combining the first and second optimized images using a three-dimensional tomographic construction algorithm.
17. A data processor comprising:
a memory storing non-transitory computer-readable instructions; and a microprocessor configured to execute the instructions to:
(i) generate a measured sinogram from a plurality of one-dimensional images captured through a slit oriented at one of a respective plurality of slit- rotation angles, relative to a subject, about a longitudinal axis substantially perpendicular to a front surface of a detector, and
(ii) generate an optimized image of the subject by iteratively determining a plurality of trial images each having a respective trial sinogram, the optimized image being one of the trial images, its corresponding trial sinogram differing from the measured sinogram by less than a predetermined tolerance.
18. The data processor of claim 17, each of the plurality of slit-rotation angles being denoted by a respective angle 6m, the one-dimensional image corresponding to 6m having pixel values Pem(x > y), the measured sinogram having pixel values Si(0m, x) = / Pem(x > y)dy, where x and y are locations along transverse axes that are (a) perpendicular to and parallel to the slit, respectively, and (b) substantially orthogonal to the longitudinal axis.
19. The data processor of claim 17, the iteratively determining being
performed, in part, according to at least one of an expectation- maximization algorithm and a maximum likelihood estimation method.
20. The data processor of claim 17, the microprocessor being further
configured to execute the instructions to:
(i) generate a second measured sinogram from a plurality of second one- dimensional images, captured through a second detector and a second slit collimator having a slit oriented at one of a respective plurality of slit-rotation angles, relative to a subject, about a longitudinal axis substantially perpendicular to a front surface of a detector,
(ii) generate a second optimized image of the subject by iteratively
determining a plurality of second trial images each having a respective second trial sinogram, the optimized image being one of the second trial images, its corresponding second trial sinogram differing from the second measured sinogram by less than a predetermined tolerance, and
(iii) generate a reconstructed volume-image by combining the first and second optimized images using a three-dimensional tomographic construction algorithm.
PCT/US2016/060675 2015-11-04 2016-11-04 Rotating-slit gamma-ray imager and associated imaging method Ceased WO2017079654A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/773,691 US10548544B2 (en) 2015-11-04 2016-11-04 Rotating-slit gamma-ray imager and associated imaging method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201562250994P 2015-11-04 2015-11-04
US62/250,994 2015-11-04

Publications (1)

Publication Number Publication Date
WO2017079654A1 true WO2017079654A1 (en) 2017-05-11

Family

ID=58662839

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2016/060675 Ceased WO2017079654A1 (en) 2015-11-04 2016-11-04 Rotating-slit gamma-ray imager and associated imaging method

Country Status (2)

Country Link
US (1) US10548544B2 (en)
WO (1) WO2017079654A1 (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060072801A1 (en) * 2004-09-24 2006-04-06 General Electric Company Method and system for iterative image reconstruction
US20070253524A1 (en) * 2006-04-28 2007-11-01 Herbert Bruder Method for scattered radiation correction in the case of an X-ray CT, and X-ray CT for applying this method
US20080061395A1 (en) * 2006-09-07 2008-03-13 General Electric Company Tileable multi-layer detector

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060072801A1 (en) * 2004-09-24 2006-04-06 General Electric Company Method and system for iterative image reconstruction
US20070253524A1 (en) * 2006-04-28 2007-11-01 Herbert Bruder Method for scattered radiation correction in the case of an X-ray CT, and X-ray CT for applying this method
US20080061395A1 (en) * 2006-09-07 2008-03-13 General Electric Company Tileable multi-layer detector

Also Published As

Publication number Publication date
US20180344269A1 (en) 2018-12-06
US10548544B2 (en) 2020-02-04

Similar Documents

Publication Publication Date Title
US7671340B2 (en) Adjustable-focal-length collimators method and system
US8487265B2 (en) Imaging detector and method of manufacturing
JP7254322B2 (en) Collimator for medical image processing system and its image reconstruction method
US11156731B2 (en) Systems and methods for improved detector assembly sizing
TWI816171B (en) Spread field imaging collimators for radiation-based imaging
Dey Improvement of performance of cardiac SPECT camera using curved detectors with pinholes
US9915737B2 (en) Systems and methods for imaging with multi-head camera
Kim et al. A novel depth-of-interaction rebinning strategy for ultrahigh resolution PET
Yao et al. Derivation of system matrix from simulation data for an animal SPECT with slit-slat collimator
US10548544B2 (en) Rotating-slit gamma-ray imager and associated imaging method
US9958552B1 (en) Systems and methods for rotation based reconstruction for multi-column detector geometries
EP3267893B1 (en) Cone beam computed tomography projection values providing system and method
US20250232489A1 (en) Adaptive pixel adjustment for spect imaging system
Hong Three-Dimensional reconstruction methods in near-field coded aperture for spect imaging system
Havelin et al. A SPECT imager with synthetic collimation
JP7209496B2 (en) nuclear medicine diagnostic equipment
Verhaeghe et al. A PET supersets data framework for exploitation of known motion in image reconstruction
Cai et al. A prototype adaptive SPECT system with self-optimized angular sampling
Peterson SPECT Imaging using Pinhole Collimation: System Design and Simulation Studies for Pre-Clinical and Clinical Imaging
CA2518153A1 (en) Method and system for creating an image of a radiation source
Alessio Statistical modeling and reconstruction of rebinned PET measurements
Ma et al. Multi-pinhole dynamic SPECT imaging: simulation and system optimization
Yoder et al. A dual plane orbit approach for improved pinhole tomography
Ovchinnikov Developing a high sensitivity and spatial resolution SPECT molecular imaging system through the use of novel multiplexing

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16863100

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16863100

Country of ref document: EP

Kind code of ref document: A1