Impact of Axial Compression for the mMR Simultaneous PET-MR Scanner Martin A Belzunce, Jim O’Doherty and Andrew J Reader King's College London, Division.

Slides:



Advertisements
Similar presentations
Topic 8. Gamma Camera (II)
Advertisements

Image Reconstruction.
Advanced GAmma Tracking Array
Evaluation of Reconstruction Techniques
1 Low-Dose Dual-Energy CT for PET Attenuation Correction with Statistical Sinogram Restoration Joonki Noh, Jeffrey A. Fessler EECS Department, The University.
PET Design: Simulation Studies using GEANT4 and GATE - Status Report - Martin Göttlich DESY.
Computed Tomography II
Implementation of PROPELLER MRI method for Diffusion Tensor Reconstruction A. Cheryauka 1, J. Lee 1, A. Samsonov 2, M. Defrise 3, and G. Gullberg 4 1 –
Completion of a Truncated Attenuation Image from the Attenuated PET Emission Data Johan Nuyts, Christian Michel, Matthias Fenchel, Girish Bal, Charles.
IMAGE QUALITY NOISE LINEARITY CROSS-FIELD UNIFORMITY IMAGE ARTIFACTS.
Statistical image reconstruction
Simulation System for Emission Tomography (SimSET): using the new block detector feature. Robert L. Harrison, William C. J. Hunter, Steven B. Gillispie,
Estimating 3D Respiratory Motion from Orbiting Views Rongping Zeng, Jeffrey A. Fessler, James M. Balter The University of Michigan Oct Funding provided.
tomos = slice, graphein = to write
Despeckle Filtering in Medical Ultrasound Imaging
Maurizio Conti, Siemens Molecular Imaging, Knoxville, Tennessee, USA
Basic principles Geometry and historical development
COMPUTED TOMOGRAPHY I – RAD 365 CT - Scan
Supplementary Material Emission Computed Tomography
Automated Image Analysis Software for Quality Assurance of a Radiotherapy CT Simulator Andrew J Reilly Imaging Physicist Oncology Physics Edinburgh Cancer.
PET data preprocessing and alternative image reconstruction strategies.
QIBA CT Volumetrics - Cross-Platform Study (Group 1C) December 23, 2009 CT Cross-platform Sizing of Phantom Lesions Quantitative Imaging Biomarker Alliance.
Coincidence to Image: PET Imaging Jennifer White Marketing Manager SNS Workshop October 13, 2003.
Design and simulation of micro-SPECT: A small animal imaging system Freek Beekman and Brendan Vastenhouw Section tomographic reconstruction and instrumentation.
0 A Fast Time Incorporating Monte-Carlo Simulation of Wire Chamber Based Small Animal PET Scanners for Detector Scatter Correction M. Dawood 1, Don Vernekohl.
Fundamental Limits of Positron Emission Tomography
Medical Image Analysis Image Reconstruction Figures come from the textbook: Medical Image Analysis, by Atam P. Dhawan, IEEE Press, 2003.
QIBA CT Volumetrics - Cross-Platform Study (Group 1C) September 2, 2009 Interclinical Comparison of CT Volumetry Quantitative Imaging Biomarker Alliance.
Fig. 3 shows how a dose distribution that is initially relatively sharp becomes blurred with time, until after 16 hours, virtually no useful information.
Scanner Validation via the SNM Clinical Trials Network Phantom Program Paul E. Christian Molecular Imaging Program Huntsman Cancer Institute University.
Single Photon Emission Computed Tomography
HEALTH Novel MR-compatible PET detectors for simultaneous PET/MRI imaging FP7-HEALTH-2009-single-stage The focus should be to develop novel.
BMI I FS05 – Class 4 “Nuclear Imaging: Math” Slide 1 Biomedical Imaging I Class 5 – Radionuclide Imaging (PET, SPECT), Part 3: Attenuation and Scatter.
Professor Brian F Hutton Institute of Nuclear Medicine University College London Emission Tomography Principles and Reconstruction.
O AK R IDGE N ATIONAL LABORATORY U.S. DEPARTMENT OF ENERGY Image Reconstruction of Restraint-Free Small Animals with Parallel and Multipinhole Collimation:
Ultrasound, Positron Emission Tomography, and Single Photon Emission Tomography Allen T. Newton, Ph.D. PAVE 2014.
Li HAN and Neal H. Clinthorne University of Michigan, Ann Arbor, MI, USA Performance comparison and system modeling of a Compton medical imaging system.
Nuclear Medicine: Tomographic Imaging – SPECT, SPECT-CT and PET-CT Katrina Cockburn Nuclear Medicine Physicist.
A Single Photon Emission Computer Tomograph for breast cancer imaging S. Vecchio a, N. Belcari a, P. Bennati b, M. Camarda a, R. Campanini c, M. N. Cinti.
Part No...., Module No....Lesson No
Conclusions Simulated fMRI phantoms with real motion and realistic susceptibility artifacts have been generated and tested using SPM2. Image distortion.
Single-Slice Rebinning Method for Helical Cone-Beam CT
Introduction In positron emission tomography (PET), each line of response (LOR) has a different sensitivity due to the scanner's geometry and detector.
Ultrasound Computed Tomography 何祚明 陳彥甫 2002/06/12.
Introduction In Positron Emission Tomography (PET), each line of response (LOR) has a different sensitivity due to the scanner's geometry and the detector's.
Profiling study of Geant4-GATE Discussion of suggested VRTs Nicolas Karakatsanis, George Loudos, Arion Chatziioannou Geant4-GATE technical meeting 12 th.
Improved Hybrid PET Imaging Using Variable Focal Length Fan-Slat Collimators Thomas C. Rust and Dan J. Kadrmas, Ph.D. Medical Imaging Research Laboratory.
1 Nuclear Medicine SPECT and PET. 2 a good book! SR Cherry, JA Sorenson, ME Phelps Physics in Nuclear Medicine Saunders, 2012.
Wai-Hoi Wong 1, Hongdi Li 2, Yuxuan Zhang 1, Rocio Ramirez 1, Hossain Baghaei 1, Shaohui An 2, Chao Wang 2, Shitao Liu 2, Yun Dong 2 1 University of Texas.
RPC February 2010 SPATIAL RESOLUTION OF HUMAN 3D RPC-PET SYSTEM 1 LIP, Laboratório de Instrumentação e Física Experimental de Partículas, Coimbra,
Software for Tomographic Image Reconstruction
Chapter-4 Single-Photon emission computed tomography (SPECT)
Computed Tomography Computed Tomography is the most significant development in radiology in the past 40 years. MRI and Ultrasound are also significant.
JESÚS SILVA-RODRÍGUEZ, PABLO AGUIAR, INÉS DOMÍNGUEZ-PRADO, MICHEL HERRANZ, ÁLVARO RUIBAL 18F-Choline: Is shine-through effect an issue for prostate SUV.
The Optimization of Reconstruction Method Reducing Partial Volume Effect in PET/CT 3D Image Acquisition Department of Nuclear Medicine, Samsung Medical.
Simulations in Medical Physics Y. TOUFIQUE*, R.CHERKAOUI EL MOURSLI*, M.KACI**, G.AMOROS**, *Université Mohammed V –Agdal, Faculté des Sciences de Rabat,
Hybrid Head motion correction in PET/MR Brian Imaging
P.E.T. Positron Emission Tomography
Positron emission tomography: a review of basic principles, scanner design and performance, and current systems  Pat Zanzonico  Seminars in Nuclear Medicine 
CASToR: Evalutation of an Efficient Generic Data Organization and Processing Code Framework for Multi-Modal and Multi-Dimensional Tomographic Reconstruction.
Prof. Stefaan Vandenberghe Enrico Clementel
SIMULATION AND IMAGE RECONSTRUCTION OF CLINICAL TOF-PET SCANNERS
Evaluation of mA Switching Method with Penalized Weighted Least-Square Noise Reduction for Low-dose CT Yunjeong Lee, Hyekyun Chung, and Seungryong Cho.
Computed Tomography Image Reconstruction
Reconstructions with TOF for in-beam PET
Detailed simulations of a full-body RPC-PET scanner
2017 Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine Theoretical Study of Benefit of Long Axial Field-of-View PET: Impact.
Basic principles Geometry and historical development
Volume 11, Issue 2, Pages (April 2016)
How can we study the magnetic distortion effect?
Presentation transcript:

Impact of Axial Compression for the mMR Simultaneous PET-MR Scanner Martin A Belzunce, Jim O’Doherty and Andrew J Reader King's College London, Division of Imaging Sciences, St Thomas’ Hospital, London, UK Purpose To evaluate the impact on the spatial resolution of reconstructing images without axial compression in comparison to the standard span- 11 reconstruction in the Siemens Biograph mMR scanner. In addition, we aim to assess the spatial resolution for a number of span values and the effect of modeling the axial compression in the system matrix. Introduction In 3D PET, an axial compression is used to reduce the data size and the computation times during the image reconstruction. This is achieved by averaging a set of sinograms with adjacent values of the oblique polar angle. This sampling scheme achieves good results in the centre of the FOV. However, there is a loss in the radial, tangential and axial resolutions at off-centre positions, which is increased in scanners with large FOVs. The Siemens Biograph mMR was the first simultaneous whole body PET-MR and it has one of the largest FOV commercially available. The system uses a sinogram-based reconstruction with an axial compression of span-11 and a component-based normalization. Additionally, for each acquisition the scanner stores a span-1 sinogram for both the prompt and delayed events. This feature brings the possibility of implementing an image reconstruction algorithm without axial compression that would improve the spatial resolution in the outer regions of the FOV, which could benefit whole body studies. In this work, we study the impact of using an image reconstruction algorithm without axial compression and the use of higher span values modeled in the system matrix. 2) Estimate an axial profile for Incident events (not uniform: scatter, attenuation) using a normalized profile for each crystal position within a block. Correct the detected profile. 3) Estimate the detection efficiency Block Profile using the corrected profile from 2). The efficiency of each axial position within a block is: 1) Total counts in the direct sinograms (detected profile) for a 5-hour 68 Ge phantom scan. 4) The axial factor for each sinogram plane of the span-1 sinogram is computed using the Block Profile: Materials and Methods Span-1 Normalization Factors Conclusions A reconstruction algorithm without axial compression was im- plemented for the Biograph mMR and its general properties were verified using a NEMA IQ phantom acquisition and using STIR as a reference. Using a line source, we showed that a meaningful improvement in resolution in the x and z axes is achieved when there is not axial compression. Including the axial compression in the system matrix can compensate the spatial resolution degradation at the cost of a span-1 forward projection and an increase in the noise for higher span values. References [1] B. Bendriem and D. W. Townsend, The Theory and Practice of 3D PET. Springer,1998. [2] G. Delso, S. Frst et al, ”Performance Measurements of the Siemens mMR Integrated Whole-Body PET/MR Scanner,” Journal of Nuclear Medicine, vol. 52, pp , December 01, [3] R. D. Badawi and P. K. Marsden, ”Developments in component-based normal- ization for 3D PET,” Phys. Med. Biol., vol. 44, pp , [4] K. Thielemans, C. Tsoumpas et al, STIR: Software for Tomographic Image Reconstruction Release 2,” Physics in Medicine and Biology, 57 (4), 2012 pp Span-1 Implementation Modeling Axial Compression in the System Matrix Impact on the Spatial Resolution The Siemens mMR scanner uses a component-based normalization. The standard reconstruction uses span-11 sinograms and therefore the normalization factors also are designed for span-11. In order to generate span-1 normalization factors, the CBN components need to be adapted to that sinogram size. A 5-hours scan of the 68 Ge uniform phantom was used to create new axial factors. The Siemens mMR scanner. It was the first simultaneous whole body PET-MR. It is composed of 8 rings of 56 detector blocks of 8x8 LSO crystal elements. The standard image reconstruction for this scanner uses 3D span-11 sinograms, with 837 sinogram planes of 344 [radial coordinate] x 252 [angle projection] bins. The system has a transaxial FOV of 59.4 cm and an axial FOV of 25.8 cm. The scanner also permits storage of span-1 sinograms. A span-1 image reconstruction algorithm was implemented (OP-OSEM). Normalization factors for this sinogram size were also generated. An assessment of the benefits of the span-1 implementation was performed using real and simulated data. The incorporation of the axial compression was included in the forward projection of the image reconstruction and compared to the span-1 reconstructions The implementation was validated evaluating the image properties of a NEMA Image Quality (IQ) phantom scan using a method capable of assessing image quality even if scatter correction is not applied to the data. 3D sinograms were simulated by projecting a NCAT and Defrise phantom and introducing them Poisson noise. The impact of the span-1 reconstruction on the spatial resolution was evaluated using a line source scan. 5) The span-1 axial factors can be converted into any span. A 3D OP-OSEM reconstruction algorithm was implemented in a C++/CUDA library for any span. A Siddon’s projector/backprojector was used. X span : X-ray transform for sinograms with span s. X 1 : X-ray transform for span-1 sinograms. C: axial compression to span s. ML-EM with axial compression in the forward model: Results for Defrise Phantom Results for NCAT Phantom We compute the SNR in the images using a spatially variant estimate of the mean value in the background of the phantom. Evolution of STD vs CRC with iterations Profiles in the z axis of the reconstructed images for Defrise phantoms with and without modeling the axial compression in the forward projection Coronal Planes for Span-17 Modeled Axial Compression Standard Reconstruction Profile centered in the lesion of the NCAT phantom with and without modeling the axial compression in the forward projection Contrast Recovery in Lesion Std Dev in Uniform Region Span-51 Span-11 modeled Span-11 Span-51 modeled Span-1 Simulated Defrise Phantom Simulated NCAT Phantom In order to validate the algorithm, a scan of the NEMA phantom was reconstructed using 1 subset (ML-EM) and 60 iterations for the span 11 and span 1 sinograms, and with STIR. A line source scan was used to evaluate the spatial resolution for the span-1 reconstruction and for a number of other span values with and without modeling the axial compression in the forward model. The FWHM in each direction was computed for each transverse, coronal and sagital plane of the reconstructed images FWHM in x in the transverse slices Effect on the FWHM of the span modelling for a number of span values Maximum intensity projections in the coronal plane