M4B1  Student Paper Competition

Thursday, Nov. 5  10:30-12:30  Golden Pacific Ballroom

Session Chair:  R.Glenn Wells, University of Ottawa Heart Institute, Canada; Patrick La Riviere, The University of Chicago, United States

Show/Hide All Abstracts

(10:30) M4B1-1, Development of a Multi-Knife-Edge Slit Collimator for Prompt Gamma Imaging During Proton Beam Therapy

J. Ready1, R. Pak1, L. Mihailescu2, K. Vetter1,2

1Department of Nuclear Engineering, UC Berkeley, Berkeley, CA, USA
2Applied Nuclear Physics, Lawrence Berkeley National Laboratory, Berkeley, CA, USA
To realize the maximum potential benefits of proton therapy, the precise location of the Bragg peak must be measured. This work presents a new method of prompt gamma verification of proton beam therapy using a novel aperture-based imaging system. This collimated system will provide 2-dimensional imaging capability for verification of proton beam range and Bragg peak dose via prompt-gamma detection. The imaging system consists of a multi-knife-edge slit collimator paired with an array of LSO scintillation detectors which have been designed, constructed, and characterized via simulations and experimental measurements. Initial simulations were performed using the TOPAS Geant4-based Monte Carlo package. Iterative reconstruction methods were combined with simulated point response functions to characterize the imaging performance of the system. Experimental characterization was performed using 2.6 MeV gamma-rays from a Th-228 source. Both simulation and experimental results indicate that this collimated system provides 2-D imaging capability in the energy range of interest for prompt-gamma dose verification. In the current configuration, with collimator-to-source distance of 13 cm, image reconstruction of point sources resulted in spatial resolution (FWHM) of approximately 4 mm in both x- and y-directions in the imaging plane. The accuracy of positioning the point source peak is less than 1 mm. The multi-slit pattern is designed to increase detection efficiency and provide spatial information in 2-dimensions -- an improvement over a single-slit collimator design. The thickness and density of the collimator will allow this detection system to perform well in an environment with high gamma flux, while ultimately providing peak determination accuracy on the order of 1 mm.

(10:45) M4B1-2, Measurements of the Time Spread of Proton Pencil Beams at a Clinical Therapy Facility

J. Petzoldt1, K. E. Roemer2, T. Kormoll1, W. Enghardt1, F. Fiedler2, S. Helmbrecht2, F. Hueso-Gonzalez2, C. Golnik1, H. Rohling2, T. Werner3, G. Pausch1

1OncoRay - National Center for Radiation Research in Oncology, Dresden, Germany
2Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany
3Technische Universitaet Dresden, Dresden, Germany
In proton therapy, the finite range of the projectiles is exploited to reduce the dosage to healthy tissue while increasing the dose inside the tumor volume compared to conventional radiotherapy. However, those benefits can be diminished by range uncertainties. An online range verification and in-vivo dosimetry is therefore highly desired. The prompt gamma ray timing (PGT) method utilizes the detection time of high energetic photons emitted during treatment with respect to the cyclotron radio frequency. The time distribution of the gamma rays contains essential information about the range of the protons. However, PGT spectra are smeared out by the time spread of the proton bunches. Knowledge about this time spread would help to disentangle the PGT measurementsand to give better input parameters to simulation procedures. At the UniversitaetsProtonenTherapieDresden (Dresden, Germany), a dedicated experiment was realized to measure the time spread of a clinical proton pencil beam created by a Cyclone 230 fixed-energy cyclotron from Ion Beam Applications. Two phoswich detectors, each composed of plastic scintillator and BGO, were placed under 90 degrees to detect coincident protons originating from elastic pp-scattering at a thin slice of PMMA. The time spread was measured for incident proton energies between 69 MeV and 225 MeV, as well as for several positions of the momentum limiting slits of the energy selection system. Additionally, the absolute transmission of protons from point of extraction to beam exit was determined. Summarizing, the measured data will help to create reliable range verification procedures using the PGT method in clinical routine.

(11:00) M4B1-3, Initial Results of Applying Automatic Channel Fault Detection and Diagnosis on Small Animal APD-Based Digital PET Scanners

J. Charest1, J.-F. Beaudoin2, M. Bergeron2, L. Arpin1, J. Cadorette2, R. Lecomte2, C.-A. Brunet3, R. Fontaine1

1Institut Interdisciplinaire d'Innovation Technologique (3IT), Universite de Sherbrooke, Sherbrooke, Quebec, Canada
2Department of Nuclear Medicine and Radiobiology, Universite de Sherbrooke, Sherbrooke, Quebec, Canada
3Department of Electrical and Computer Engineering, Universite de Sherbrooke, Sherbrooke, Quebec, Canada
Optimal image quality in small animal positron emission tomography (PET) is critical to ensure accuracy and reliability of results obtained in biological studies. Indeed, unstable image quality over time can jeopardize longitudinal studies. This is why quality control (QC) procedures are of the utmost importance in order to keep PET scanners at an optimal performance level. Unfortunately, as the scanner technology evolves increasing the number of acquisition channels, so does the scanner operator's effort to keep up with adequate QC procedures. With scanners using one-to-one crystal to photodetector coupling to achieve enhanced spatial resolution and contrast to noise ratio (CNR), the QC workload rapidly increases to unmanageable levels due to the number of independent channels involved. An intelligent system (IS) was proposed to help reduce the QC workload by performing automatic channel fault detection and diagnosis. The IS consists of four high-level modules that employ machine learning methods to perform their tasks: Parameter extraction, Fault detection, Fault prioritization and Fault diagnosis. Ultimately, the IS presents a prioritized list to the operator containing the faulty channels and proposes actions that should be taken to correct them. To validate that the IS can perform QC procedures with minimal operator intervention, it was deployed on a LabPET scanner in Sherbrooke and image quality metrics were extracted before and after the channel corrections proposed by the IS where applied. After a single iteration of corrections on sub-optimal scanner settings, a 6.3 % increase in the CNR was observed as well as a 7.0 % decrease of the uniformity percentage standard deviation. These results indicate that the IS can improve scanner performance and further iterations are expected to make the scanner converge towards optimal settings.

(11:15) M4B1-4, Limited Data CT Reconstructions with a Prior-Image-Guided, Adaptive TV Penalty

D. S. Rigie, P. J. La Riviere

Radiology, University of Chicago, Chicago, IL, United States
We present a spatially adaptive form of the total-variation (TV) penalty that incorporates edge information from a prior image and demonstrate its utility in reconstructing various kinds of limited data. Specifically, we investigate sparse-view, limited-angle, and interior reconstruction problems that arise in some recently proposed spectral CT geometries. In this work, we conduct an inverse-crime study to compare the proposed penalty to the conventional, isotropic total variation. Though we do not achieve exact recovery, the images reconstructed with the proposed, prior-image-guided penalty have substantially smaller errors than those obtained without the prior image using an ordinary TV penalty. This prior-image-guided reconstruction approach may facilitate more flexibility in spectral CT system design, due to its ability to overcome severe data limitations.

(11:30) M4B1-5, The ClearPET/XPAD Prototype: Development of a Simultaneous PET/CT Scanner for Mice

M. Hamonet1, M. Dupont1, T. Fabiani1, F. Cassol1, Y. Boursiser1, A. Bonissent1, F. Debarbieux2, L. Bidaut3, C. Morel1, G. Pottier4

1CPPM CNRS/IN2P3, Aix-Marseille Université, Marseille, France
2INT CNRS, Aix-Marseille Université, Marseille, France
3CRIF, University of Dundee and Ninewells Hospital, Scotland, UK
4Inserm/CEA, Universite Paris Sud, Paris, France
The combination of Positron Emission Tomography (PET) and X-ray Computerized Tomography (CT) for PET/CT imaging has been an essential line of research in the previous decades and led to a rapid expansion of this technique in clinics and preclinics. However, even if the two modalities can be juxtaposed on the same gantry, the original concept invented by Dave Townsend foresees to do both PET and CT imaging at the same time, while imaging a common Field-of-View. This will allow to know the exact position of the animal during the scan and possibly to correct for the animal movements. Therefore we developed the ClearPET/XPAD prototype: the first simultaneous PET/CT scanner for mice combining on the same rotating gantry ClearPET modules and the hybrid pixels camera XPAD3 facing an X-ray source. We present the first simultaneous PET/CT scans of 68Ge and 22Na point sources, of the micro Derenzo phantom filled with 6.5 MBq [18F]FDG and a solution of iodine to enhance CT contrast, and of a mouse injected retro-orbitally with 16 MBq [18F]FDG. These first images demonstrates the feasibility to acquire simultaneously PET and CT data from a common Field-of-View as Dave Townsend’s invention of PET/CT arose originally from the observation of empty spaces between the detector arrays in the first rotating positron tomograph (PRT-1) that he developed in the early 90s in Geneva.

(11:45) M4B1-6, A Joint Estimation Method for Kinetic Modeling of Simultaneously Acquired PET/MRI Signals

M. Q. Wilks, G. El Fakhri, N. M. Alpert, Q. Li

Center For Advanced Medical Imaging Sciences, Massachusetts General Hospital, Boston, MA, USA
The advent of combination PET/MRI now allows for data acquired simultaneously from multiple modalities to be modeled in a unified manner, and for model parameters to be jointly estimated. Such estimation methods have the potential to greatly increase utility of image derived kinetics, as joint optimization schemes can help overcome low signal-to-noise regimes, as measurement and reconstruction error will be independent between modalities. Additionally, the increased measurements made in a dual-modality scanner can allow for more complicated models to be used, and accurate estimates to be made on previously unidentifiable parameters. In this work we propose and test such a joint estimation method based around the Alternating Direction Method of Multipliers (ADMM) optimization scheme. This method is broadly applicable to many multi-modality imaging protocols, under the sole constraint that there exists a non-empty set of parameters in each of the models that have a defined (non-trivial) linear relationship with one another. As a test case of this method we have used the relatively simple model of blood flow imaging on a PET/MR scanner of 15O-H20 and gadolinium kinetics. Dynamic PET/MR measurements were simulated with Gaussian noise and fit either with standard, independent, cost-functions applied to the simulated PET and MR data, or with an ADMM-based joint estimation protocol. The joint estimation method reduced the variance and bias of several important flow parameters over multiple (n=500) noise realizations, at marginal increases in computational time.

(12:00) M4B1-7, Imaging Performance of a Proof-of-Concept Prototype Time-of-Flight PET System Based on High-Quantum-Efficiency Multi-Anode PMTs

J.-W. Son, K. Y. Kim, H. S. Yoon, J. Y. Won, G. B. Ko, M. S. Lee, J. S. Lee

Department of Nuclear Medicine, Seoul National University, Seoul, South Korea
We present the design and the imaging performance of a proof-of-concept prototype time-of-flight (TOF) positron emission tomography (PET) scanner based on the advanced high-quantum-efficiency (33 % at 420 nm) position-sensitive multi-anode (8×8) photomultiplier tubes (MA-PMTs) equipped with super-bialkali photocathode (H10966A-100; Hamamatsu Photonics K. K., Hamamatsu, Japan) coupled with small cross-sectional (3 × 3 × 20 mm3) LGSO scintillation crystals and a FPGA-based time-to-digital converter with 14 ps measurement precision. The number of detector elements of the prototype system is 9000 and the transaxial and axial field-of-views of the scanner are about 52 cm and 4.6 cm, respectively. The intrinsic performances of the scanner are as follows: 340 ± 45 ps (full-width-at-half-maximum (FWHM)) coincident resolving time, 11.5 ± 0.9 % (FWHM) energy resolution, and 5.3 ± 1 distance-to-width ratio.
We acquired images of a NEMA IEC body phantom (sphere-to-background ratio of 8:1) and calculated the contrast recovery coefficient and background variability of each sphere to quantitatively evaluate and compare the imaging performance with and without TOF information. We also acquired images of a 3D Hoffman brain phantom for qualitative assessment. All data were corrected for random, normalization, scatter and attenuation. Images were reconstructed with a listmode ordered-subset-expectation-maximization algorithm.
The images showed excellent quality and better results (both qualitatively and quantitatively) were obtained with TOF information. We demonstrated that the developed PET system has superior imaging performance — mainly due to its fine time, energy, and spatial resolution achieved with the advanced high QE MA-PMTs. As the next generation TOF PET scanners require concurrent measurement of DOI information without compromising the accuracy of TOF assessment, we are making an attempt to add the DOI encoding capability to the scanner.

(12:15) M4B1-8, Coupled Motion and Activity Estimation from PET and MR Data with Motion Model Based Parameter Reduction

D. R. Balfour, C. Kolbitsch, A. J. Reader, A. P. King, P. K. Marsden

Imaging Sciences and Biomedical Engineering, King's College London, London, London, United Kindom
We propose a technique that estimates both the image and the respiratory motion states of all PET gates during PET reconstruction. This is done using an analytical gradient derived from the Poisson log-likelihood involving a motion model. This is incorporated into reconstruction using an motion-correction image reconstruction (MCIR) framework coupled with optimisation of the analytical gradient. The use of a motion model in this way reduces the number of additional parameters to be optimised for motion correction to 1 per PET gates. This was tested with realistic PET simulations and real volunteer motion data, with 5, 10, and 15 PET gates. The proposed method successfully recovered lesion SUV peak and shape in all cases.