LIP Laboratory of Instrumentation and Experimental Particles Physics

Lisbon, Portugal

LIP Laboratory of Instrumentation and Experimental Particles Physics

Lisbon, Portugal
SEARCH FILTERS
Time filter
Source Type

Cao L.,German Cancer Research Center | Bugalho R.,LIP Laboratory of Instrumentation and Experimental Particles Physics | Ferreira C.S.,LIP Laboratory of Instrumentation and Experimental Particles Physics | Ferreira C.S.,University of Lisbon | And 4 more authors.
IEEE Transactions on Nuclear Science | Year: 2014

A fast list-mode MLEM reconstruction algorithm including low-noise random correction for the Clear-PEM imaging system is presented. In this approach, the sensitivity matrix is estimated based on planar phantom acquisitions, rather than tracing along lines of response (LORs). The algorithm is implemented by distributing the calculation of the acquired list-mode events into different threads yielding significantly increased computational speed. A list-mode random correction approach is presented in which reconstructed data are corrected according to a smooth correction image estimated from delayed coincidence data. Monte Carlo simulations based on GATE were performed to validate the proposed approach. A Derenzo phantom study shows that the 1.2-mm hot rods can be resolved using the presented algorithm. A breast-torso phantom study as well as a study based on acquired patient data demonstrate its feasibility in a clinical environment. This method reaches a significant higher calculation efficiency in which the total reconstruction time is less than 30 seconds for a typical clinical study. Results show that bias from random coincidences is largely suppressed. The presented algorithm achieves a fast and low-noise reconstructions without compromising intrinsic system resolution and sensitivity. © 1963-2012 IEEE.


Almeida I.P.,University of Lisbon | Ferreira N.C.,University of Coimbra | Ortigao C.,LIP Laboratory of Instrumentation and Experimental Particles Physics | Bugalho R.,LIP Laboratory of Instrumentation and Experimental Particles Physics | Varela J.,LIP Laboratory of Instrumentation and Experimental Particles Physics
2014 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2014 | Year: 2014

We aim at developing practical ways to measure the parameters necessary for data corrections and calibration of a PET scanner with variable geometry (ClearPEM), in order to achieve accurate quantitative images for a large range of imaging tasks using few phantom measurements. We used a decaying cylindrical phantom to determine calibration and dead time factors for a set of acquisition geometries that was obtained by varying the distance between detectors. New factors for intermediate distances can then be obtained by interpolation. Normalization correction factors were measured with a planar source of activity using the same distances. The results allowed to assess the dead time correction and calibration factors, indicating that with two simple phantom acquisitions it is possible to calibrate the scanner and obtain the data needed to apply data corrections in order to obtain quantitative images for a large range of imaging applications. © 2014 IEEE.


Cao L.,German Cancer Research Center | Bugalho R.,LIP Laboratory of Instrumentation and Experimental Particles Physics | Ferreira C.S.,LIP Laboratory of Instrumentation and Experimental Particles Physics | Ferreira C.S.,University of Lisbon | And 3 more authors.
IEEE Nuclear Science Symposium Conference Record | Year: 2013

The Clear-PEM system is a dedicated PET scanner optimized for breast imaging. Because of the limited field-of-view (FOV) and the moderate timing resolution, the imaging results of the system might be degraded by random noise. This noise is more pronounced at the region that is near to the torso, which could cause false-positive or inconclusive diagnostics. Because of the high number of lines of responses (LORs) comparing to the number of acquired coincidences, list-mode reconstruction is required to maintain efficiency and accuracy. A new acquisition strategy is presented in this abstract in order to largely increase the statistics of acquired random events, without the requirement of hardware to collect single counts. During data acquisition, a large coincidence window of 90 ns is set in the readout electronics. All data within this window are collected in list-mode and, afterwards, classified by the acquisition software into prompt counts (0-4 ns), ignored counts (4-20 ns) and random counts (20-90 ns). A smooth correction image is estimated using those collected random counts. Reconstruction is afterwards performed considering the correction image with multiplication of the ratio between coincidence window width and random window width. An experimental study was performed on a breast-torso phantom. Results show that this approach can increase the statistics of recorded random coincidences over 17 folds, leading to a pronounced improvement of random correction effect. Because of the adoption of a 90 ns coincidence trigger in the hardware that can filter out most of the counts, it yields a much less computational burden to the acquisition system in comparison to the correction method using single count rate. © 2013 IEEE.


Cao L.,German Cancer Research Center | Bugalho R.,LIP Laboratory of Instrumentation and Experimental Particles Physics | Matela N.,University of Lisbon | Martins M.,University of Lisbon | And 3 more authors.
IEEE Nuclear Science Symposium Conference Record | Year: 2012

A dedicated implementation of list-mode maximum-likelihood expectation-maximization (MLEM) reconstruction for the ClearPEM system is presented. The system is composed of two face-to-face detectors, which can be rotated to acquire data from different angular positions. Due to the specific design with irregular sampling and depth of interaction capability, the possible number of lines of response (LOR) is significantly greater than the number of detected events in a standard clinical study. Because reconstruction methods based on data histogramming to sinogram lead to a high computational cost and/or a loss of the intrinsical system resolution, it is necessary to consider the processing of events in list-mode during the reconstruction. The presented method adopted EM algorithm to maximize the logarithmic likelihood function that is expressed in list-mode. The voxel efficiency is corrected by pre-calculated efficiency maps based on flood phantom acquisitions. The method is also implemented with parallelization by distributing the calculation of the acquired events into different threads for significantly increasing computational speed. The results of a Derenzo phantom study show that the presented algorithm can achieve a similar result as 3D-OSEM reconstruction based on data histogramming with significantly lower reconstruction time (6 times faster with one thread, 20 times faster with 8 threads distributed in 8 CPU cores). In clinical studies with lower acquired events, the acceleration ratio can be even higher. The result from a breast phantom study shows that lesions with 15 mm in diameter, each, as well as a small lesion with 5 mm in diameter are clearly visible and can be characterized. The mouse imaging studies show also great potential of the system in small animal applications. © 2011 IEEE.


Cao L.,German Cancer Research Center | Bugalho R.,LIP Laboratory of Instrumentation and Experimental Particles Physics | Ortigao C.,LIP Laboratory of Instrumentation and Experimental Particles Physics | Varela J.,LIP Laboratory of Instrumentation and Experimental Particles Physics | Peter J.,German Cancer Research Center
IEEE Nuclear Science Symposium Conference Record | Year: 2012

A dedicated random correction algorithm is presented in this study for positron emission mammography (PEM) systems. PEM refers to a specified PET system that is optimized for breast imaging by its small FOV. Clinical imaging results from such systems, however, may be degraded by strong statistical noise caused from random coincidences, especially in the region that is near to the torso, due to the high amount of activity uptake outside the FOV, the low geometrical sensitivity of the detector elements near to the torso, and the large solid angle acceptance of random coincidence events. Because of the low statistics of detected coincidence events against the extremely high number of LORs, list-mode reconstruction algorithms are suggested for PEM systems. The conventional random correction methods cannot be directly implemented or can induce an even higher statistical noise. The correction method by single count rate requires a high hardware cost to record single events and needs an accurate calibration to reach a non-bias correction. The new random correction algorithm presented in this study can be implemented into list-mode reconstruction without single count acquisition. This algorithm estimates in a first step a smooth correction image with the delayed coincidences data. This correction image is then used to estimate the mean random coincidence rate for each detected event during the iterative list-mode reconstruction routine. The approach is tested on a ClearPEM system developed by the Crystal Clear Collaboration. Experimental data are acquired by two face-to-face detectors at four angular positions with a total acquisition time of 20 min for each breast. Results show that the proposed algorithm can largely suppress the statistical noise in the region near the torso. © 2012 IEEE.

Loading LIP Laboratory of Instrumentation and Experimental Particles Physics collaborators
Loading LIP Laboratory of Instrumentation and Experimental Particles Physics collaborators