Computer-aided staging of lymphoma patients with FDG PET/CT imaging based on textural information

We have designed a computer-aided diagnosis system to discriminate between hypermetabolic cancer lesions and hypermetabolic inflammatory or physiological but noncancerous processes in FDG PET/CT exams of lymphoma patients. Detection performance of the support vector machine (SVM) classifier was assessed based on feature sets including 105 positron emission tomography (PET) and Computed tomography (CT) characteristics derived from the clinical practice and from more sophisticated texture analysis. An original feature selection method based on combining different filter methods was proposed. The evaluation database consisted of 156 lymphomatous and 32 suspicious but nonlymphomatous regions of interest. Different types of training databases including either the PET and CT features or the PET features only, with or without feature selection, were evaluated to assess the added value of multimodality and texture information on classification performance. An optimization study was conducted for each classifier separately to select the best combination of parameters. Promising classification performance was achieved by the SVM classifier combined with the 12 most discriminant PET and CT features with a value of the area under the receiver operating curve of 0.91.

Comparison of bootstrap resampling methods for 3-D PET imaging

Two groups of bootstrap methods have been proposed to estimate the statistical properties of positron emission tomography (PET) images by generating multiple statistically equivalent data sets from few data samples. The first group generates resampled data based on a parametric approach assuming that data from which resampling is performed follows a Poisson distribution while the second group consists of nonparametric approaches. These methods either require a unique original sample or a series of statistically equivalent data that can be list-mode files or sinograms. Previous reports regarding these bootstrap approaches suggest different results. This work compares the accuracy of three of these bootstrap methods for 3-D PET imaging based on simulated data. Two methods are based on a unique file, namely a list-mode based nonparametric (LMNP) method and a sinogram based parametric (SP) method. The third method is a sinogram-based nonparametric (SNP) method. Another original method (extended LMNP) was also investigated, which is an extension of the LMNP methods based on deriving a resampled list-mode file by drawings events from multiple original list-mode files. Our comparison is based on the analysis of the statistical moments estimated on the repeated and resampled data. This includes the probability density function and the moments of order 1 and 2. Results show that the two methods based on multiple original data (SNP and extended LMNP) are the only methods that correctly estimate the statistical parameters. Performances of the LMNP and SP methods are variable. Simulated data used in this study were characterized by a high noise level. Differences among the tested strategies might be reduced with clinical data sets with lower noise.

Characterization of neuro-endocrine tumors in an athymic nude mouse model using dedicated synchronization strategies for T2-weigted MR imaging at 7T

Endocrine tumours, with digestive localization, are tumours with variable forecast which are independent of their invasive and metastatic extensions. The experimental model of endocrine tumors with liver dissemination is available for evaluation of new medical therapeutics such as antiangiogenic therapy. MRI is a non invasive modality allowing in vivo examinations and is suitable to follow liver lesion evolution during longitudinal study on animal models. The goal of this study was to assess the detection level and to characterize the liver lesions in an athymic nude mouse model, using a dedicated MRI protocol and an optimized synchronization strategy at high magnetic field strength. The experiments were performed at 7T. To detect liver lesions, respiratory-triggered T2-weighted MR images is the sequence of choice. With conventional acquisition strategies used on small animal MR systems, trigger signal is performed at each respiratory cycle and thus, the T2 contrast is not freely controlled. Additionally, the slice number is limited by the expiration delay. To overcome these drawbacks, we proposed an original strategy enabling true T2-weighted imaging with minimal movement artifacts, regardless of the respiratory period and the number of slices. This protocol was used to carry out a longitudinal follow-up of hepatic lesions in 8 nude mice at stages D7, D12, D17 and D24. The fraction of lesion over the total liver volume was quantified. Moreover, the characterization of cystic or non-cystic type of lesions was achieved using various TE leading to T2 maps. In conclusion, the level of lesion detection and characterization of liver lesions was performed using a devoted protocol with original synchronization strategy dedicated to high field MRI. MR imaging could be used with relevance in the evaluation of new therapeutics protocol for treatment of liver lesions in neuroendocrine tumors using small animal model.

Validation of PET-SORTEO Monte Carlo simulations for the geometries of the MicroPET R4 and Focus 220 PET scanners

PET-SORTEO is a Monte Carlo-based simulator that enables the fast generation of realistic PET data for the geometry of the ECAT EXACT HR+ scanner. In order to address the increasing need for simulation models of animal PET imaging systems, our aim is to adapt and configure this simulation tool for small animal PET scanners, especially for the widely distributed microPET R4 and Focus 220 systems manufactured by Siemens Preclinical Solutions. We propose a simulation model that can produce realistic rodent images in order to evaluate and optimize acquisition and reconstruction protocols. The first part of this study presents the validation of SORTEO against the geometries of the R4 and the Focus 220 systems. This validation is carried out against actual measurements performed on the R4 scanner at the Montreal Neurological Institute in Canada and on the Focus 220 system of the department of radiopharmaceuticals of the Austrian Research Center in Seibersdorf. The comparison of simulated and experimental performance measurements includes spatial resolution, energy spectra, scatter fraction and count rates. In the second part of the study, we demonstrate the ability to rapidly generate realistic whole-body radioactive distributions using the MOBY phantom and give comparative example case studies of the same rodent model simulated with PET-SORTEO for the R4 and Focus 220 systems.

Evaluation of Multiclass Model Observers in PET LROC Studies

A localization ROC (LROC) study was conducted to evaluate nonprewhitening matched-filter (NPW) and channelized NPW (CNPW) versions of a multiclass model observer as predictors of human tumor-detection performance with PET images. Target localization is explicitly performed by these model observers. Tumors were placed in the liver, lungs, and background soft tissue of a mathematical phantom, and the data simulation modeled a full-3D acquisition mode. Reconstructions were performed with the FORE+AWOSEM algorithm. The LROC study measured observer performance with 2D images consisting of either coronal, sagittal, or transverse views of the same set of cases. Versions of the CNPW observer based on two previously published difference-of-Gaussian channel models demonstrated good quantitative agreement with human observers. One interpretation of these results treats the CNPW observer as a channelized Hotelling observer with implicit internal noise.

Assessment of non-reperfused and reperfused myocardial infarction using diffusible or deposited radiolabelled perfusion imaging agents

PURPOSE

Incomplete microvascular reperfusion is often observed in patients undergoing thrombolytic therapy or angioplasty for acute myocardial infarction and has important prognostic implications. We compared the myocardial uptake of diffusible ((201)Tl) and deposited ((99m)TcN-NOET) perfusion imaging agents in the setting of experimental infarction.

METHODS

Rats were subjected to permanent coronary occlusion (OCC, n=10) or to 45-min occlusion and reperfusion (REP, n=17). Seven days later, the tracers were co-injected and the animals were euthanised 15 min (all ten rats in the OCC group and 12 rats in the REP group) or 120 min (five rats from the REP group, euthanised at this time point to evaluate any redistribution of the tracers: REP-RED group) afterwards. Infarct size determination and (99m)TcN-NOET/(201)Tl ex vivo imaging were performed. Regional flow and tissue oedema were quantified using radioactive microspheres and (99m)Tc-DTPA, respectively.

RESULTS

(99m)TcN-NOET and (201)Tl defect magnitudes were similar in OCC animals (0.11+/-0.01 vs 0.13+/-0.01). In REP animals, (201)Tl defect magnitude (0.25+/-0.02) was significantly lower than the magnitude of (99m)TcN-NOET and flow defects (0.14+/-0.03 and 0.17+/-0.01, respectively; p<0.05), despite the lack of (201)Tl redistribution (REP-RED animals). (99m)Tc-DTPA indicated the presence of oedema in the reperfused area. Blood distribution studies showed that, unlike (99m)TcN-NOET, (201)Tl plasma activity was mostly unbound to plasma proteins.

CONCLUSION

(99m)TcN-NOET and (201)Tl delineated the non-viable area in chronic non-reperfused and reperfused myocardial infarction. The significantly decreased (201)Tl defect in reperfused infarction was likely due to partial diffusion of the tracer from the plasma into the oedema present in the infarcted area. Deposited perfusion tracers might be better suited than diffusible agents for the assessment of regional flow following reperfusion of myocardial infarction.

GATE: a simulation toolkit for PET and SPECT

Monte Carlo simulation is an essential tool in emission tomography that can assist in the design of new medical imaging devices, the optimization of acquisition protocols and the development or assessment of image reconstruction algorithms and correction techniques. GATE, the Geant4 Application for Tomographic Emission, encapsulates the Geant4 libraries to achieve a modular, versatile, scripted simulation toolkit adapted to the field of nuclear medicine. In particular, GATE allows the description of time-dependent phenomena such as source or detector movement, and source decay kinetics. This feature makes it possible to simulate time curves under realistic acquisition conditions and to test dynamic reconstruction algorithms. This paper gives a detailed description of the design and development of GATE by the OpenGATE collaboration, whose continuing objective is to improve, document and validate GATE by simulating commercially available imaging systems for PET and SPECT. Large effort is also invested in the ability and the flexibility to model novel detection systems or systems still under design. A public release of GATE licensed under the GNU Lesser General Public License can be downloaded at http:/www-lphe.epfl.ch/GATE/. Two benchmarks developed for PET and SPECT to test the installation of GATE and to serve as a tutorial for the users are presented. Extensive validation of the GATE simulation platform has been started, comparing simulations and measurements on commercially available acquisition systems. References to those results are listed. The future prospects towards the gridification of GATE and its extension to other domains such as dosimetry are also discussed.

GATE: a simulation toolkit for PET and SPECT

Monte Carlo simulation is an essential tool in emission tomography that can assist in the design of new medical imaging devices, the optimization of acquisition protocols and the development or assessment of image reconstruction algorithms and correction techniques. GATE, the Geant4 Application for Tomographic Emission, encapsulates the Geant4 libraries to achieve a modular, versatile, scripted simulation toolkit adapted to the field of nuclear medicine. In particular, GATE allows the description of time-dependent phenomena such as source or detector movement, and source decay kinetics. This feature makes it possible to simulate time curves under realistic acquisition conditions and to test dynamic reconstruction algorithms. This paper gives a detailed description of the design and development of GATE by the OpenGATE collaboration, whose continuing objective is to improve, document and validate GATE by simulating commercially available imaging systems for PET and SPECT. Large effort is also invested in the ability and the flexibility to model novel detection systems or systems still under design. A public release of GATE licensed under the GNU Lesser General Public License can be downloaded at http:/www-lphe.epfl.ch/GATE/. Two benchmarks developed for PET and SPECT to test the installation of GATE and to serve as a tutorial for the users are presented. Extensive validation of the GATE simulation platform has been started, comparing simulations and measurements on commercially available acquisition systems. References to those results are listed. The future prospects towards the gridification of GATE and its extension to other domains such as dosimetry are also discussed.

A hybrid scatter correction for 3D PET based on an estimation of the distribution of unscattered coincidences: implementation on the ECAT EXACT HR+

We implemented a hybrid scatter-correction method for 3D PET that combines two scatter-correction methods in a complementary way. The implemented scheme uses a method based on the discrimination of the energy of events (the estimation of trues method (ETM)) and an auxiliary method (the single scatter simulation method (SSSI) or the convolution-subtraction method (CONV)) in an attempt to increase the accuracy of the correction over a wider range of acquisitions. The ETM takes into account the scatter from outside the field-of-view (FOV), which is not estimated with the auxiliary method. On the other hand, the auxiliary method accounts for events that have scattered with small angles, which have an energy that cannot be discriminated from that of unscattered events using the ETM. The ETM uses the data acquired in an upper energy window above the photopeak (550-650 keV) to obtain a noisy estimate of the unscattered events in the standard window (350-650 keV). Our implementation uses the auxiliary method to correct the residual scatter in the upper window. After appropriate scaling, the upper window data are subtracted from the total coincidences acquired in the standard window, resulting in the final scatter estimate, after smoothing. In this work we compare the hybrid method with the corrections used by default in the 2D and 3D modes of the ECAT EXACT HR+ using phantom measurements. Generally, the contrast was better with the hybrid method, although the relative errors of quantification were similar. We conclude that hybrid techniques such as the one implemented in this work can provide an accurate, general-purpose and practical way to correct the scatter in 3D PET, taking into account the scatter from outside the FOV.