Albert Hirtl

PET based volume segmentation with emphasis on the iterative TrueX algorithm.

PURPOSE To assess the influence of reconstruction algorithms for positron emission tomography (PET) based volume quantification. The specifically detected activity in the threshold defined volume was investigated for different reconstruction algorithms as a function of volume size and signal to background ratio (SBR), especially for volumes smaller than 1ml. Special attention was given to the Siemens specific iterative reconstruction algorithm TrueX. METHODS Measurements were performed with a modified in-house produced IEC body phantom on a Siemens Biograph 64 True Point PET/CT scanner (Siemens, Medical Systems) for six different SBRs (2.1, 3.8, 4.9, 6.7, 8.9, 9.4 and without active background (BG)). The phantom consisted of a water-filled cavity with built-in plastic spheres (0.27, 0.52, 1.15, 2.57, 5.58 and 11.49ml). The following reconstruction algorithms available on the Siemens Syngo workstation were evaluated: Iterative OSEM (OSEM) (4 iterations, 21 subsets), iterative TrueX (TrueX) (4 iterations, 21 subsets) and filtered backprojection (FBP). For the threshold based volume segmentation the software Rover (ABX, Dresden) was used. RESULTS For spheres larger than 2.5ml a constant threshold (standard deviation (SD) 10%) level was found for a given SBR and reconstruction algorithm and therefore a mean threshold for the largest three spheres was calculated. This threshold could be approximated by a function inversely proportional to the SBR. The threshold decreased with increasing SBR for all sphere sizes. For the OSEM algorithm the threshold for small spheres with 0.27, 0.52 and 1.15ml varied between 17% and 44% (depending on sphere size). The threshold for the TrueX algorithm was substantially lower (up to 17%) than for the OSEM algorithm for all sphere sizes. The maximum activity in a specific volume yielded the true activity for the OSEM algorithm when using a SBR independent correction factor C, which depended on sphere size. For the largest three volumes a constant factor C=1.10±0.03 was found. For smaller volumes, C increased exponentially due to the partial volume effect. For the TrueX algorithm the maximum activity overestimated the true activity. CONCLUSION The threshold values for PET based target volume segmentation increased with increasing sphere size for all tested algorithms. True activity values of spheres in the phantom could be extracted using experimentally determined correction factors C. The TrueX algorithm has to be used carefully for quantitative comparison (e.g. follow-up) and multicenter studies.

A pencil beam algorithm for helium ion beam therapy

PURPOSE To develop a flexible pencil beam algorithm for helium ion beam therapy. Dose distributions were calculated using the newly developed pencil beam algorithm and validated using Monte Carlo (MC) methods. METHODS The algorithm was based on the established theory of fluence weighted elemental pencil beam (PB) kernels. Using a new real-time splitting approach, a minimization routine selects the optimal shape for each sub-beam. Dose depositions along the beam path were determined using a look-up table (LUT). Data for LUT generation were derived from MC simulations in water using GATE 6.1. For materials other than water, dose depositions were calculated by the algorithm using water-equivalent depth scaling. Lateral beam spreading caused by multiple scattering has been accounted for by implementing a non-local scattering formula developed by Gottschalk. A new nuclear correction was modelled using a Voigt function and implemented by a LUT approach. Validation simulations have been performed using a phantom filled with homogeneous materials or heterogeneous slabs of up to 3 cm. The beams were incident perpendicular to the phantoms surface with initial particle energies ranging from 50 to 250 MeV/A with a total number of 10(7) ions per beam. For comparison a special evaluation software was developed calculating the gamma indices for dose distributions. RESULTS In homogeneous phantoms, maximum range deviations between PB and MC of less than 1.1% and differences in the width of the distal energy fall off of the Bragg-Peak from 80% to 20% of less than 0.1 mm were found. Heterogeneous phantoms using layered slabs satisfied a γ-index criterion of 2%/2mm of the local value except for some single voxels. For more complex phantoms using laterally arranged bone-air slabs, the γ-index criterion was exceeded in some areas giving a maximum γ-index of 1.75 and 4.9% of the voxels showed γ-index values larger than one. The calculation precision of the presented algorithm was considered to be sufficient for clinical practice. Although only data for helium beams was presented, the performance of the pencil beam algorithm for proton beams was comparable. CONCLUSIONS The pencil beam algorithm developed for helium ions presents a suitable tool for dose calculations. Its calculation speed was evaluated to be similar to other published pencil beam algorithms. The flexible design allows easy customization of measured depth-dose distributions and use of varying beam profiles, thus making it a promising candidate for integration into future treatment planning systems. Current work in progress deals with RBE effects of helium ions to complete the model.

Variation of system performance, quality control standards and adherence to international FDG-PET/CT imaging guidelines. A national survey of PET/CT operations in Austria.

AIM To gather information on clinical operations, quality control (QC) standards and adoption of guidelines for FDG-PET/CT imaging in Austrian PET/CT centres. METHODS A written survey composed of 68 questions related to A) PET/CT centre and installation, B) standard protocol parameters for FDG-PET/CT imaging of oncology patients, and C) standard QC procedures was conducted between November and December 2013 among all Austrian PET/CT centres. In addition, a NEMA-NU2 2012 image quality phantom test was performed using standard whole-body imaging settings on all PET/CT systems with a lesion-to-background ratio of 4. Recovery coefficients (RC) were calculated for each lesion and PET/CT system. RESULTS A) 13 PET/CT systems were installed in 12 nuclear medicine departments at public hospitals. B) Average fasting prior to FDG-PET/CT was 7.6 (4-12) h. All sites measured blood glucose levels while using different cut-off levels (64%: 150 mg/dl). Weight-based activity injection was performed at 83% sites with a mean FDG activity of 4.1 MBq/kg. Average FDG uptake time was 55 (45-75) min. All sites employed CT contrast agents (variation from 1%-95% of the patients). All sites reported SUV-max. C) Frequency of QC tests varied significantly and QC phantom measurements revealed significant differences in RCs. CONCLUSION Significant variations in FDG-PET/CT protocol parameters among all Austrian PET/CT users were observed. Subsequently, efforts need to be put in place to further standardize imaging protocols. At a minimum clinical PET/CT operations should ensure compliance with existing guidelines. Further, standardized QC procedures must be followed to improve quantitative accuracy across PET/CT centres.

Evaluating Treatment Response of Radioembolization in Intermediate-Stage Hepatocellular Carcinoma Patients Using 18F-Fluoroethylcholine PET/CT

The aim of this study was to evaluate 18F-fluoroethylcholine PET/CT as a metabolic imaging technique for the assessment of treatment response to 90Y radioembolization in patients with locally advanced hepatocellular carcinoma (HCC). Methods: Thirty-four HCC patients undergoing 78 18F-fluoroethylcholine PET/CT scans were identified for this study. Patients with initial or follow-up metastastic disease (n = 9) were excluded at the time point of the metastatic occurrence as well as patients with negative α-fetoprotein (AFP; n = 1), resulting in 24 patients and 57 scans that were eligible. All patients were scheduled for radioembolization and underwent 1 pretherapeutic and at least 1 posttherapeutic 18F-fluoroethylcholine PET/CT scan. Volume-of-interest analysis and volume-of-interest subtractions were performed. Maximum, mean, and peak standardized uptake value (SUV) analysis was performed, and the total intrahepatic 18F-fluoroethylcholine positive tumor volume (FEC-PTV) and tumor-to-background ratio were assessed. Statistical analysis was performed using a decreasing AFP of at least 20% as a standard of reference for therapy response including receiver-operating-characteristic analyses as well as descriptive and correlation analyses and multiple logistic regression. Results: Fourteen follow-up examinations were categorized as responder and 19 follow-up examinations as nonresponder. Absolute AFP values did not correlate with SUV parameters (P = 0.055). In receiver-operating-characteristic analyses, the initial mean SUV, Δmaximum SUV, and Δtumor-to-background ratio demonstrated the highest area under the curve, 0.84 (P = 0.009), 0.83 (P = 0.011), and 0.83 (P = 0.012), respectively, resulting in a positive prediction of 82%, 83%, and 91% at the respective cutoff points. When multiple logistic regression analysis was applied, this resulted in an area under the curve of 0.90 (P = 0.001), with a positive prediction of 94% and a sensitivity of 94%. The FEC-PTV did not reach significance in the presented dataset. Conclusion: 18F-fluoroethylcholine PET/CT demonstrates a high potential for follow-up assessment in the context of radioembolization in patients with locally advanced, but nonmetastatic, HCC and initially elevated AFP, possibly enabling early therapy monitoring independent of morphology.

Influence of PET reconstruction parameters on the TrueX algorithm. A combined phantom and patient study.

UNLABELLED With the increasing use of functional imaging in modern radiotherapy (RT) and the envisaged automated integration of PET into target definition, the need for reliable quantification of PET is growing. Reconstruction algorithms in new PET scanners employ point-spread-function (PSF) based resolution recovery, however, their impact on PET quantification still requires thorough investigation. PATIENTS, MATERIAL, METHODS Measurements were performed on a Siemens PET/CT using an IEC phantom filled with varying activity. Data were reconstructed using the OSEM (Gauss filter) and the PSF TrueX (Gauss and Allpass filter) algorithm with all available products of iterations (i) and subsets (ss). The recovery coeffcient (RC) and threshold defining the real sphere volume were determined for all settings and compared to the clinical standard (4i21ss). PET acquisitions of eight lung patients were reconstructed using all algorithms with 4i21ss. Volume size and tracer uptake were determined with different segmentation methods. RESULTS The threshold for the TrueX was lower (up to 40%) than for the OSEM. The RC for the different algorithms and filters varied. TrueX was more sensitive to permutations of i and ss and only the RC of the OSEM stabilised with increasing number. For patient scans the difference of the volume and activity between TrueX and OSEM could be reduced by applying an adapted threshold and activity correction. CONCLUSION The TrueX algorithm results in excellent diagnostic image quality, however, guidelines for native algorithms have to be extended for PSF based reconstruction methods. For appropriate tumour delineation, for the TrueX a lower threshold than the 42% recommended for the OSEM is necessary. These filter dependent thresholds have to be verified for different scanners prior to using them in multicenter trials.

Influence of PET reconstruction parameters on the TrueX algorithm

UNLABELLED With the increasing use of functional imaging in modern radiotherapy (RT) and the envisaged automated integration of PET into target definition, the need for reliable quantification of PET is growing. Reconstruction algorithms in new PET scanners employ point-spread-function (PSF) based resolution recovery, however, their impact on PET quantification still requires thorough investigation. PATIENTS, MATERIAL, METHODS Measurements were performed on a Siemens PET/CT using an IEC phantom filled with varying activity. Data were reconstructed using the OSEM (Gauss filter) and the PSF TrueX (Gauss and Allpass filter) algorithm with all available products of iterations (i) and subsets (ss). The recovery coeffcient (RC) and threshold defining the real sphere volume were determined for all settings and compared to the clinical standard (4i21ss). PET acquisitions of eight lung patients were reconstructed using all algorithms with 4i21ss. Volume size and tracer uptake were determined with different segmentation methods. RESULTS The threshold for the TrueX was lower (up to 40%) than for the OSEM. The RC for the different algorithms and filters varied. TrueX was more sensitive to permutations of i and ss and only the RC of the OSEM stabilised with increasing number. For patient scans the difference of the volume and activity between TrueX and OSEM could be reduced by applying an adapted threshold and activity correction. CONCLUSION The TrueX algorithm results in excellent diagnostic image quality, however, guidelines for native algorithms have to be extended for PSF based reconstruction methods. For appropriate tumour delineation, for the TrueX a lower threshold than the 42% recommended for the OSEM is necessary. These filter dependent thresholds have to be verified for different scanners prior to using them in multicenter trials.

Variation of system performance, quality control standards and adherence to international FDG-PET/CT imaging guidelines

AIM To gather information on clinical operations, quality control (QC) standards and adoption of guidelines for FDG-PET/CT imaging in Austrian PET/CT centres. METHODS A written survey composed of 68 questions related to A) PET/CT centre and installation, B) standard protocol parameters for FDG-PET/CT imaging of oncology patients, and C) standard QC procedures was conducted between November and December 2013 among all Austrian PET/CT centres. In addition, a NEMA-NU2 2012 image quality phantom test was performed using standard whole-body imaging settings on all PET/CT systems with a lesion-to-background ratio of 4. Recovery coefficients (RC) were calculated for each lesion and PET/CT system. RESULTS A) 13 PET/CT systems were installed in 12 nuclear medicine departments at public hospitals. B) Average fasting prior to FDG-PET/CT was 7.6 (4-12) h. All sites measured blood glucose levels while using different cut-off levels (64%: 150 mg/dl). Weight-based activity injection was performed at 83% sites with a mean FDG activity of 4.1 MBq/kg. Average FDG uptake time was 55 (45-75) min. All sites employed CT contrast agents (variation from 1%-95% of the patients). All sites reported SUV-max. C) Frequency of QC tests varied significantly and QC phantom measurements revealed significant differences in RCs. CONCLUSION Significant variations in FDG-PET/CT protocol parameters among all Austrian PET/CT users were observed. Subsequently, efforts need to be put in place to further standardize imaging protocols. At a minimum clinical PET/CT operations should ensure compliance with existing guidelines. Further, standardized QC procedures must be followed to improve quantitative accuracy across PET/CT centres.

New Approaches for Improvement of TOF-PET

Abstract We present results of simulations on the influence of photon propagation and the Cherenkov effect on the time resolution of LSO:Ce scintillators. The influence of the scintillator length on the coincidence time resolution is shown. Furthermore, the impact of the depth of interaction on the time resolution, the light output and the arrival time distribution at the photon detector is simulated and it is shown how these information can be used for time walk correction.

Technical Note: Fully-automated analysis of Jaszczak phantom measurements as part of routine SPECT quality control

Purpose Inspection and quantitative validation of tomographic imaging properties of SPECT systems, i.e., spatial resolution, contrast, and inhomogeneity must be performed in regular intervals. Typically, the modular Jaszczak phantom is used for that purpose, as it offers the possibility to investigate all three system properties with a single measurement. The interpretation of the measurement is performed visually, thus, being insensitive to subtle changes in system performance. To overcome this limitation, a fully‐automated software for the objective analysis of Jaszczak phantom measurements is proposed here. Methods The software was developed as an ImageJ plugin and offers a number of sequential evaluation steps: automatic determination of the type of Jaszczak phantom, calculation of sector and sphere contrast, detection of ring artifacts using either the Hough transform, followed by a threshold‐based decision criterion, or Students t‐test. Monte Carlo simulations were used to estimate the detectability limits for ring artifacts. Results The software successfully calculated sector and sphere contrasts and reliably determined ring artifacts present in the homogeneity part of the Jaszczak phantom, based on automatic identification of the phantom type. Conclusion Given the quantitative nature of the produced output, results from one imaging system can easily be compared to another in an objective way. The advantage of the software is clearly that the information provided is objective and does not rely on the experience level of the user.

Influence of PET reconstruction technique and matrix size on qualitative and quantitative assessment of lung lesions on [18F]-FDG-PET: A prospective study in 37 cancer patients

PURPOSE To evaluate the influence of point spread function (PSF)-based reconstruction and matrix size for PET on (1) lung lesion detection and (2) standardized uptake values (SUV). METHODS This prospective study included oncological patients who underwent [18F]-FDG-PET/CT for staging. PET data were reconstructed with a 2D ordered subset expectation maximization (OSEM) algorithm, and a 2D PSF-based algorithm (TrueX), separately with two matrix sizes (168×168 and 336×336). The four PET reconstructions (TrueX-168; OSEM-168; TrueX-336; and OSEM-336) were read independently by two raters, and PET-positive lung lesions were recorded. Blinded to the PET findings, a third independent rater assessed lung lesions with diameters of >4mm on CT. Subsequently, PET and CT were reviewed side-by side in consensus. Multi-factorial logistic regression analyses and two-way repeated measures analyses of variance (ANOVA) were performed. RESULTS Thirty-seven patients with 206 lung lesions were included. Lesion-based PET sensitivities differed significantly between reconstruction algorithms (P<0.001) and between reconstruction matrices (P=0.022). Sensitivities were 94.2% and 88.3% for TrueX-336; 88.3% and 85.9% for TrueX-168; 67.8% and 66.3% for OSEM-336; and 67.0% and 67.9% for OSEM-168; for rater 1 and rater 2, respectively. SUVmax and SUVmean were significantly higher for images reconstructed with 336×336 matrices than for those reconstructed with 168×168 matrices (P<0.001). CONCLUSION Our results demonstrate that PSF-based PET reconstruction, and, to a lesser degree, higher matrix size, improve detection of metabolically active lung lesions. However, PSF-based PET reconstructions and larger matrix sizes lead to higher SUVs, which may be a concern when PET data from different institutions are compared.