Comparison of four synthetic CT generators for brain and prostate MR-only workflow in radiotherapy.

BACKGROUND: The interest in MR-only workflows is growing with the introduction of artificial intelligence in the synthetic CT generators converting MR images into CT images. The aim of this study was to evaluate several commercially available sCT generators for two anatomical localizations. METHODS: Four sCT generators were evaluated: one based on the bulk density method and three based on deep learning methods. The comparison was performed on large patient cohorts (brain: 42 patients and pelvis: 52 patients). It included geometric accuracy with the evaluation of Hounsfield Units (HU) mean error (ME) for several structures like the body, bones and soft tissues. Dose evaluation included metrics like the Dmean ME for bone structures (skull or femoral heads), PTV and soft tissues (brain or bladder or rectum). A 1%/1 mm gamma analysis was also performed. RESULTS: HU ME in the body were similar to those reported in the literature. Dmean ME were smaller than 2% for all structures. Mean gamma pass rate down to 78% were observed for the bulk density method in the brain. Performances of the bulk density generator were generally worse than the artificial intelligence generators for the brain but similar for the pelvis. None of the generators performed best in all the metrics studied. CONCLUSIONS: All four generators can be used in clinical practice to implement a MR-only workflow but the bulk density method clearly performed worst in the brain.

Lung stereotactic body radiation therapy: personalized PTV margins according to tumor location and number of four-dimensional CT scans.

OBJECTIVES: To characterise the motion of pulmonary tumours during stereotactic body radiation therapy (SBRT) and to evaluate different margins when creating the planning target volume (PTV) on a single 4D CT scan (4DCT). METHODS: We conducted a retrospective single-site analysis on 30 patients undergoing lung SBRT. Two 4DCTs (4DCT1 and 4DCT2) were performed on all patients. First, motion was recorded for each 4DCT in anterior-posterior (AP), superior-inferior (SI) and rightleft (RL) directions. Then, we used 3 different margins (3,4 and 5 mm) to create the PTV, from the internal target volume (ITV) of 4DCT1 only (PTV D1 + 3, PTV D1 + 4, PTV D1 + 5). We compared, using the Dice coefficient, the volumes of these 3 PTVs, to the PTV actually used for the treatment (PTVttt). Finally, new treatment plans were calculated using only these 3 PTVs. We studied the ratio of the D2%, D50% and D98% between each new plan and the plan actually used for the treatment (D2% PTVttt, D50% PTVttt, D50% ITVttt D98% PTVttt). RESULTS: 30 lesions were studied. The greatest motion was observed in the SI axis (8.8 ± 6.6 [0.4-25.8] mm). The Dice index was higher when comparing PTVttt to PTV D1 + 4 mm (0.89 ± 0.04 [0.82-0.98]). Large differences were observed when comparing plans relative to PTVttt and PTV D1 + 3 for D98% PTVttt (0.85 ± 0.24 [0.19-1.00]). and also for D98% ITVttt (0.93 ± 0.12 [0.4-1.0]).D98% PTVttt (0.85 ± 0.24 [0.19-1.00], p value = 0.003) was statistically different when comparing plans relative to PTVttt and PTV D1 + 3. No stastistically differences were observed when comparing plans relative to PTVttt and PTV D1 + 4. A difference greater than 10% relative to D98% PTVttt was found for only in one UL lesion, located under the carina. CONCLUSION: A single 4DCT appears feasible for upper lobe lesions located above the carina, using a 4-mm margin to generate the PTV. ADVANCE IN KNOWLEDGE: Propostion of a personalized SBRT treatment (number of 4DCT, margins) according to tumor location (above or under the carina).

Monte Carlo modeling of gamma cameras for I-131 imaging in targeted radiotherapy.

INTRODUCTION: Dosimetric studies for targeted radiotherapy require the quantification of activity from scintigraphic images. Quantitative imaging is difficult to achieve because of several effects that can lead to errors in activity estimates, some of which are more apparent when I-131 is considered as a source. An evaluation of these phenomena was performed by modeling the gamma camera and its behavior using Monte Carlo simulations. Two gamma cameras were modeled: DST-XLi and Millennium VG Hawk-Eye (GEMS), and two Monte Carlo codes were used: MCNP (LANL) and GATE (openGate collaboration). GATE is a dedicated single photon emission computed tomography/positron emission tomography (SPECT)/(PET) software based on Geant4 (CERN, Geneve). MATERIALS AND METHODS: Gamma-camera modeling was performed in 2 steps: first without a collimator, then with a high-energy, all-purpose (HEAP) collimator according to the specifications given by the manufacturer (the simulation took the hexagonal shape of collimator holes into account). Simulated and measured energy spectra from point sources in air were compared (with or without a collimator). Spatial resolution was obtained from line sources in air at various distances from the detector heads. The photons detected in the 20% energy window from a point source were analyzed in order to determine the amount of primary photons, scattered photons (in the collimator), and septal photons (i.e., photons that crossed the collimator septa without interacting). RESULTS: Both codes agree well with experimental measurements for the two gamma cameras considered in this study. This allowed us to validate gamma-camera modeling and also served as a benchmark of GATE (new code) versus MCNP (reference code). As shown previously by Dewaraja et al., septal penetration is an important source of image degradation when HEAP collimators are used for I-131 imaging. With the DST-XLi, and for a point source in air, our simulations have shown that 53% of scattered (30%) and septal penetration (23%) photons are detected in the 20% window. CONCLUSION: The modeling of two gamma cameras (DST-XLi and Millennium VG Hawk-Eye) has been performed with two Monte Carlo codes (MCNP and Gate). Results obtained with the two Monte Carlo codes agree well with experimental results. As already indicated by several authors, septal penetration and scattered photons in the collimator have a major impact on I-131 scintigraphic imaging.