Search

Introduction Respiratory motion models establish a correspondence between respiratory‐correlated (RC) 4‐dimensional (4D) imaging and respiratory surrogates, to estimate time‐resolved (TR) 3D breathing motion. To evaluate the performance of motion models on real patient data, a validation framework based on magnetic resonance imaging (MRI) is proposed, entailing the use of RC 4DMRI to build the model, and on both (i) TR 2D cine‐MRI and (ii) additional 4DMRI data for testing intra‐/inter‐fraction breathing motion variability. Methods Repeated MRI data were acquired in 7 patients with abdominal lesions. The considered model relied on deformable image registration (DIR) for building the model and compensating for inter‐fraction baseline variations. Both 2D and 3D validation were performed, by comparing model estimations with the ground truth 2D cine‐MRI and 4DMRI respiratory phases, respectively. Results The median DIR error was comparable to the voxel size (1.33 × 1.33 × 5 mm3), with higher values in the presence of large inter‐fraction motion (median value: 2.97 mm). In the 2D validation, the median estimation error on anatomical landmarks’ position resulted below 4 mm in every scenario, whereas in the 3D validation it was 1.33 mm and 4.21 mm when testing intra‐ and inter‐fraction motion, respectively. The range of motion described in the cine‐MRI was comparable to the motion of the building 4DMRI, being always above the estimation error. Overall, the model performance was dependent on DIR error, presenting reduced accuracy when inter‐fraction baseline variations occurred. Conclusions Results suggest the potential of the proposed framework in evaluating global motion models for organ motion management in MRI‐guided radiotherapy.

Multiparametric MRI is a remarkable imaging method for the assessment of patho-physiological processes. In particular, brain tumor characterization has taken advantage of the development of advanced techniques such as Diffusion- (DWI) and Perfusion- (PWI) Weighted Imaging, but a thorough analysis of meningiomas is still lacking despite the variety of computational methods proposed. We compute perfusion and diffusion parametric maps relying on a well-defined methodological workflow, investigating possible correlations between pure and diffusion-based perfusion parameters in a cohort of 26 patients before proton therapy. A preliminary investigation of meningioma staging biomarkers based on IntraVoxel Incoherent Motion and Dynamic Susceptibility Contrast is also reported. We observed significant differences between the gross target volume and the normal appearing white matter for every investigated parameter, confirming the higher vascularization of the neoplastic tissue. DWI and PWI parameters appeared to be weakly correlated and we found that diffusion parameters - the perfusion fraction in particular - could be promising biomarkers for tumor staging.

Skull-base chordoma (SBC) can be treated with carbon ion radiotherapy (CIRT) to improve local control (LC). The study aimed to explore the role of multi-parametric radiomic, dosiomic and clinical features as prognostic factors for LC in SBC patients undergoing CIRT. Before CIRT, 57 patients underwent MR and CT imaging, from which tumour contours and dose maps were obtained. MRI and CT-based radiomic, and dosiomic features were selected and fed to two survival models, singularly or by combining them with clinical factors. Adverse LC was given by in-field recurrence or tumour progression. The dataset was split in development and test sets and the models&rsquo, performance evaluated using the concordance index (C-index). Patients were then assigned a low- or high-risk score. Survival curves were estimated, and risk groups compared through log-rank tests (after Bonferroni correction &alpha, = 0.0083). The best performing models were built on features describing tumour shape and dosiomic heterogeneity (median/interquartile range validation C-index: 0.80/024 and 0.79/0.26), followed by combined (0.73/0.30 and 0.75/0.27) and CT-based models (0.77/0.24 and 0.64/0.28). Dosiomic and combined models could consistently stratify patients in two significantly different groups. Dosiomic and multi-parametric radiomic features showed to be promising prognostic factors for LC in SBC treated with CIRT.

Ocular proton therapy (OPT) is acknowledged as a therapeutic option for the treatment of ocular melanomas. OPT clinical workflow is deeply based on x-ray image guidance procedures, both for treatment planning and patient setup verification purposes. An optimized eye orientation relative to the proton beam axis is determined during treatment planning and it is reproduced during treatment by focusing the patient gaze on a fixation light conveniently positioned in space. Treatment geometry verification is routinely performed through stereoscopic radiographic images while real time patient gaze reproducibility is qualitatively monitored by visual control of eye surface images acquired by dedicated optical cameras. We described an approach to quantitatively evaluate the stability of patients’ gaze direction over an OPT treatment course at the National Centre of Oncological Hadrontherapy (Centro Nazionale di Adroterapia Oncologica, CNAO, Pavia, Italy).

To assess early microstructural changes of meningiomas treated with proton therapy through quantitative analysis of intravoxel incoherent motion (IVIM) and diffusion-weighted imaging (DWI) parameters. Seventeen subjects with meningiomas that were eligible for proton therapy treatment were retrospectively enrolled. Each subject underwent a magnetic resonance imaging (MRI) including DWI sequences and IVIM assessments at baseline, immediately before the 1st (t0), 10th (t10), 20th (t20), and 30th (t30) treatment fraction and at follow-up. Manual tumor contours were drawn on T2-weighted images by two expert neuroradiologists and then rigidly registered to DWI images. Median values of the apparent diffusion coefficient (ADC), true diffusion (D), pseudo-diffusion (D*), and perfusion fraction (f) were extracted at all timepoints. Statistical analysis was performed using the pairwise Wilcoxon test. Statistically significant differences from baseline to follow-up were found for ADC, D, and D* values, with a progressive increase in ADC and D in conjunction with a progressive decrease in D*. MRI during treatment showed statistically significant differences in D values between t0 and t20 (p = 0.03) and t0 and t30 (p = 0.02), and for ADC values between t0 and t20 (p = 0.04), t10 and t20 (p = 0.02), and t10 and t30 (p = 0.035). Subjects that showed a volume reduction greater than 15% of the baseline tumor size at follow-up showed early D changes, whereas ADC changes were not statistically significant. IVIM appears to be a useful tool for detecting early microstructural changes within meningiomas treated with proton therapy and may potentially be able to predict tumor response.

An Eye Tracking System (ETS) is used at CNAO for providing a stable and reproducible ocular proton therapy (OPT) set-up, featuring a fixation light (FL) and monitoring stereo-cameras embedded in a rigid case. The aim of this work is to propose an ETS set-up simulation algorithm, that automatically provides the FL positioning in space, according to patient-specific gaze direction and avoiding interferences with patient, beam and collimator. Two configurations are provided: one in the CT room for acquiring images required for treatment planning with the patient lying on a couch, and one related to the treatment room with the patient sitting in front of the beam. Algorithm validation was performed reproducing ETS simulation (CT) and treatment (room) set-up for 30 patients previously treated at CNAO. The positioning accuracy of the device was quantified through a set of 14 control points applied to the ETS case and localizable both in the CT volume and in room X-ray images. Differences between the position of ETS reference points estimated by the algorithm and those measured by imaging systems are reported. The corresponding gaze direction deviation is on average 0.2° polar and 0.3° azimuth for positioning in CT room and 0.1° polar and 0.4° azimuth in the treatment room. The simulation algorithm was embedded in a clinically usable software application, which we assessed as capable of ensuring ETS positioning with an average accuracy of 2 mm in CT room and 1.5 mm in treatment room, corresponding to gaze direction deviations consistently lower than 1°.

BACKGROUND AND PURPOSE To compare RECIST 1.1 with volume modifications in patients with sacral chordoma not suitable for surgery treated with carbon ions radiotherapy (CIRT) alone. To evaluate patients pain before and after CIRT. To detect if baseline Apparent Diffusion Coefficient values (ADC) from Diffusion Weighted sequences could predict response to treatment. MATERIAL AND METHODS Patients included had one cycle of CIRT and underwent MRI before and after treatment. For each MRI, lesion maximum diameter and volume were obtained, and ADC values were analyzed within the whole lesion volume. Patients pain was evaluated with Numerical Rating Scale (NRS), considering the upper tumor level at baseline MRIs. RESULTS 39 patients were studied (mean follow-up 18 months). Considering RECIST 1.1 there was not a significant reduction in tumor diameters (p = 0.19), instead there was a significant reduction in tumor volume (p

Purpose To generate virtual 4DCT from 4DMRI with field of view (FOV) extended to the entire involved patient anatomy, in order to evaluate its use in carbon ion radiation therapy (CIRT) of the abdominal site in a clinical scenario. Materials and methods The virtual 4DCT was generated by deforming a reference CT in order to (1) match the anatomy depicted in the 4DMRI within its FOV, by calculating deformation fields with deformable image registration to describe inter-fractional and breathing motion, and (2) obtain physically plausible deformation outside of the 4DMRI FOV, by propagating and modulating the previously obtained deformation fields. The implemented method was validated on a digital anthropomorphic phantom, for which a ground truth (GT) 4DCT was available. A CIRT treatment plan was optimized at the end-exhale reference CT and the RBE-weighted dose distribution was recalculated on both the virtual and GT 4DCTs. The method estimation error was quantified by comparing the virtual and GT 4DCTs and the corresponding recomputed doses. The method was then evaluated on 8 patients with pancreas or liver tumors treated with CIRT using respiratory gating at end-exhale. The clinical treatment plans adopted at the National Center for Oncological Hadrontherapy (CNAO, Pavia, Italy) were considered and the dose distribution was recomputed on all respiratory phases of the planning and virtual 4DCTs. By comparing the two datasets and the corresponding dose distributions, the geometrical and dosimetric impact of organ motion was assessed. Results For the phantom, the error outside of the 4DMRI FOV was up to 4.5 mm, but it remained sub-millimetric in correspondence to the target within the 4DMRI FOV. Although the impact of motion on the target D95% resulted in variations ranging from 22% to 90% between the planned dose and the doses recomputed on the GT 4DCT phases, the corresponding estimation error was ≤2.2%. In the patient cases, the variation of the baseline tumor position between the planning and the virtual end-exhale CTs presented a median (interquartile range) value of 6.0 (4.9) mm. For baseline variations larger than 5 mm, the tumor D95% variation between the plan and the dose recomputed on the end-exhale virtual CT resulted larger than 10%. Median variations higher than 10% in the target D95% and gastro-intestinal OARs D2% were quantified at the end-inhale, whereas close to the end-exhale phase, limited variations of relevant dose metrics were found for both tumor and OARs. Conclusions The negligible impact of the geometrical inaccuracy in the estimated anatomy outside of the 4DMRI FOV on the overall dosimetric accuracy suggests the feasibility of virtual 4DCT with extended FOV in CIRT of the abdominal site. In the analyzed patient group, inter-fractional variations such as baseline variation and breathing variability were quantified, demonstrating the method capability to support treatment planning in gated CIRT of the abdominal site.

Purpose To evaluate if baseline ADC from DWI sequences could predict response to treatment in patients with sacral chordoma not suitable for surgery treated with carbon ion radiotherapy (CIRT) alone compared with volume changes. Methods Fifty-nine patients with sacral chordoma not suitable for surgery underwent one cycle of CIRT alone and a minimum of 12-months follow-up. All patients underwent MRI before treatment (baseline), every three months in the first two years after treatment, and every six months afterwards. For each MRI, lesion volume was obtained and median, kurtosis, and skewness ADC were analyzed within the whole lesion volume. Volume changes between baseline and the last available follow-up were used to divide patients with partial response, progression of disease and stable disease (PR, PD, and SD). Results Ten patients were excluded since DWI sequences from baseline MRI were not available. ADC maps obtained from baseline DWI examinations of 50 lesions in the remaining 49 patients were considered. Seven lesions were categorized as PD, 30 PR, and 13 SD. PD showed significantly higher median ADC values at baseline (p = 0.003) compared with both PR and SD (1665vs1253vs1263 *10−6 mm²/s), and more negative skewness values (-0.26vs0.26vs0.08), although not significantly different (p = 0.16). Conclusions Preliminary results suggest that baseline ADC could predict response to treatment with CIRT, particularly to detect potential non-responder patients.

Purpose In image-guided particle radiotherapy of abdominal lesions, respiratory motion hinders treatment accuracy. In this study, 2D cineMRI data were used to quantify the tumor (GTV) motion and to evaluate the clinical approach based on deriving an internal target volume (ITV) from a planning 4DCT for gating treatments. Methods Seven patients with abdominal lesions were treated with carbon-ion therapy at the National Centre of Oncological Hadron-therapy (Italy). The MR scan was performed on the same day of the 4DCT acquisition. For four patients, an additional MR was acquired approximately after 1 week. The cineMRI combined with deformable image registration algorithm was used to quantify tumor motion. Afterwards, two ITVs were defined considering (1) all phases (ITVFB) and (2) only phases within the gating window (ITVG), and then compared with the clinical (4DCT-derived) ITVs (ITVCG and ITVCFB). Results Tumor residual motion estimated by cineMRI data in the two MRI sessions resulted not significantly different from 4DCT, although cineMRI accounted for cycle-to-cycle variations. The ITV normalized for the GTV median values were higher for ITVFB with respect to ITVG, ITVCFB and ITVCG. The Hausdorff distances with respect to the GTV were up to 10.55 mm, 3.13 mm, 5.56 mm and 2.51 mm, for ITVFB, ITVG, ITVCFB and ITVCG, respectively. According to both metrics, ITVCG and ITVG were not found significantly different. Conclusions CineMRI acquisitions allowed to quantify organ motion without delivering additional dose to the patient and to verify treatment margins in gated carbon-ion therapy of abdominal lesions.

Objective: Accurate patient positioning is crucial in particle therapy due to the geometrical selectivity of particles. We report and discuss the National Center for Oncological Hadrontherapy (CNAO) experience in positioning accuracy and stability achieved with solid thermoplastic masks fixed on index base plates and assessed by daily orthogonal X-ray imaging. Methods: Positioning data were retrospectively collected (between 2012 and 2018) and grouped according to the treated anatomical site. 19696 fractions of 1325 patients were evaluated. The study was designed to assess: (i) the number of fractions in which a single correction vector was applied(SCV); (ii) the number of fractions in which further setup verification was performed (SV); (iii) the number of fractions in which SV lead to an additional correction within (MCV5min) 5 minutes from the first setup correction; (iv) the systematic (Σ) and random (σ) error components of the correction vectors applied. Results: A SCV was applied in 71.5% of fractions, otherwise SV was required. In 30.6% of fractions with SV, patient position was not further revised. In the remaining fractions, MCV5min were applied mainly in extracranial and cranial sites respectively. Interfraction Σ was ≤ 1.7 mm/0.7° and σ was ≤ 1.2 mm/0.6° in cranial sites while in extracranial sites Σ was ≤ 5.5 mm/0.9° and σ was ≤4.4 mm/0.9°. Setup residuals were submillimetric in all sites. In cranial patients, maximum intrafractional Σ was 0.8 mm/0.4°. Conclusion: This report extensively quantifies inter- and intrafraction setup accuracy on an institutional basis and confirms the need of image guidance to fully benefit from the geometrical selectivity of particles. Advances in knowledge: The reported analysis provides a board institutional data set on the evaluation of patient immobilization and bony anatomy alignment for several particle therapy clinical indications.

Particle therapy (PT) has shown positive therapeutic results in local control of locally advanced pancreatic lesions. PT effectiveness is highly influenced by target localization accuracy both in space, since the pancreas is located in proximity to radiosensitive vital organs, and in time as it is subject to substantial breathing‐related motion. The purpose of this preliminary study was to quantify pancreas range of motion under typical PT treatment conditions. Three common immobilization devices (vacuum cushion, thermoplastic mask, and compressor belt) were evaluated on five male patients in prone and supine positions. Retrospective four‐dimensional magnetic resonance imaging data were reconstructed for each condition and the pancreas was manually segmented on each of six breathing phases. A k‐means algorithm was then applied on the manually segmented map in order to obtain clusters representative of the three pancreas segments: head, body, and tail. Centers of mass (COM) for the pancreas and its segments were computed, as well as their displacements with respect to a reference breathing phase (beginning exhalation). The median three‐dimensional COM displacements were in the range of 3 mm. Latero–lateral and superior–inferior directions had a higher range of motion than the anterior–posterior direction. Motion analysis of the pancreas segments showed slightly lower COM displacements for the head cluster compared to the tail cluster, especially in prone position. Statistically significant differences were found within patients among the investigated setups. Hence a patient‐specific approach, rather than a general strategy, is suggested to define the optimal treatment setup in the frame of a millimeter positioning accuracy. PACS number(s): 87.55.‐x, 87.57.nm, 87.61

A clinical trial named PROMETHEUS is currently ongoing for inoperable hepatocellular carcinoma (HCC) at the Heidelberg Ion Beam Therapy Center (HIT, Germany). In this framework, 4D PET-CT datasets are acquired shortly after the therapeutic treatment to compare the irradiation induced PET image with a Monte Carlo PET prediction resulting from the simulation of treatment delivery. The extremely low count statistics of this measured PET image represents a major limitation of this technique, especially in presence of target motion. The purpose of the study is to investigate two different 4D PET motion compensation strategies towards the recovery of the whole count statistics for improved image quality of the 4D PET-CT datasets for PET-based treatment verification. The well-known 4D-MLEM reconstruction algorithm, embedding the motion compensation in the reconstruction process of 4D PET sinograms, was compared to a recently proposed pre-reconstruction motion compensation strategy, which operates in sinogram domain by applying the motion compensation to the 4D PET sinograms. With reference to phantom and patient datasets, advantages and drawbacks of the two 4D PET motion compensation strategies were identified. The 4D-MLEM algorithm was strongly affected by inverse inconsistency of the motion model but demonstrated the capability to mitigate the noise-break-up effects. Conversely, the pre-reconstruction warping showed less sensitivity to inverse inconsistency but also more noise in the reconstructed images. The comparison was performed by relying on quantification of PET activity and ion range difference, typically yielding similar results. The study demonstrated that treatment verification of moving targets could be accomplished by relying on the whole count statistics image quality, as obtained from the application of 4D PET motion compensation strategies. In particular, the pre-reconstruction warping was shown to represent a promising choice when combined with intra-reconstruction smoothing.

Purpose To derive personalized tumour control probability (TCP) models, using diffusion-weighted (DW-) MRI for defining initial tumour cellular density in skull-base chordoma patients undergoing carbon-ion radiotherapy (CIRT). Materials and methods 67 patients affected by skull-base chordoma were enrolled for a standardized CIRT treatment (70.4 Gy (RBE) prescription dose). Local control information was clinically assessed. For 20 of them, apparent diffusion coefficient (ADC) maps were computed from DW-MRI and then converted into cellular density. Radiosensitivity parameters (α, β) were estimated from the available data through an optimization procedure, taking advantage of a relationship observed between local control and the dose received by at least the 98% of the gross tumour volume. These parameters were fed into two poissonian TCP models, based on the LQ model, being the first (TCPLIT) computed from literature parameters and the second (TCPADC) enriched by a personalized initial cellular density derived from ADC maps. Results The inclusion of the cellular density derived from ADC into TCPADC yielded slightly higher dose values at which TCP = 0.5 (D50 = 38.91 Gy (RBE)) with respect to TCPLIT (D5034.16 Gy (RBE)). This suggested a more conservative approach, even if the prognostic power of TCPADC and TCPLIT, tested with respect to local control, was equivalent in terms of sensitivity (0.867) and specificity (0.600). Conclusions Both TCPADC and TCPLIT exhibited good agreement with a clinically validated information of local control, the former providing more conservative predictions.

Purpose Apparent diffusion coefficient (ADC) plays a key-role in the context of quantitative diffusion-weighted MRI (DW-MRI) for response assessment in particle-therapy. With this purpose, threshold values defined on ADC could be clinically used to discriminate respondent and not-respondent patients. However, concordant ADC measurements are required for their acceptance as quantitative biomarkers [1] . This study aims at retrospectively investigating the role of ADC maps estimation by comparing different offline strategies with respect to an online map, directly derived from the clinical scanner. Methods DW-MRI at 7 b-values (0, 50, 100, 150, 200, 400, 1000 s/mm2) were acquired during 58 examinations of the brain using a clinical MR scanner (3T). ADC was computed online and offline. For the latter case, DW-MR images at two (ADC_2b: 0–1000 s/mm2; ADC_200b: 200–1000 s/mm2), three (ADC_3b: 50-400–1000 s/mm2) and all (ADC_7b) b-values were used to linearly fit a mono-exponential model exploiting least-square optimization. Voxel-wise absolute differences were computed between any offline and the online ADC for each exam (ΔADC_∗∗b), both on the whole image and on the manually delineated gross tumour volume (GTV). Median values of such differences were then compared using the Friedman test (α = 0.01). Results Evaluating the whole DW-MRI images, ΔADC_2b did not show any statically significant difference with respect to ΔADC_3b, whereas all other pairs did. Focusing on the GTV, differences between any pair were found to be significantly different. The smallest differences were found for ΔADC_7b ( median WHOLE [iqr] = 0.011 [0.009] and median GTV [iqr] = 0.006 [0.001] 10 - 3 mm2/s) and the highest for Δ ADC_200b ( median WHOLE [iqr] = 0.274 [0.034] and median GTV [iqr] = 0.231 [0.034] 10 - 3 mm2/s), whereas Δ ADC_3b and Δ ADC_2b showed intermediate results. Conclusions From this preliminary evaluation, online ADC seems to make use of all available DW images, thus better relating to ADC_7b. Attention must be paid when comparing discriminative threshold values based on online ADC maps and studies aiming at defining the optimal b-values combination for different target tissues should be encouraged.

The purpose of this paper is to describe the experiments and methods that led to the geometrical interpretation of new-generation commercial PET-CT scanners, finalized to off-line PET-based treatment verification in ion beam therapy. Typically, the geometrical correspondence between the image domain (i.e., the dicom PET) and the sinogram domain (i.e., the PET raw data) is not explicitly described by scanner vendors. Hence, the proposed characterization can be applied to commercial PET-CT scanners used in biomedical research, for the development of technologies and methods requiring the use of PET raw data, without having access to confidential information from the vendors.

A novel strategy for 4D PET optimization in the sinogram domain is proposed, aiming at motion model application before image reconstruction (“sinogram warping” strategy). Compared to state-of-the-art 4D-MLEM reconstruction, the proposed strategy is able to optimize the image SNR, avoiding iterative direct and inverse warping procedures, which are typical of the 4D-MLEM algorithm. A full-count statistics sinogram of the motion-compensated 4D PET reference phase is generated by warping the sinograms corresponding to the different PET phases. This is achieved relying on a motion model expressed in the sinogram domain. The strategy was tested on the anthropomorphic 4D PET–CT NCAT phantom in comparison with the 4D-MLEM algorithm, with particular reference to robustness to PET–CT co-registrations artefacts. The MLEM reconstruction of the warped sinogram according to the proposed strategy exhibited better accuracy (up to +40.90 % with respect to the ideal value), whereas images reconstructed according to the 4D-MLEM reconstruction resulted in less noisy (down to −26.90 % with respect to the ideal value) but more blurred. The sinogram warping strategy demonstrates advantages with respect to 4D-MLEM algorithm. These advantages are paid back by introducing approximation of the deformation field, and further efforts are required to mitigate the impact of such an approximation in clinical 4D PET reconstruction.

The purpose of the study is to evaluate the performance of a novel strategy, referred to as “virtual 4D PET”, aiming at the optimization of hybrid 4D CT-PET scan for radiotherapy treatment planning. The virtual 4D PET strategy applies 4D CT motion modeling to avoid time-resolved PET image acquisition. This leads to a reduction of radioactive tracer administered to the patient and to a total acquisition time comparable to free-breathing PET studies. The proposed method exploits a motion model derived from 4D CT, which is applied to the free-breathing PET to recover respiratory motion and motion blur. The free-breathing PET is warped according to the motion model, in order to generate the virtual 4D PET. The virtual 4D PET strategy was tested on images obtained from a 4D computational anthropomorphic phantom. The performance was compared to conventional motion compensated 4D PET. Tests were also carried out on clinical 4D CT-PET scans coming from seven lung and liver cancer patients. The virtual 4D PET strategy was able to recover lesion motion, with comparable performance with respect to the motion compensated 4D PET. The compensation of the activity blurring due to motion was successfully achieved in terms of spill out removal. Specific limitations were highlighted in terms of partial volume compensation. Results on clinical 4D CT-PET scans confirmed the efficacy in 4D PET count statistics optimization, as equal to the free-breathing PET, and recovery of lesion motion. Compared to conventional motion compensation strategies that explicitly require 4D PET imaging, the virtual 4D PET strategy reduces clinical workload and computational costs, resulting in significant advantages for radiotherapy treatment planning.

The synergy between in-room imaging and optical tracking, in co-operation with highly accurate robotic patient handling represents a concept for patient-set-up which has been implemented at CNAO (Centro Nazionale di Adroterapia Oncologica). In-room imaging is based on a double oblique X-ray projection system; optical tracking consists of the detection of the position of spherical markers placed directly on the patient's skin or on the immobilization devices. These markers are used as external fiducials during patient positioning and dose delivery. This study reports the results of a comparative analysis between in-room imaging and optical tracking data for patient positioning within the framework of high-precision particle therapy. Differences between the optical tracking system (OTS) and the imaging system (IS) were on average within the expected localization accuracy. On the first 633 fractions for head and neck (HN) set-up procedures, the corrections applied by the IS, after patient positioning using the OTS only, were for the mostly sub-millimetric regarding the translations (0.4 ± 1.1 mm) and sub-gradual regarding the rotations (0.0° ± 0.8°). On the first 236 fractions for pelvis localizations the amplitude of the corrections applied by the IS after preliminary optical set-up correction were moderately higher and more dispersed (translations: 1.3 ± 2.9 mm, rotations 0.1 ± 0.9°). Although the indication of the OTS cannot replace information provided by in-room imaging devices and 2D-3D image registration, the reported data show that OTS preliminary correction might greatly support image-based patient set-up refinement and also provide a secondary, independent verification system for patient positioning.