Detecting outliers beyond tolerance limits derived from statistical process control in patient-specific quality assurance.

BACKGROUND: Tolerance limit is defined on pre-treatment patient specific quality assurance results to identify "out of the norm" dose discrepancy in plan. An out-of-tolerance plan during measurement can often cause treatment delays especially if replanning is required. In this study, we aim to develop an outlier detection model to identify out-of-tolerance plan early during treatment planning phase to mitigate the above-mentioned risks. METHODS: Patient-specific quality assurance results with portal dosimetry for stereotactic body radiotherapy measured between January 2020 and December 2021 were used in this study. Data were divided into thorax and pelvis sites and gamma passing rates were recorded using 2%/2 mm, 2%/1 mm, and 1%/1 mm gamma criteria. Statistical process control method was used to determine six different site and criterion-specific tolerance and action limits. Using only the inliers identified with our determined tolerance limits, we trained three different outlier detection models using the plan complexity metrics extracted from each treatment field-robust covariance, isolation forest, and one class support vector machine. The hyperparameters were optimized using the F1-score calculated from both the inliers and validation outliers' data. RESULTS: 308 pelvis and 200 thorax fields were used in this study. The tolerance (action) limits for 2%/2 mm, 2%/1 mm, and 1%/1 mm gamma criteria in the pelvis site are 99.1% (98.1%), 95.8% (91.1%), and 91.7% (86.1%), respectively. The tolerance (action) limits in the thorax site are 99.0% (98.7%), 97.0% (96.2%), and 91.5% (87.2%). One class support vector machine performs the best among all the algorithms. The best performing model in the thorax (pelvis) site achieves a precision of 0.56 (0.54), recall of 1.0 (1.0), and F1-score of 0.72 (0.70) when using the 2%/2 mm (2%/1 mm) criterion. CONCLUSION: The model will help the planner to identify an out-of-tolerance plan early so that they can refine the plan further during the planning stage without risking late discovery during measurement.

Repurposing DailyQA3 for an efficient and spot position sensitive daily quality assurance tool for proton therapy.

INTRODUCTION: Daily quality assurance is an integral part of a radiotherapy workflow to ensure the dose is delivered safely and accurately to the patient. It is performed before the first treatment of the day and needs to be time and cost efficient for a multiple gantries proton center. In this study, we introduced an efficient method to perform QA for output constancy, range verification, spot positioning accuracy and imaging and proton beam isocenter coincidence with DailyQA3. METHODS: A stepped acrylic block of specific dimensions is fabricated and placed on top of the DailyQA3 device. Treatment plans comprising of two different spread-out Bragg peaks and five individual spots of 1.0 MU each are designed to be delivered to the device. A mathematical framework to measure the 2D distance between the detectors and individual spot is introduced and play an important role in realizing the spot positioning and centering QA. Lastly, a 5 months trends of the QA for two gantries are presented. RESULTS: The outputs are monitored by two ion chambers in the DailyQA3 and a tolerance of ± 3 % $ \pm 3\% $ are used. The range of the SOBPs are monitored by the ratio of ion chamber signals and a tolerance of ± 1 mm $ \pm 1\ {\mathrm{mm}}$ is used. Four diodes at ± 10 cm $ \pm 10\ {\mathrm{cm}}$ from the central ion chambers are used for spot positioning QA, while the central ion chamber is used for imaging and proton beam isocenter coincidence QA. Using the framework, we determined the absolute signal threshold corresponding to the offset tolerance between the individual proton spot and the detector. A 1.5 mm $1.5\ {\mathrm{mm}}$ tolerances are used for both the positioning and centering QA. No violation of the tolerances is observed in the 5 months trends for both gantries. CONCLUSION: With the proposed approach, we can perform four QA items in the TG224 within 10 min.

Effects of modern aesthetic dental fillings on proton therapy.

Background and purpose: High-density dental fillings pose a non-negligible impact on head and neck cancer treatment. For proton therapy, stopping power ratio (SPR) prediction will be significantly impaired by the associated image artifacts. Dose perturbation is also inevitable, compromising the treatment plan quality. While plenty of work has been done on metal or amalgam fillings, none has touched on composite resin (CR) and glass ionomer cement (GIC) which have seen an increasing usage. Hence, this work aims to provide a detailed characterisation of SPR and dose perturbation in proton therapy caused by CR and GIC. Materials and methods: Four types of fillings were used: CR, Fuji Bulk (FB), Fuji II (FII) and Fuji IX (FIX). The latter three belong to GIC category. Measured SPR were compared with SPR predicted using single-energy computed tomography (SECT) and dual-energy computed tomography (DECT). Dose perturbation of proton beams with lower- and higher-energy levels was also quantified using Gafchromic films. Results: The measured SPR for CR, FB, FII and FIX were 1.68, 1.77, 1.77 and 1.76, respectively. Overall, DECT could predict SPR better than SECT. The lowest percentage error achieved by DECT was 19.7 %, demonstrating the challenge in estimating SPR, even for fillings with relatively lower densities. For both proton beam energies and all four fillings of about 4.5 mm thickness, the maximum dose perturbation was 3 %. Conclusion: This study showed that dose perturbation by CR and GIC was comparatively small. We have measured and recommended the SPR values for overriding the fillings in TPS.

Real-time gated proton therapy with a reduced source to imager distance: Commissioning and quality assurance.

INTRODUCTION: Real-time gated proton therapy (RGPT) is a motion management technique unique to the Hitachi particle therapy system. It uses pulsed fluoroscopy to track an implanted fiducial marker. There are currently no published guidelines on how to conduct the commissioning and quality assurance. In this work we reported on our centre's commissioning workflow and our daily and monthly QA procedures. METHODS: Six commissioning measurements were designed for RGPT. The measurements include imaging qualities, fluoroscopic exposures, RGPT marker tracking accuracy, temporal gating latency, fiducial marker tracking fidelity and an end-to-end proton dosimetry measurement. Daily QA consists of one measurement on marker localization accuracy. Four months daily QA trends are presented. Monthly QA consists of three measurementson the gating latency, fluoroscopy imaging quality and dosimetry verification of gating operation with RGPT. RESULTS: The RGPT was successfully commissioned in our centre. The air kerma rates were within 15 % from specifications and the marker tracking accuracies were within 0.245 mm. The gating latencies for turning the proton beam on and off were 119.5 and 50.0 ms respectively. The 0.4x10.0 mm2 Gold AnchorTM gave the best tracking results with visibility up to 30 g/cm2. Gamma analysis showed that dose distribution of a moving and static detectors had a passing rate of more than 95 % at 3 %/3mm. The daily marker localization QA results were all less than 0.2 mm. CONCLUSION: This work could serve as a good reference for other upcoming Hitachi particle therapy centres who are interested to use RGPT as their motion management solution.

First clinical experience following the consensus guide for calibrating a proton stopping power ratio curve in a new proton centre.

BACKGROUND AND PURPOSE: This work introduces the first assessment of CT calibration following the ESTRO's consensus guidelines and validating the HLUT through the irradiation of biological material. METHODS: Two electron density phantoms were scanned with two CT scanners using two CT scan energies. The stopping power ratio (SPR) and mass density (MD) HLUTs for different CT scan energies were derived using Schneider's and ESTRO's methods. The comparison metric in this work is based on the Water-Equivalent Thickness (WET) difference between the treatment planning system and biological irradiation measurement. The SPR HLUTs were compared between the two calibration methods. To assess the accuracy of using MD HLUT for dose calculation in the treatment planning system, MD vs SPR HLUT was compared. Lastly, the feasibility of using a single SPR HLUT to replace two different energy CT scans was explored. RESULTS: The results show a WET difference of less than 3.5% except for the result in the Bone region between Schneider's and ESTRO's methods. Comparing MD and SPR HLUT, the results from MD HLUT show less than a 3.5% difference except for the Bone region. However, the SPR HLUT shows a lower mean absolute percentage difference as compared to MD HLUT between the measured and calculated WET difference. Lastly, it is possible to use a single SPR HLUT for two different CT scan energies since both WET differences are within 3.5%. CONCLUSION: This is the first report on calibrating an HLUT following the ESTRO's guidelines. While our result shows incremental improvement in range uncertainty using the ESTRO's guideline, the prescriptional approach of the guideline does promote harmonization of CT calibration protocols between different centres.

Deformable anthropomorphic pelvis phantom for dose accumulation verification.

Objective.The validation of deformable image registration (DIR) for contour propagation is often done using contour-based metrics. Meanwhile, dose accumulation requires evaluation of voxel mapping accuracy, which might not be accurately represented by contour-based metrics. By fabricating a deformable anthropomorphic pelvis phantom, we aim to (1) quantify the voxel mapping accuracy for various deformation scenarios, in high- and low-contrast regions, and (2) identify any correlation between dice similarity coefficient (DSC), a commonly used contour-based metric, and the voxel mapping accuracy for each organ.Approach. Four organs, i.e. pelvic bone, prostate, bladder and rectum (PBR), were 3D printed using PLA and a Polyjet digital material, and assembled. The latter three were implanted with glass bead and CT markers within or on their surfaces. Four deformation scenarios were simulated by varying the bladder and rectum volumes. For each scenario, nine DIRs with different parameters were performed on RayStation v10B. The voxel mapping accuracy was quantified by finding the discrepancy between true and mapped marker positions, termed the target registration error (TRE). Pearson correlation test was done between the DSC and mean TRE for each organ.Main results. For the first time, we fabricated a deformable phantom purely from 3D printing, which successfully reproduced realistic anatomical deformations. Overall, the voxel mapping accuracy dropped with increasing deformation magnitude, but improved when more organs were used to guide the DIR or limit the registration region. DSC was found to be a good indicator of voxel mapping accuracy for prostate and rectum, but a comparatively poorer one for bladder. DSC > 0.85/0.90 was established as the threshold of mean TRE ⩽ 0.3 cm for rectum/prostate. For bladder, extra metrics in addition to DSC should be considered.Significance. This work presented a 3D printed phantom, which enabled quantification of voxel mapping accuracy and evaluation of correlation between DSC and voxel mapping accuracy.

Predicting dice similarity coefficient of deformably registered contours using Siamese neural network.

Objective. Automatic deformable image registration (DIR) is a critical step in adaptive radiotherapy. Manually delineated organs-at-risk (OARs) contours on planning CT (pCT) scans are deformably registered onto daily cone-beam CT (CBCT) scans for delivered dose accumulation. However, evaluation of registered contours requires human assessment, which is time-consuming and subjects to high inter-observer variability. This work proposes a deep learning model that allows accurate prediction of Dice similarity coefficients (DSC) of registered contours in prostate radiotherapy.Approach. Our dataset comprises 20 prostate cancer patients with 37-39 daily CBCT scans each. The pCT scans and planning contours were deformably registered to each corresponding CBCT scan to generate virtual CT (vCT) scans and registered contours. The DSC score, which is a common contour-based validation metric for registration quality, between the registered and manual contours were computed. A Siamese neural network was trained on the vCT-CBCT image pairs to predict DSC. To assess the performance of the model, the root mean squared error (RMSE) between the actual and predicted DSC were computed.Main results. The model showed promising results for predicting DSC, giving RMSE of 0.070, 0.079 and 0.118 for rectum, prostate, and bladder respectively on the holdout test set. Clinically, a low RMSE implies that the predicted DSC can be reliably used to determine if further DIR assessment from physicians is required. Considering the event where a registered contour is classified as poor if its DSC is below 0.6 and good otherwise, the model achieves an accuracy of 92% for the rectum. A sensitivity of 0.97 suggests that the model can correctly identify 97% of poorly registered contours, allowing manual assessment of DIR to be triggered.Significance. We propose a neural network capable of accurately predicting DSC of deformably registered OAR contours, which can be used to evaluate eligibility for plan adaptation.

Implementing dispersion measurement as part of scanning proton therapy commissioning and quality assurance.

Introduction. Dispersion in an accelerator quantifies the deviation of the proton trajectory when there is a momentum deviation. We present for the first time a safe method of measuring dispersion in the clinic, using a scintillator detector and the momentum deviations within a spill. This is an important accelerator quantity as we found that this is the reason behind the large dose fluctuation in our absolute dosimetry measurement.Methods. Dispersions are measured for nine energies in a Hitachi ProBeat system at three locations (isocenter and at two profile monitors) and at two gantry angles (0 and 90 degrees) by first measuring the spot position and momentum drift within a spill. The spot position drift is measured by the XRV-4000 at the isocenter, and by the two profile monitors located at 0.57 and 2.27 m from the isocenter. The momentum drift is calculated from the intra-spill range drift which is measured using the Ranger accessory. The dispersion at isocenter and its gradient are calculated using the weighted least square regression on the measured dispersions at the three locations. A constraint is formulated on the dispersion and its gradient to ensure minimal intra-spill spot position deviation around the isocenter.Results. The measured intra-spill range and spot positional drift at isocenter are less than0.25mmand0.7mmrespectively. The momentum spread calculated from the range drift are less than 0.08%. The dispersion at the isocenter ranged from0.50to4.30mand the zero-crossing happens upstream of isocenter for all energies. 2 of the 9 energies (168.0 and 187.5 MeV) violated the constraint and has an intra-spill spot positional deviation greater than1.0within5cmfrom the isocenter.Conclusion. This measurement is recommended as part of commissioning and annual quality assurance for accelerator monitoring and to ensure intra-spill spot deviations remain low.

Anatomical change during radiotherapy for head and neck cancer, and its effect on delivered dose to the spinal cord.

BACKGROUND AND PURPOSE: The impact of weight loss and anatomical change during head and neck (H&N) radiotherapy on spinal cord dosimetry is poorly understood, limiting evidence-based adaptive management strategies. MATERIALS AND METHODS: 133 H&N patients treated with daily mega-voltage CT image-guidance (MVCT-IG) on TomoTherapy, were selected. Elastix software was used to deform planning scan SC contours to MVCT-IG scans, and accumulate dose. Planned (DP) and delivered (DA) spinal cord D2% (SCD2%) were compared. Univariate relationships between neck irradiation strategy (unilateral vs bilateral), T-stage, N-stage, weight loss, and changes in lateral separation (LND) and CT slice surface area (SSA) at C1 and the superior thyroid notch (TN), and ΔSCD2% [(DA - DP) D2%] were examined. RESULTS: The mean value for (DA - DP) D2% was -0.07 Gy (95%CI -0.28 to 0.14, range -5.7 Gy to 3.8 Gy), and the mean absolute difference between DP and DA (independent of difference direction) was 0.9 Gy (95%CI 0.76-1.04 Gy). Neck treatment strategy (p = 0.39) and T-stage (p = 0.56) did not affect ΔSCD2%. Borderline significance (p = 0.09) was seen for higher N-stage (N2-3) and higher ΔSCD2%. Mean reductions in anatomical metrics were substantial: weight loss 6.8 kg; C1LND 12.9 mm; C1SSA 12.1 cm2; TNLND 5.3 mm; TNSSA 11.2 cm2, but no relationship between weight loss or anatomical change and ΔSCD2% was observed (all r2 < 0.1). CONCLUSIONS: Differences between delivered and planned spinal cord D2% are small in patients treated with daily IG. Even patients experiencing substantial weight loss or anatomical change during treatment do not require adaptive replanning for spinal cord safety.

Applying physical science techniques and CERN technology to an unsolved problem in radiation treatment for cancer: the multidisciplinary 'VoxTox' research programme.

The VoxTox research programme has applied expertise from the physical sciences to the problem of radiotherapy toxicity, bringing together expertise from engineering, mathematics, high energy physics (including the Large Hadron Collider), medical physics and radiation oncology. In our initial cohort of 109 men treated with curative radiotherapy for prostate cancer, daily image guidance computed tomography (CT) scans have been used to calculate delivered dose to the rectum, as distinct from planned dose, using an automated approach. Clinical toxicity data have been collected, allowing us to address the hypothesis that delivered dose provides a better predictor of toxicity than planned dose.