The effects of computed tomography image characteristics and knot spacing on the spatial accuracy of B-spline deformable image registration in the head and neck geometry

Quantification of deformable image registration uncertainties for dose accumulation on head and neck cancer proton treatments

PURPOSE

Head and neck cancer (HNC) patients in radiotherapy require adaptive treatment plans due to anatomical changes. Deformable image registration (DIR) is used in adaptive radiotherapy, e.g. for deformable dose accumulation (DDA). However, DIR's ill-posedness necessitates addressing uncertainties, often overlooked in clinical implementations. DIR's further clinical implementation is hindered by missing quantitative commissioning and quality assurance tools. This study evaluates one pathway for more quantitative DDA uncertainties.

METHODS

For five HNC patients, each with multiple repeated CTs acquired during treatment, a simultaneous-integrated boost (SIB) plan was optimized. Recalculated doses were warped individually using multiple DIRs from repeated to reference CTs, and voxel-by-voxel dose ranges determined an error-bar for DDA. Followed by evaluating, a previously proposed early-stage DDA uncertainty estimation method tested for lung cancer, which combines geometric DIR uncertainties, dose gradients and their directional dependence, in the context of HNC.

RESULTS

Applying multiple DIRs show dose differences, pronounced in high dose gradient regions. The patient with largest anatomical changes (-13.1 % in ROI body volume), exhibited 33 % maximum uncertainty in contralateral parotid, with 54 % of voxels presenting an uncertainty >5 %. Accumulation over multiple CTs partially mitigated uncertainties. The estimation approach predicted 92.6 % of voxels within ±5 % to the reference dose uncertainty across all patients.

CONCLUSIONS

DIR variations impact accumulated doses, emphasizing DDA uncertainty quantification's importance for HNC patients. Multiple DIR dose warping aids in quantifying DDA uncertainties. An estimation approach previously described for lung cancer was successfully validated for HNC, for SIB plans, presenting different dose gradients, and for accumulated treatments.

Glenoid bone strain after anatomical total shoulder arthroplasty: In vitro measurements with micro-CT and digital volume correlation

Glenoid implant loosening remains a major source of failure and concern after anatomical total shoulder arthroplasty (aTSA). It is assumed to be associated with eccentric loading and excessive bone strain, but direct measurement of bone strain after aTSA is not available yet. Therefore, our objective was to develop an in vitro technique for measuring bone strain around a loaded glenoid implant. A custom loading device (1500 N) was designed to fit within a micro-CT scanner, to use digital volume correlation for measuring displacement and calculating strain. Errors were evaluated with three pairs of unloaded scans. The average displacement random error of three pairs of unloaded scans was 6.1 µm. Corresponding systematic and random errors of strain components were less than 806.0 µε and 2039.9 µε, respectively. The average strain accuracy (MAER) and precision (SDER) were 694.3 µε and 440.3 µε, respectively. The loaded minimum principal strain (8738.9 µε) was 12.6 times higher than the MAER (694.3 µε) on average, and was above the MAER for most of the glenoid bone volume (98.1%). Therefore, this technique proves to be accurate and precise enough to eventually compare glenoid implant designs, fixation techniques, or to validate numerical models of specimens under similar loading.

Free-to-use DIR solutions in radiotherapy: Benchmark against commercial platforms through a contour-propagation study

PURPOSE

A contour propagation study has been conducted to benchmark three algorithms for Deformable Image Registration (DIR) freely available online against well-established commercial solutions.

METHODS

ElastiX, BRAINS and Plastimach, available as modules in the open source platform 3DSlicer, were tested as the recent AAPM Task group 132 guidelines proposes. The overlap of the DIR-mapped ROIs in four computational anthropomorphic phantoms was measured. To avoid bias every algorithm was left to run without any human interaction nor particular registration strategy. The accuracy of the algorithms was measured using the Dice Similarity Coefficient (DSC) and Mean Distance to Conformity (MDC) metrics. The registration quality was compared to the recommended geometrical accuracy suggested by AAPM TG132 and to the results of a large population-based study performed with commercial DIR solutions.

RESULTS

The considered free-to-use DIR solutions generally meet acceptable accuracy and good overlap (DSC > 0.85). Mild failures (DSC < 0.75) were detected only for the smallest structures. In case of extremely severe deformations acceptable accuracy was not met (MDC > 3 mm). The morphing capability of the tested algorithms equals those of commercial systems when the user interaction is avoided. Underperformances were detected only in cases where a specific registration strategy is mandatory to obtain a satisfying match.

CONCLUSIONS

All of the considered algorithms show performances not inferior to previously published data and have the potential to be good candidates for use in the clinical routine. The results and conclusions only apply to the considered phantoms and should not be considered to be generally applicable and extendable to patient cases.

An automated, quantitative, and case-specific evaluation of deformable image registration in computed tomography images

A prerequisite for adaptive dose-tracking in radiotherapy is the assessment of the deformable image registration (DIR) quality. In this work, various metrics that quantify DIR uncertainties are investigated using realistic deformation fields of 26 head and neck and 12 lung cancer patients. Metrics related to the physiologically feasibility (the Jacobian determinant, harmonic energy (HE), and octahedral shear strain (OSS)) and numerically robustness of the deformation (the inverse consistency error (ICE), transitivity error (TE), and distance discordance metric (DDM)) were investigated. The deformable registrations were performed using a B-spline transformation model. The DIR error metrics were log-transformed and correlated (Pearson) against the log-transformed ground-truth error on a voxel level. Correlations of r  ⩾  0.5 were found for the DDM and HE. Given a DIR tolerance threshold of 2.0 mm and a negative predictive value of 0.90, the DDM and HE thresholds were 0.49 mm and 0.014, respectively. In conclusion, the log-transformed DDM and HE can be used to identify voxels at risk for large DIR errors with a large negative predictive value. The HE and/or DDM can therefore be used to perform automated quality assurance of each CT-based DIR for head and neck and lung cancer patients.

Influence of dosimetric and clinical criteria on the requirement of artificial nutrition during radiotherapy of head and neck cancer patients

PURPOSE/OBJECTIVE(S)

Intensification of radiotherapy and chemotherapy for head-and-neck cancer (HNC) may lead to increased rates of long term dysphagia as a severe side effect. Mucositis and consequent swallowing problems require artificial nutrition in many HNC patients undergoing radiotherapy or chemoradiation. It is unknown, which predict factors for prophylactic PEG tube insertion appear useful.

MATERIALS/METHODS

From an institutional database, 101 patients (72 male, 29 female, mean age 59.5years) were identified who underwent radiotherapy or chemoradiation for HNC. Primary end point of the investigation was the need for artificial nutrition for more than 4days during radiotherapy. Dose volume parameters of defined normal tissue structures potentially of relevance for swallowing ability as well as clinical factors were used to develop a predictive model using a binary multiple logistic regression model.

RESULTS

Whereas several dosimetric and clinical factors were significant predictors for the need of artificial nutrition on univariate analysis, on multivariate analysis only three factors remained independently significant: mean dose to the oropharynx+1cm circumferential margin, ECOG performance state (0-1 vs. 2-4), and the use of chemotherapy (yes vs. no).

CONCLUSIONS

Using a 3 parameter model we could distinguish HNC-patients with different risks for the need of artificial nutrition during radiotherapy. After independent validation, the model could be helpful to decision on prophylactic PEG tube insertion.