ABSTRACT
STUDY DESIGN: Retrospective analysis of a registered cohort of patients treated and irradiated for metastases in the spinal column in a single institute. OBJECTIVE: This is the first study to develop and internally validate radiomics features for predicting six-month survival probability for patients with spinal bone metastases (SBM). BACKGROUND DATA: Extracted radiomics features from routine clinical CT images can be used to identify textural and intensity-based features unperceivable to human observers and associate them with a patient survival probability or disease progression. METHODS: A study was conducted on 250 patients treated for metastases in the spinal column irradiated for the first time between 2014 and 2016, at the MAASTRO clinic in Maastricht, the Netherlands. The first 150 available patients were used to develop the model and the subsequent 100 patient were considered as a test set for the model. A bootstrap (B = 400) stepwise model selection, which combines both the forward and backward variable elimination procedure, was used to select the most useful predictive features from the training data based on the Akaike information criterion (AIC). The stepwise selection procedure was applied to the 400 bootstrap samples, and the results were plotted as a histogram to visualize how often each variable was selected. Only variables selected more than 90 % of the time over the bootstrap runs were used to build the final model.A prognostic index (PI) called radiomics score (radscore) and clinical score (clinscore) was calculated for each patient. The prognostic index was not scaled, the original values were used which can be extracted from the model directly or calculated as a linear combination of the variables in the model multiplied by the respective beta value for each patient. RESULTS: The clinical model had a good discrimination power. The radiomics model, on the other hand, had an inferior performance with no added predictive power to the clinical model. The internal imaging characteristics do not seem to have a value in the prediction of survival. However, the Shape features were excluded from further analyses in our study since all biopsies had a standard shape hence no variability.
ABSTRACT
Monte Carlo proton dose calculations requires mass densities calculated from the patient CT image. This work investigates the impact of different single-energy CT (SECT) and dual-energy CT (DECT) to density conversion methods in proton dose distributions for brain tumours.Material and methods: Head CT scans for four patients were acquired in SECT and DECT acquisition modes. Commercial software was used to reconstruct DirectDensity™ images in Relative Electron Densities (RED, [Formula: see text]) and to obtain DECT-based pseudo-monoenergetic images (PMI). PMI and SECT images were converted to RED using piecewise linear interpolations calibrated on a head-sized phantom, these fits were referred to as "PMI2RED" and "CT2RED". Two DECT-based calibration methods ("Hünemohr-15it" and "Saito-15it") were also investigated. [Formula: see text] images were converted to mass-densities ([Formula: see text]) to investigate [Formula: see text]differences and one representative patient case was used to make a proton treatment plan. Using CT2RED as reference method, dose distribution differences in the target and in five organs-at-risk (OARs) were quantified.Results: In the phantom study, Saito-15it and Hünemohr-15it produced the lowest [Formula: see text]root-mean-square error (0.7%) and DirectDensity™ the highest error (2.7%). The proton plan evaluated in the Saito-15it and Hünemohr-15it datasets showed the largest relative differences compared to initial CT2RED plan down to -6% of the prescribed dose. Compared to CT2RED, average range differences were calculated: -0.1 ± 0.3 mm for PMI2RED; -0.8 ± 0.4 mm for Hünemohr-15it, and -1.2 ± 0.4 mm for Saito-15it.Conclusion: Given the wide choice of available conversion methods, studies investigating the density accuracy for proton dose calculations are necessary. However, there is still a gap between performing accuracy studies in reference [Formula: see text]phantoms and applying these methods in human CT images. For this treatment case, the PMI2RED method was equivalent to the conventional CT2RED method in terms of dose distribution, CTV coverage and OAR sparing, whereas Hünemohr-15it and Saito-15it presented the largest differences.