Image feature reproducibility and redundancy depended on both the CT machine and the CT image type. For each image type, the authors found a set of cross-machine reproducible, nonredundant, and informative image features that would be useful for future image-based models. Compared to end-exhale 4D-CT and breath-hold 3D-CT, average 4D-CT derived image features showed superior multimachine reproducibility and are the best candidates for clinical correlation.

We have successfully incorporated CT radiomics features into a framework for building predictive RP models via LASSO logistic regression. Addition of normal lung image features produced superior model performance relative to traditional dosimetric and clinical predictors of RP, suggesting that pretreatment CT radiomics features should be considered in the context of RP prediction.

No abstract

The ability of a commercially available dual bias, dual MOSFET dosimetry system to measure therapeutic doses reproducibly throughout its vendor-defined dose-based lifetime has been evaluated by characterizing its sensitivity variation to integrated/cumulative doses from,high-energy (6 and 15 MV) photon radiotherapy beams. The variation of sensitivity as a function of total integrated dose was studied for three different dose-per-fraction levels; namely, 50, 200, and 1200 cGy/fraction. In standard sensitivity mode (i.e., measurements involving dose-per-fraction levels > or =100 cGy), the response of the MOSFET system to identical irradiations increased with integrated dose for both energies investigated. Dose measurement reproducibility for the low (i.e., 50 cGy) dose fractions was within 2.1% (if the system was calibrated before each in-phantom measurement) and 3.1% [if the system was calibrated prior to first use, with no intermediate calibration(s)]. Similarly, dose measurement reproducibility was between 2.2% and 6.6% for the conventional (i.e., 200 cGy) dose fractions and between 1.8% and 7.9% for escalated (i.e., 1200 cGy) dose fractions. The results of this study suggest that, due to the progressively increasing sensitivity resulting from the dual-MOSFET design, frequent calibrations are required to achieve measurement accuracy of < or =3% (within one standard deviation).

No abstract