Background and purpose
A rapid learning approach has been proposed to extract and apply knowledge from routine care data rather than solely relying on clinical trial evidence. To validate this in practice we deployed a previously developed decision support system (DSS) in a typical, busy clinic for non-small cell lung cancer (NSCLC) patients.
Material and methods
Gender, age, performance status, lung function, lymph node status, tumor volume and survival were extracted without review from clinical data sources for lung cancer patients. With these data the DSS was tested to predict overall survival.
Results
3919 lung cancer patients were identified with 159 eligible for inclusion, due to ineligible histology or stage, non-radical dose, missing tumor volume or survival. The DSS successfully identified a good prognosis group and a medium/poor prognosis group (2 year OS 69% vs. 27/30%, p < 0.001). Stage was less discriminatory (2 year OS 47% for stage I–II vs. 36% for stage IIIA–IIIB, p = 0.12) with most good prognosis patients having higher stage disease. The DSS predicted a large absolute overall survival benefit (~40%) for a radical dose compared to a non-radical dose in patients with a good prognosis, while no survival benefit of radical radiotherapy was predicted for patients with a poor prognosis.
Conclusions
A rapid learning environment is possible with the quality of clinical data sufficient to validate a DSS. It uses patient and tumor features to identify prognostic groups in whom therapy can be individualized based on predicted outcomes. Especially the survival benefit of a radical versus non-radical dose predicted by the DSS for various prognostic groups has clinical relevance, but needs to be prospectively validated.
Conventional 4DCBCT captures 1320 projections across 4 min. Adaptive 4DCBCT has been developed to reduce imaging dose and scan time. This study investigated reconstruction algorithms that best complement adaptive 4DCBCT acquisition for reducing imaging dose and scan time whilst maintaining or improving image quality compared to conventional 4DCBCT acquisition using real patient data from the first 10 adaptive 4DCBCT patients. Adaptive 4DCBCT was implemented in the ADaptive CT Acquisition for Personalized Thoracic imaging clinical trial. Adaptive 4DCBCT modulates gantry rotation speed and kV acquisition rate in response to the patient’s real-time respiratory signal, ensuring even angular spacing between projections at each respiratory phase. We examined the first 10 lung cancer radiotherapy patients that received adaptive 4DCBCT. Fast, 200-projection scans over 60–80 s, and slower, 600-projection scans over ∼240 s, were obtained after routine patient treatment and compared against conventional 4DCBCT acquisition. Adaptive 4DCBCT acquisitions were reconstructed using Feldkamp−Davis−Kress (FDK), McKinnon–Bates (MKB), Motion Compensated FDK (MCFDK) and Motion Compensated MKB (MCMKB) algorithms. Reconstructions were assessed via, Structural SIMilarity (SSIM), Signal-to-Noise-Ratio (SNR), Contrast-to-Noise-Ratio (CNR), Tissue Interface Sharpness of Diaphragm (TIS-D) and Tumor (TIS-T). The 200- and 600-projection adaptive 4DCBCT acquisition corresponded to 85% and 55% reduction in imaging dose, shorter and similar scan times of approximately 90 s and 236 s respectively, compared to conventional 4DCBCT acquisition. 200- and 600-projection adaptive 4DCBCT reconstructions achieved more than 0.900 SSIM relative to conventional 4DCBCT acquisition. Compared to conventional 4DCBCT acquisition, 200-projection adaptive 4DCBCT reconstructions achieved higher SNR, CNR, TIS-T, TIS-D with motion compensated algorithms, MCFDK (208%, 159%, 174%, 247%) and MCMKB (214%, 173%, 266%, 245%) respectively. The 200-projection adaptive 4DCBCT MCFDK- and MCMKB-reconstruction results show image quality improvements are possible even with 85% fewer projections acquired. We established acquisition-reconstruction protocols that provide substantial reductions in imaging time and dose whilst improving image quality.
Purpose
Dose deposition measurements for parallel MRI‐linacs have previously only shown comparisons between 0 T and a single available magnetic field. The Australian MRI‐Linac consists of a magnet coupled with a dual energy linear accelerator and a 120 leaf Multi‐Leaf Collimator with the radiation beam parallel to the magnetic field. Two different magnets, with field strengths of 1 and 1.5 T, were used during prototyping. This work aims to characterize the impact of the magnetic field at 1 and 1.5 T on dose deposition, possible by comparing dosimetry measured at both magnetic field strengths to measurements without the magnetic field.
Methods
Dose deposition measurements focused on a comparison of beam quality (TPR20/10), PDD, profiles at various depths, surface doses, and field size output factors. Measurements were acquired at 0, 1, and 1.5 T. Beam quality was measured using an ion chamber in solid water at isocenter with appropriate TPR20/10 buildup. PDDs and profiles were acquired via EBT3 film placed in solid water either parallel or perpendicular to the radiation beam. Films at surface were used to determine surface dose. Output factors were measured in solid water using an ion chamber at isocenter with 10 cm solid water buildup.
Results
Beam quality was within ±0.5% of the 0 T value for the 1 and 1.5 T magnetic field strengths. PDDs and profiles showed agreement for the three magnetic field strengths at depths beyond 20 mm. Deposited dose increased at shallower depths due to electron focusing. Output factors showed agreement within 1%.
Conclusion
Dose deposition at depth for a parallel MRI‐linac was not significantly impacted by either a 1 or 1.5 T magnetic field. PDDs and profiles at shallow depths and surface dose measurements showed significant differences between 0, 1, and 1.5 T due to electron focusing.
CORRESPONDENCE dopamine. In our two healthy volunteers the oppo site may have happened, and the combination of amphetamine with haloperidol may have caused marked dystonia by potentiation of dopamine release.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.