Ivan Vazquez | Searcharxiv

arXiv2022

Deep Learning-Based Dose Prediction for Automated, Individualized Quality Assurance of Head and Neck Radiation Therapy Plans

Purpose: This study aimed to use deep learning-based dose prediction to assess head and neck (HN) plan quality and identify suboptimal plans. Methods: A total of 245 VMAT HN plans were created using RapidPlan knowledge-based planning (KBP). A subset of 112 high-quality plans was selected under the supervision of an HN radiation oncologist. We trained a 3D Dense Dilated U-Net architecture to predict 3-dimensional dose distributions using 3-fold cross-validation on 90 plans. Model inputs included CT images, target prescriptions, and contours for targets and organs at risk (OARs). The model's performance was assessed on the remaining 22 test plans. We then tested the application of the dose prediction model for automated review of plan quality. Dose distributions were predicted on 14 clinical plans. The predicted versus clinical OAR dose metrics were compared to flag OARs with suboptimal normal tissue sparing using a 2 Gy dose difference or 3% dose-volume threshold. OAR flags were compared to manual flags by 3 HN radiation oncologists. Results: The predicted dose distributions were of comparable quality to the KBP plans. The differences between the predicted and KBP-planned D1%, D95%, and D99% across the targets were within -2.53%(SD=1.34%), -0.42%(SD=1.27%), and -0.12%(SD=1.97%), respectively, and the OAR mean and maximum doses were within -0.33Gy(SD=1.40Gy) and -0.96Gy(SD=2.08Gy). For the plan quality assessment study, radiation oncologists flagged 47 OARs for possible plan improvement. There was high interphysician variability; 83% of physician-flagged OARs were flagged by only one of 3 physicians. The comparative dose prediction model flagged 63 OARs, including 30 of 47 physician-flagged OARs. Conclusion: Deep learning can predict high-quality dose distributions, which can be used as comparative dose distributions for automated, individualized assessment of HN plan quality.

physics.med-ph↗

arXiv2020

Hints for icy pebble migration feeding an oxygen-rich chemistry in the inner planet-forming region of disks

We present a synergic study of protoplanetary disks to investigate links between inner disk gas molecules and the large-scale migration of solid pebbles. The sample includes 63 disks where two types of measurements are available: i) spatially-resolved disk images revealing the radial distribution of disk pebbles (mm-cm dust grains), from millimeter observations with ALMA or the SMA, and ii) infrared molecular emission spectra as observed with Spitzer. The line flux ratios of H2O with HCN, C2H2, and CO2 all anti-correlate with the dust disk radius R$_{dust}$, expanding previous results found by Najita et al. (2013) for HCN/H2O and the dust disk mass. By normalization with the dependence on accretion luminosity common to all molecules, only the H2O luminosity maintains a detectable anti-correlation with disk radius, suggesting that the strongest underlying relation is between H2O and R$_{dust}$. If R$_{dust}$ is set by large-scale pebble drift, and if molecular luminosities trace the elemental budgets of inner disk warm gas, these results can be naturally explained with scenarios where the inner disk chemistry is fed by sublimation of oxygen-rich icy pebbles migrating inward from the outer disk. Anti-correlations are also detected between all molecular luminosities and the infrared index n$_{13-30}$, which is sensitive to the presence and size of an inner disk dust cavity. Overall, these relations suggest a physical interconnection between dust and gas evolution both locally and across disk scales. We discuss fundamental predictions to test this interpretation and study the interplay between pebble drift, inner disk depletion, and the chemistry of planet-forming material.

astro-ph.EP↗