Premium
WE‐D‐BRE‐06: Quantification of Dose‐Response for High Grade Esophagtis Patients Using a Novel Voxel‐To‐Voxel Method
Author(s) -
Niedzielski J,
Yang J,
Martel M,
Tucker S,
Gomez D,
Briere T,
Court L
Publication year - 2014
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4889396
Subject(s) - voxel , nuclear medicine , esophagitis , medicine , dosimetry , radiation treatment planning , volume rendering , population , radiology , radiation therapy , artificial intelligence , computer science , environmental health , reflux , visualization , disease
Purpose: Radiation induces an inflammatory response in the esophagus, discernible on CT studies. This work objectively quantifies the voxel esophageal radiation‐response for patients with acute esophagitis. This knowledge is an important first‐step towards predicting the effect of complex dose distributions on patient esophagitis symptoms. Methods: A previously validated voxel‐based methodology of quantifying radiation esophagitis severity was used to identify the voxel dose‐response for 18 NSCLC patients with severe esophagitis (CTCAE grading criteria, grade2 or higher). The response is quantified as percent voxel volume change for a given dose. During treatment (6–8 weeks), patients had weekly 4DCT studies and esophagitis scoring. Planning CT esophageal contours were deformed to each weekly CT using a demons DIR algorithm. An algorithm using the Jacobian Map from the DIR of the planning CT to all weekly CTs was used to quantify voxel‐volume change, along with corresponding delivered voxel dose, to the planning voxel. Dose for each voxel for each time‐point was calculated on each previous weekly CT image, and accumulated using DIR. Thus, for each voxel, the volume‐change and delivered dose was calculated for each time‐point. The data was binned according to when the volume‐change first increased by a threshold volume (10%–100%, in 10% increments), and the average delivered dose calculated for each bin. Results: The average dose resulting in a voxel volume increase of 10–100% was 21.6 to 45.9Gy, respectively. The mean population dose to give a 50% volume increase was 36.3±4.4Gy, (range:29.8 to 43.5Gy). The average week of 50% response was 4.1 (range:4.9 to 2.8 weeks). All 18 patients showed similar dose to first response curves, showing a common trend in the initial inflammatoryresponse. Conclusion: We extracted the dose‐response curve of the esophagus on a voxel‐to‐voxel level. This may be useful for estimating the esophagus response (and patient symptoms) to complicated dose distributions.