Wei Zhao1, Liyue Shen1, Bin Han1, Yong Yang1, Kai Cheng1, Diego A S Toesca1, Albert C Koong2, Daniel T Chang1, Lei Xing3. 1. Department of Radiation Oncology, Stanford University, Stanford, California. 2. Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas. 3. Department of Radiation Oncology, Stanford University, Stanford, California. Electronic address: lei@stanford.edu.
Abstract
PURPOSE: Deep learning is an emerging technique that allows us to capture imaging information beyond the visually recognizable level of a human being. Because of the anatomic characteristics and location, on-board target verification for radiation delivery to pancreatic tumors is a challenging task. Our goal was to use a deep neural network to localize the pancreatic tumor target on kV x-ray images acquired using an on-board imager for image guided radiation therapy. METHODS AND MATERIALS: The network is set up in such a way that the input is either a digitally reconstructed radiograph image or a monoscopic x-ray projection image acquired by the on-board imager from a given direction, and the output is the location of the planning target volume in the projection image. To produce a sufficient number of training x-ray images reflecting the vast number of possible clinical scenarios of anatomy distribution, a series of changes were introduced to the planning computed tomography images, including deformation, rotation, and translation, to simulate inter- and intrafractional variations. After model training, the accuracy of the model was evaluated by retrospectively studying patients who underwent pancreatic cancer radiation therapy. Statistical analysis using mean absolute differences (MADs) and Lin's concordance correlation coefficient were used to assess the accuracy of the predicted target positions. RESULTS: MADs between the model-predicted and the actual positions were found to be less than 2.60 mm in anteroposterior, lateral, and oblique directions for both axes in the detector plane. For comparison studies with and without fiducials, MADs are less than 2.49 mm. For all cases, Lin's concordance correlation coefficients between the predicted and actual positions were found to be better than 93%, demonstrating the success of the proposed deep learning for image guided radiation therapy. CONCLUSIONS: We demonstrated that markerless pancreatic tumor target localization is achievable with high accuracy by using a deep learning technique approach.
PURPOSE: Deep learning is an emerging technique that allows us to capture imaging information beyond the visually recognizable level of a human being. Because of the anatomic characteristics and location, on-board target verification for radiation delivery to pancreatic tumors is a challenging task. Our goal was to use a deep neural network to localize the pancreatic tumor target on kV x-ray images acquired using an on-board imager for image guided radiation therapy. METHODS AND MATERIALS: The network is set up in such a way that the input is either a digitally reconstructed radiograph image or a monoscopic x-ray projection image acquired by the on-board imager from a given direction, and the output is the location of the planning target volume in the projection image. To produce a sufficient number of training x-ray images reflecting the vast number of possible clinical scenarios of anatomy distribution, a series of changes were introduced to the planning computed tomography images, including deformation, rotation, and translation, to simulate inter- and intrafractional variations. After model training, the accuracy of the model was evaluated by retrospectively studying patients who underwent pancreatic cancer radiation therapy. Statistical analysis using mean absolute differences (MADs) and Lin's concordance correlation coefficient were used to assess the accuracy of the predicted target positions. RESULTS: MADs between the model-predicted and the actual positions were found to be less than 2.60 mm in anteroposterior, lateral, and oblique directions for both axes in the detector plane. For comparison studies with and without fiducials, MADs are less than 2.49 mm. For all cases, Lin's concordance correlation coefficients between the predicted and actual positions were found to be better than 93%, demonstrating the success of the proposed deep learning for image guided radiation therapy. CONCLUSIONS: We demonstrated that markerless pancreatic tumor target localization is achievable with high accuracy by using a deep learning technique approach.
Authors: Liv Veldeman; Indira Madani; Frank Hulstaert; Gert De Meerleer; Marc Mareel; Wilfried De Neve Journal: Lancet Oncol Date: 2008-04 Impact factor: 41.316
Authors: Andrew J Shepard; Charles K Matrosic; Jeffrey L Radtke; Sydney A Jupitz; Wesley S Culberson; Bryan P Bednarz Journal: Med Phys Date: 2018-10-10 Impact factor: 4.071
Authors: Mouen A Khashab; Katherine J Kim; Erik J Tryggestad; Aaron T Wild; Teboh Roland; Vikesh K Singh; Anne Marie Lennon; Eun Ji Shin; Mark A Ziegler; Reem Z Sharaiha; Marcia Irene Canto; Joseph M Herman Journal: Gastrointest Endosc Date: 2012-11 Impact factor: 9.427
Authors: Priya Jayachandran; A Yuriko Minn; Jacques Van Dam; Jeffrey A Norton; Albert C Koong; Daniel T Chang Journal: Int J Radiat Oncol Biol Phys Date: 2009-10-30 Impact factor: 7.038
Authors: Kenny H Cha; Lubomir Hadjiiski; Heang-Ping Chan; Alon Z Weizer; Ajjai Alva; Richard H Cohan; Elaine M Caoili; Chintana Paramagul; Ravi K Samala Journal: Sci Rep Date: 2017-08-18 Impact factor: 4.379