Thibault Marin1, Yue Zhuo1, Rita Maria Lahoud1, Fei Tian1, Xiaoyue Ma1, Fangxu Xing1, Maryam Moteabbed2, Xiaofeng Liu1, Kira Grogg1, Nadya Shusharina3, Jonghye Woo1, Ruth Lim1, Chao Ma1, Yen-Lin E Chen2, Georges El Fakhri4. 1. Gordon Center for Medical Imaging, Department of Radiology, Massachusetts General Hospital, Boston, United States; Harvard Medical School, Boston, United States. 2. Gordon Center for Medical Imaging, Department of Radiology, Massachusetts General Hospital, Boston, United States; Department of Radiation Oncology, Massachusetts General Hospital, Boston, United States; Harvard Medical School, Boston, United States. 3. Department of Radiation Oncology, Massachusetts General Hospital, Boston, United States; Harvard Medical School, Boston, United States. 4. Gordon Center for Medical Imaging, Department of Radiology, Massachusetts General Hospital, Boston, United States; Harvard Medical School, Boston, United States. Electronic address: elfakhri.georges@mgh.harvard.edu.
Abstract
BACKGROUND AND PURPOSE: The delineation of the gross tumor volume (GTV) is a critical step for radiation therapy treatment planning. The delineation procedure is typically performed manually which exposes two major issues: cost and reproducibility. Delineation is a time-consuming process that is subject to inter- and intra-observer variability. While methods have been proposed to predict GTV contours, typical approaches ignore variability and therefore fail to utilize the valuable confidence information offered by multiple contours. MATERIALS AND METHODS: In this work we propose an automatic GTV contouring method for soft-tissue sarcomas from X-ray computed tomography (CT) images, using deep learning by integrating inter- and intra-observer variability in the learned model. Sixty-eight patients with soft tissue and bone sarcomas were considered in this evaluation, all underwent pre-operative CT imaging used to perform GTV delineation. Four radiation oncologists and radiologists performed three contouring trials each for all patients. We quantify variability by defining confidence levels based on the frequency of inclusion of a given voxel into the GTV and use a deep convolutional neural network to learn GTV confidence maps. RESULTS: Results were compared to confidence maps from the four readers as well as ground-truth consensus contours established jointly by all readers. The resulting continuous Dice score between predicted and true confidence maps was 87% and the Hausdorff distance was 14 mm. CONCLUSION: Results demonstrate the ability of the proposed method to predict accurate contours while utilizing variability and as such it can be used to improve clinical workflow.
BACKGROUND AND PURPOSE: The delineation of the gross tumor volume (GTV) is a critical step for radiation therapy treatment planning. The delineation procedure is typically performed manually which exposes two major issues: cost and reproducibility. Delineation is a time-consuming process that is subject to inter- and intra-observer variability. While methods have been proposed to predict GTV contours, typical approaches ignore variability and therefore fail to utilize the valuable confidence information offered by multiple contours. MATERIALS AND METHODS: In this work we propose an automatic GTV contouring method for soft-tissue sarcomas from X-ray computed tomography (CT) images, using deep learning by integrating inter- and intra-observer variability in the learned model. Sixty-eight patients with soft tissue and bone sarcomas were considered in this evaluation, all underwent pre-operative CT imaging used to perform GTV delineation. Four radiation oncologists and radiologists performed three contouring trials each for all patients. We quantify variability by defining confidence levels based on the frequency of inclusion of a given voxel into the GTV and use a deep convolutional neural network to learn GTV confidence maps. RESULTS: Results were compared to confidence maps from the four readers as well as ground-truth consensus contours established jointly by all readers. The resulting continuous Dice score between predicted and true confidence maps was 87% and the Hausdorff distance was 14 mm. CONCLUSION: Results demonstrate the ability of the proposed method to predict accurate contours while utilizing variability and as such it can be used to improve clinical workflow.
Authors: Dian Wang; Walter Bosch; David G Kirsch; Rawan Al Lozi; Issam El Naqa; David Roberge; Steven E Finkelstein; Ivy Petersen; Michael Haddock; Yen-Lin E Chen; Naoyuki G Saito; Ying J Hitchcock; Aaron H Wolfson; Thomas F DeLaney Journal: Int J Radiat Oncol Biol Phys Date: 2011-01-27 Impact factor: 7.038
Authors: Dian Wang; Qiang Zhang; Burton L Eisenberg; John M Kane; X Allen Li; David Lucas; Ivy A Petersen; Thomas F DeLaney; Carolyn R Freeman; Steven E Finkelstein; Ying J Hitchcock; Manpreet Bedi; Anurag K Singh; George Dundas; David G Kirsch Journal: J Clin Oncol Date: 2015-02-09 Impact factor: 44.544
Authors: Chris McIntosh; Leigh Conroy; Alejandro Berlin; Thomas G Purdie; Michael C Tjong; Tim Craig; Andrew Bayley; Charles Catton; Mary Gospodarowicz; Joelle Helou; Naghmeh Isfahanian; Vickie Kong; Tony Lam; Srinivas Raman; Padraig Warde; Peter Chung Journal: Nat Med Date: 2021-06-03 Impact factor: 53.440
Authors: Carlos E Cardenas; Rachel E McCarroll; Laurence E Court; Baher A Elgohari; Hesham Elhalawani; Clifton D Fuller; Mona J Kamal; Mohamed A M Meheissen; Abdallah S R Mohamed; Arvind Rao; Bowman Williams; Andrew Wong; Jinzhong Yang; Michalis Aristophanous Journal: Int J Radiat Oncol Biol Phys Date: 2018-02-07 Impact factor: 7.038