Kaicheng Pan1, Lei Zhao2, Song Gu3, Yi Tang1, Jiahao Wang4, Wen Yu2, Lucheng Zhu4, Qi Feng5, Ruipeng Su6, Zhiyong Xu2, Xiadong Li4, Zhongxiang Ding5, Xiaolong Fu2, Shenglin Ma1, Jun Yan6, Shigong Kang6, Tao Zhou7, Bing Xia8. 1. Department of Radiation Oncology, The Affiliated Hangzhou Hospital of Nanjing Medical University, Hangzhou, China. 2. Department of Radiation Oncology, Shanghai Chest Hospital, Shanghai Jiao Tong University, Shanghai, China. 3. Department of Radiation Oncology, Hangzhou Yikang Chinese Medicine Oncology Hospital, Hangzhou, China. 4. Department of Radiation Oncology, Key Laboratory of Clinical Cancer Pharmacology and Toxicology Research of Zhejiang Province, Affiliated Hangzhou Cancer Hospital, Hangzhou, China. 5. Department of Radiology, Key Laboratory of Clinical Cancer Pharmacology and Toxicology Research of Zhejiang Province, Affiliated Hangzhou First People's Hospital, Hangzhou, China. 6. Beijing Allcure Medical Technology Group Co., Ltd., Beijing, China. 7. Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, Jinan, China. zt1973sd@163.com. 8. Department of Radiation Oncology, The Affiliated Hangzhou Hospital of Nanjing Medical University, Hangzhou, China. bingxia_hzch@163.com.
Abstract
BACKGROUND: Whole brain radiotherapy (WBRT) can impair patients' cognitive function. Hippocampal avoidance during WBRT can potentially prevent this side effect. However, manually delineating the target area is time-consuming and difficult. Here, we proposed a credible approach of automatic hippocampal delineation based on convolutional neural networks. METHODS: Referring to the hippocampus contouring atlas proposed by RTOG 0933, we manually delineated (MD) the hippocampus on the MRI data sets (3-dimensional T1-weighted with slice thickness of 1 mm, n = 175), which were used to construct a three-dimensional convolutional neural network aiming for the hippocampus automatic delineation (AD). The performance of this AD tool was tested on three cohorts: (a) 3D T1 MRI with 1-mm slice thickness (n = 30); (b) non-3D T1-weighted MRI with 3-mm slice thickness (n = 19); (c) non-3D T1-weighted MRI with 1-mm slice thickness (n = 11). All MRIs confirmed with normal hippocampus has not been violated by any disease. Virtual radiation plans were created for AD and MD hippocampi in cohort c to evaluate the clinical feasibility of the artificial intelligence approach. Statistical analyses were performed using SPSS version 23. P < 0.05 was considered significant. RESULTS: The Dice similarity coefficient (DSC) and Average Hausdorff Distance (AVD) between the AD and MD hippocampi are 0.86 ± 0.028 and 0.18 ± 0.050 cm in cohort a, 0.76 ± 0.035 and 0.31 ± 0.064 cm in cohort b, 0.80 ± 0.015 and 0.24 ± 0.021 cm in cohort c, respectively. The DSC and AVD in cohort a were better than those in cohorts b and c (P < 0.01). There is no significant difference between the radiotherapy plans generated using the AD and MD hippocampi. CONCLUSION: The AD of the hippocampus based on a deep learning algorithm showed satisfying results, which could have a positive impact on improving delineation accuracy and reducing work load.
BACKGROUND: Whole brain radiotherapy (WBRT) can impair patients' cognitive function. Hippocampal avoidance during WBRT can potentially prevent this side effect. However, manually delineating the target area is time-consuming and difficult. Here, we proposed a credible approach of automatic hippocampal delineation based on convolutional neural networks. METHODS: Referring to the hippocampus contouring atlas proposed by RTOG 0933, we manually delineated (MD) the hippocampus on the MRI data sets (3-dimensional T1-weighted with slice thickness of 1 mm, n = 175), which were used to construct a three-dimensional convolutional neural network aiming for the hippocampus automatic delineation (AD). The performance of this AD tool was tested on three cohorts: (a) 3D T1 MRI with 1-mm slice thickness (n = 30); (b) non-3D T1-weighted MRI with 3-mm slice thickness (n = 19); (c) non-3D T1-weighted MRI with 1-mm slice thickness (n = 11). All MRIs confirmed with normal hippocampus has not been violated by any disease. Virtual radiation plans were created for AD and MD hippocampi in cohort c to evaluate the clinical feasibility of the artificial intelligence approach. Statistical analyses were performed using SPSS version 23. P < 0.05 was considered significant. RESULTS: The Dice similarity coefficient (DSC) and Average Hausdorff Distance (AVD) between the AD and MD hippocampi are 0.86 ± 0.028 and 0.18 ± 0.050 cm in cohort a, 0.76 ± 0.035 and 0.31 ± 0.064 cm in cohort b, 0.80 ± 0.015 and 0.24 ± 0.021 cm in cohort c, respectively. The DSC and AVD in cohort a were better than those in cohorts b and c (P < 0.01). There is no significant difference between the radiotherapy plans generated using the AD and MD hippocampi. CONCLUSION: The AD of the hippocampus based on a deep learning algorithm showed satisfying results, which could have a positive impact on improving delineation accuracy and reducing work load.
Authors: Dian Wang; Walter Bosch; David Roberge; Steven E Finkelstein; Ivy Petersen; Michael Haddock; Yen-Lin E Chen; Naoyuki G Saito; David G Kirsch; Ying J Hitchcock; Aaron H Wolfson; Thomas F DeLaney Journal: Int J Radiat Oncol Biol Phys Date: 2011-06-15 Impact factor: 7.038
Authors: M Mukesh; R Benson; R Jena; A Hoole; T Roques; C Scrase; C Martin; G A Whitfield; J Gemmill; S Jefferies Journal: Br J Radiol Date: 2012-08 Impact factor: 3.039
Authors: Vinai Gondi; Stephanie L Pugh; Wolfgang A Tome; Chip Caine; Ben Corn; Andrew Kanner; Howard Rowley; Vijayananda Kundapur; Albert DeNittis; Jeffrey N Greenspoon; Andre A Konski; Glenn S Bauman; Sunjay Shah; Wenyin Shi; Merideth Wendland; Lisa Kachnic; Minesh P Mehta Journal: J Clin Oncol Date: 2014-10-27 Impact factor: 44.544
Authors: Jonathan A Ledermann; Allan Hackshaw; Stan Kaye; Gordon Jayson; Hani Gabra; Iain McNeish; Helena Earl; Tim Perren; Martin Gore; Mojca Persic; Malcolm Adams; Lindsay James; Graham Temple; Michael Merger; Gordon Rustin Journal: J Clin Oncol Date: 2011-08-22 Impact factor: 44.544
Authors: David J Madden; Julia Spaniol; Matthew C Costello; Barbara Bucur; Leonard E White; Roberto Cabeza; Simon W Davis; Nancy A Dennis; James M Provenzale; Scott A Huettel Journal: J Cogn Neurosci Date: 2009-02 Impact factor: 3.225
Authors: Eric L Chang; Jeffrey S Wefel; Kenneth R Hess; Pamela K Allen; Frederick F Lang; David G Kornguth; Rebecca B Arbuckle; J Michael Swint; Almon S Shiu; Moshe H Maor; Christina A Meyers Journal: Lancet Oncol Date: 2009-10-02 Impact factor: 41.316