Takafumi Nemoto1, Natsumi Futakami2, Masamichi Yagi3, Etsuo Kunieda2, Takeshi Akiba2, Atsuya Takeda4, Naoyuki Shigematsu5. 1. Department of Radiology, Keio University School of Medicine, Shinanomachi 35, Shinjuku-ku, Tokyo 160-8582, Japan; Division of Radiation Oncology, Saiseikai Yokohamashi Tobu Hospital, Shimosueyoshi 3-6-1, Tsurumi-ku, Yokohama-shi, Kanagawa 230-8765, Japan. Electronic address: takatohoku@gmail.com. 2. Department of Radiation Oncology, Tokai University School of Medicine, Shimokasuya 143, Isehara-shi, Kanagawa 259-1143, Japan. 3. HPC&AI Business Dept., Platform Technical Engineer Div., System Platform Solution Unit, Fujitsu Limited, World Trade Center Building, 4-1, Hamamatsucho 2-chome, Minato-ku, Tokyo 105-6125, Japan. 4. Radiation Oncology Center, Ofuna Chuo Hospital, Kamakura-shi 247-0056, Japan. 5. Department of Radiology, Keio University School of Medicine, Shinanomachi 35, Shinjuku-ku, Tokyo 160-8582, Japan.
Abstract
PURPOSE: Deep learning has shown great efficacy for semantic segmentation. However, there are difficulties in the collection, labeling and management of medical imaging data, because of ethical complications and the limited number of imaging studies available at a single facility. This study aimed to find a simple and low-cost method to increase the accuracy of deep learning semantic segmentation for radiation therapy of prostate cancer. METHODS: In total, 556 cases with non-contrast CT images for prostate cancer radiation therapy were examined using a two-dimensional U-Net. Initially, all slices were used for the input data. Then, we removed slices of the cranial portions, which were beyond the margins of the bladder and rectum. Finally, the ground truth labels for the bladder and rectum were added as channels to the input for the prostate training dataset. RESULTS: The highest mean dice similarity coefficients (DSCs) for each organ in the test dataset of 56 cases were 0.85 ± 0.05, 0.94 ± 0.04 and 0.85 ± 0.07 for the prostate, bladder and rectum, respectively. Removal of the cranial slices from the original images significantly increased the DSC of the rectum from 0.83 ± 0.09 to 0.85 ± 0.07 (p < 0.05). Adding bladder and rectum information to prostate training without removing the slices significantly increased the DSC of the prostate from 0.79 ± 0.05 to 0.85 ± 0.05 (p < 0.05). CONCLUSIONS: These cost-free approaches may be useful for new applications, which may include updated models and datasets. They may be applicable to other organs at risk (OARs) and clinical targets such as elective nodal irradiation.
PURPOSE: Deep learning has shown great efficacy for semantic segmentation. However, there are difficulties in the collection, labeling and management of medical imaging data, because of ethical complications and the limited number of imaging studies available at a single facility. This study aimed to find a simple and low-cost method to increase the accuracy of deep learning semantic segmentation for radiation therapy of prostate cancer. METHODS: In total, 556 cases with non-contrast CT images for prostate cancer radiation therapy were examined using a two-dimensional U-Net. Initially, all slices were used for the input data. Then, we removed slices of the cranial portions, which were beyond the margins of the bladder and rectum. Finally, the ground truth labels for the bladder and rectum were added as channels to the input for the prostate training dataset. RESULTS: The highest mean dice similarity coefficients (DSCs) for each organ in the test dataset of 56 cases were 0.85 ± 0.05, 0.94 ± 0.04 and 0.85 ± 0.07 for the prostate, bladder and rectum, respectively. Removal of the cranial slices from the original images significantly increased the DSC of the rectum from 0.83 ± 0.09 to 0.85 ± 0.07 (p < 0.05). Adding bladder and rectum information to prostate training without removing the slices significantly increased the DSC of the prostate from 0.79 ± 0.05 to 0.85 ± 0.05 (p < 0.05). CONCLUSIONS: These cost-free approaches may be useful for new applications, which may include updated models and datasets. They may be applicable to other organs at risk (OARs) and clinical targets such as elective nodal irradiation.
Authors: Kerstin Johnsson; Johan Brynolfsson; Hannicka Sahlstedt; Nicholas G Nickols; Matthew Rettig; Stephan Probst; Michael J Morris; Anders Bjartell; Mathias Eiber; Aseem Anand Journal: Eur J Nucl Med Mol Imaging Date: 2021-08-31 Impact factor: 10.057
Authors: Ana Barragán-Montero; Umair Javaid; Gilmer Valdés; Dan Nguyen; Paul Desbordes; Benoit Macq; Siri Willems; Liesbeth Vandewinckele; Mats Holmström; Fredrik Löfman; Steven Michiels; Kevin Souris; Edmond Sterpin; John A Lee Journal: Phys Med Date: 2021-05-09 Impact factor: 2.685