| Presentation | 2022-01-26 Deep Learning based 2D/3D deformable Image Registration for Abdominal Organs Ryuto Miura, Megumi Nakao, Mitsuhiro Nakamura, Tetsuya Matsuda, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | 2D/3D image registration is a problem that solves the deformation and alignment of a pre-treatment 3D image to a 2D projection image, which is available for treatment support and biomedical analysis. Conventional optimization-based methods widely studied for skeletal structures have problems due to calculation cost and unstable convergence characteristics. Specifically, as the abdominal organs are greatly deformed, and the contours are not detected on X-ray images, no studies have reported 3D image reconstruction from a single 2D projected image. In this study, we propose a supervised deep learning framework that achieves 2D/3D deformable image registration between the 3D image and a single viewpoint 2D projected image. The proposed method learns the translation from the target 2D projection images and the initial 3D image to 3D displacement fields by 3D U-Net. We registered 3D-CT images to the digitally reconstructed radiographs generated from abdominal 4D-CT images and confirmed that the CT images reflecting the respiratory organ motion were reconstructed. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | 2D/3D registration / deformable image registration / displacement field / Convolutional Neural Network |
| Paper # | MI2021-62 |
| Date of Issue | 2022-01-18 (MI) |
| Conference Information | |
| Committee | MI |
|---|---|
| Conference Date | 2022/1/25(3days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Hidekata Hontani(Nagoya Inst. of Tech.) |
| Vice Chair | Hideaki Haneishi(Chiba Univ.) / Takayuki Kitasaka(Aichi Inst. of Tech.) |
| Secretary | Hideaki Haneishi(Yamaguchi Univ.) / Takayuki Kitasaka(Univ. of Hyogo) |
| Assistant | Hotaka Takizawa(Tsukuba Univ.) / Yoshito Otake(NAIST) |
| Paper Information | |
| Registration To | Technical Committee on Medical Imaging |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Deep Learning based 2D/3D deformable Image Registration for Abdominal Organs |
| Sub Title (in English) | |
| Keyword(1) | 2D/3D registration |
| Keyword(2) | deformable image registration |
| Keyword(3) | displacement field |
| Keyword(4) | Convolutional Neural Network |
| 1st Author's Name | Ryuto Miura |
| 1st Author's Affiliation | Kyoto University(Kyoto Univ.) |
| 2nd Author's Name | Megumi Nakao |
| 2nd Author's Affiliation | Kyoto University(Kyoto Univ.) |
| 3rd Author's Name | Mitsuhiro Nakamura |
| 3rd Author's Affiliation | Kyoto University(Kyoto Univ.) |
| 4th Author's Name | Tetsuya Matsuda |
| 4th Author's Affiliation | Kyoto University(Kyoto Univ.) |
| Date | 2022-01-26 |
| Paper # | MI2021-62 |
| Volume (vol) | vol.121 |
| Number (no) | MI-347 |
| Page | pp.pp.70-75(MI), |
| #Pages | 6 |
| Date of Issue | 2022-01-18 (MI) |