컨텐츠 시작
학술대회/행사
초록검색
| 제출번호(No.) | 0257 |
|---|---|
| 분류(Section) | Special Session |
| 분과(Session) | Algebraic geometry and computer vision (SS-09) |
| 영문제목 (Title(Eng.)) |
A mobile application that reconstructs a 3D bony shape from un-calibrated radiographs |
| 저자(Author(s)) |
Kibeom Youn1, Moon Seok Park1, Jehee Lee1 Seoul National University1 |
| 초록본문(Abstract) | Computed tomography (CT)provides benefits in accurate diagnosis of bony deformities. However, the potential adverse effect of radiation exposure has become a concern, especially in a pediatric population as children are more susceptible to the effects of radiation. To reduce radiation dose while maintaining the accuracy of diagnosis, some systems have been proposed to reconstruct 3D bony shapes from calibrated bi-planar X-ray images. However, there are some problems to apply those systems to the actual medical environment. We propose a mobile application that reconstructs the 3D shape from uncalibrated radiographs. The user was required to take photos of AP and LAT radiographs with a iPad and segment the silhouette of the femur. If the user labels a part of the bone through the touch interface, the remaining bone parts are automatically extracted by the graph-cut algorithm. Each radiographic image is automatically aligned using a iterative PNP algorithm to match the silhouette of the statistical shape model, and then the shape of the femur is optimized. |
| 분류기호 (MSC number(s)) |
65D18 |
| 키워드(Keyword(s)) | 3D reconstruction, camera calibration, computed tomography, statistical shape model |
| 강연 형태 (Language of Session (Talk)) |
English |