| Presentation | 2007-01-26 Verification of Effectiveness of Stabilization Terms in Multi-Level Statistical Shape Model Toshiyuki OKADA, Ryuji SHIMADA, Masahiko NAKAMOTO, Yoshinobu SATO, Nobuhiko SUGANO, Masatoshi HORI, Hideki YOSHIKAWA, Hironobu NAKAMURA, Shinichi TAMURA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Accuracy of shape approximation using statistical shape models is essentially limited. To overcome the limit, multi-level statistical shape model (ML-SSM) is proposed. One problem of ML-SSM is, however, inconsistency among patches. In this paper, we propose a new method to avoid intra-level inconsistency in ML-SSM. We assume that the patches overlap around boundaries at each intra-level. To maintain the intra-level consistency, we introduce the following constraints. (1) The overlapping region between patches adhere each other. (2) The each patch shape is not far from its mean shape. In our experiments for the proximal femur and the liver. ML-SSM can approximate the unknown shape better by combining appropriate consistency terms into the cost function. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Hierarchical statistical shape model / Proximal femur / Liver / Principal component analysis / Segmentation |
| Paper # | MI2006-91 |
| Date of Issue |
| Conference Information | |
| Committee | MI |
|---|---|
| Conference Date | 2007/1/19(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Medical Imaging (MI) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Verification of Effectiveness of Stabilization Terms in Multi-Level Statistical Shape Model |
| Sub Title (in English) | |
| Keyword(1) | Hierarchical statistical shape model |
| Keyword(2) | Proximal femur |
| Keyword(3) | Liver |
| Keyword(4) | Principal component analysis |
| Keyword(5) | Segmentation |
| 1st Author's Name | Toshiyuki OKADA |
| 1st Author's Affiliation | Department of Computer Science, Osaka University Graduate School of Information Science and Technology:Division of Image Analysis, Osaka University Graduate School of Medicine() |
| 2nd Author's Name | Ryuji SHIMADA |
| 2nd Author's Affiliation | Department of Information and Computer Sciences, Osaka University School of Engineering Science |
| 3rd Author's Name | Masahiko NAKAMOTO |
| 3rd Author's Affiliation | Division of Image Analysis, Osaka University Graduate School of Medicine:Department of Computer Science, Osaka University Graduate School of Information Science and Technology |
| 4th Author's Name | Yoshinobu SATO |
| 4th Author's Affiliation | Division of Image Analysis, Osaka University Graduate School of Medicine:Department of Computer Science, Osaka University Graduate School of Information Science and Technology |
| 5th Author's Name | Nobuhiko SUGANO |
| 5th Author's Affiliation | Department of Orthopaedic Surgery, Osaka University Graduate School of Medicine |
| 6th Author's Name | Masatoshi HORI |
| 6th Author's Affiliation | Department of Radiology, Osaka University Graduate School of Medicine |
| 7th Author's Name | Hideki YOSHIKAWA |
| 7th Author's Affiliation | Department of Orthopaedic Surgery, Osaka University Graduate School of Medicine |
| 8th Author's Name | Hironobu NAKAMURA |
| 8th Author's Affiliation | Department of Radiology, Osaka University Graduate School of Medicine |
| 9th Author's Name | Shinichi TAMURA |
| 9th Author's Affiliation | Division of Image Analysis, Osaka University Graduate School of Medicine:Department of Computer Science, Osaka University Graduate School of Information Science and Technology |
| Date | 2007-01-26 |
| Paper # | MI2006-91 |
| Volume (vol) | vol.106 |
| Number (no) | 509 |
| Page | pp.pp.- |
| #Pages | 4 |
| Date of Issue | |