2013 Fiscal Year Final Research Report
Mathematical Foundations of Computational Anatomy
Project Area | Computational anatomy for computer-aided diagnosis and therapy :Frontiers of medical image sciences |
Project/Area Number |
21103002
|
Research Category |
Grant-in-Aid for Scientific Research on Innovative Areas (Research in a proposed research area)
|
Allocation Type | Single-year Grants |
Review Section |
Science and Engineering
|
Research Institution | The University of Tokyo |
Principal Investigator |
|
Co-Investigator(Kenkyū-buntansha) |
HONTANI Hidekata 名古屋工業大学, 大学院工学系研究科, 教授 (60282688)
IMIYA Atsushi 千葉大学, 総合メディア基盤センター, 教授 (10176505)
MATSUZOE Hiroshi 名古屋工業大学, 大学院工学系研究科, 准教授 (90315177)
HAYASHI Naoto 東京大学, 医学部附属病院, 特任准教授 (10261992)
|
Co-Investigator(Renkei-kenkyūsha) |
OHTOMO Kuni 東京大学, 医学部附属病院, 教授 (80126010)
|
Project Period (FY) |
2009-07-23 – 2014-03-31
|
Keywords | 計算解剖学 / 医用画像 / 解剖学的ランドマーク / 統計的形状モデル / 形状平均 / 組み合わせ最適化 / 変分原理 / 変形型指数分布族 |
Research Abstract |
For the goal of the project; robust computational understanding of clinical images, our researches covered mathematical methods for representation of the "Computational Anatomy Model" which statistically describes individual variations of the organ shapes, and clinical image understanding methods by using the model. The model was constructed as a set of structured points, that is, point distribution model with anatomical features. By using the clinical CT images in our database, various researches from mathematical foundation to clinical application have been performed. The studies covered (1) anatomical landmark detection initiating clinical image understanding, (2) robust image segmentation by a point distribution model with local appearance and with sparse representation, (3) spatio-temporal shape average definition by variational method, and (4) introducing deformed exponential families for treating statistical outliers such as anatomical abnormalities.
|
Research Products
(31 results)