发明公开
EP1646982A2 METHOD FOR ROBUST SCALE-SPACE ANALYSIS OF 3D LOCAL STRUCTURES IN MEDICAL IMAGES
审中-公开
VERFAHREN ZUR ROBUSTEN MASSSTABSRAUMANALSESE VON LOKALEN 3D-STRUKTUREN IN MEDIZINISCHEN BILDERN
- 专利标题： METHOD FOR ROBUST SCALE-SPACE ANALYSIS OF 3D LOCAL STRUCTURES IN MEDICAL IMAGES
- 专利标题（中）： VERFAHREN ZUR ROBUSTEN MASSSTABSRAUMANALSESE VON LOKALEN 3D-STRUKTUREN IN MEDIZINISCHEN BILDERN
-
申请号： EP04778656.1申请日： 2004-07-19
-
公开(公告)号： EP1646982A2公开(公告)日： 2006-04-19
- 发明人: OKADA, Kazunori , COMANICIU, Dorin , BOGONI, Luca
- 申请人： Siemens Medical Solutions USA, Inc.
- 申请人地址： 51, Valley Stream Parkway Malvern, PA 19355-1406 US
- 专利权人： Siemens Medical Solutions USA, Inc.
- 当前专利权人： Siemens Medical Solutions USA, Inc.
- 当前专利权人地址： 51, Valley Stream Parkway Malvern, PA 19355-1406 US
- 代理机构： Vigars, Christopher Ian
- 优先权： US488603P 20030718; US892646 20040716
- 国际公布： WO2005008592 20050127
- 主分类号： G06T7/00
- IPC分类号： G06T7/00
摘要：
A method for determining the location, shape and orientation of a tumor in a medical image includes finding a plurality of spatial extrema µ of a D-dimensional spatial signal ƒ for a set of bandwidths H by performing (11) mean shift-based gradient-ascent iterations for a set of bandwidths H and then determining (12) a D-dimensional spread and orientation of the signal about each extrema µ by estimating a covariance Σ of the signal ƒ for each extrema µ . The optimal estimate of µ and Σ is determined by performing (13) a Jensen-Shannon divergence on the full set µ of and Σ .
信息查询
