Please use this identifier to cite or link to this item:
|Title:||Metric- and rank-based similarity learning in medical image computing||Authors:||Huang, Wei||Keywords:||DRNTU::Engineering::Electrical and electronic engineering::Electronic systems||Issue Date:||2011||Abstract:||While a great number of medical images are still being examined and analysed visually and qualitatively by clinicians in their clinical diagnosis nowadays, the evergrowing amount of medical images accompanying routine diagnosis and treatment does increase the demand and acceptance of incorporating medical image computing techniques for more objective and quantitative clinical analyses to assist clinicians' decision making. Although studies on medical image computing in diverse medical applications are becoming ever more intensive, the number of fully automated and robust medical image computing methods, which can be conveniently deployed by clinicians in their routine diagnosis, is limited. It is due to the fact that, a large number of contemporary methods still rely heavily on a number of parameters that are usually pre-defined experimentally by system developers. These somewhat "ad-hoc" methods need to be changed to cope with diverse clinical needs more easily and conveniently.||Description:||142 p.||URI:||http://hdl.handle.net/10356/58107||Rights:||Nanyang Technological University||Fulltext Permission:||restricted||Fulltext Availability:||With Fulltext|
|Appears in Collections:||EEE Theses|
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.