Morphological classification of odontogenic keratocysts using Bouligand-Minkowski fractal descriptors

详细信息    查看全文

• 作者：Joao B. Florindo^a ; ^b ; ^{jbflorindo@ime.unicamp.br} ; ; Odemir M. Bruno^a ; ^{bruno@ifsc.usp.br} ; ; Gabriel Landini^c ; ^{G.Landini@bham.ac.uk}
• 关键词：Computer-aided detection and diagnosis ; Microscopy ; Pattern recognition and classification ; Odontogenic cyst
• 刊名：Computers in Biology and Medicine
• 出版年：2017
• 出版时间：1 February 2017
• 年：2017
• 卷：81
• 期：Complete
• 页码：1-10
• 全文大小：935 K
• 卷排序：81

文摘

Bouligand–Minkowski descriptors are proposed to identify three groups of oral cysts. Images from the stained epithelial lining of these lesions are used for the analysis. The case-wise classification accuracy of radicular cysts and keratocysts was 100%. For sporadic and syndromic keratocysts the classification accuracy was 71%.