A fully automated tortuosity quantification system with application to corneal nerve fibres in confocal microscopy images
详细信息 查看全文
- 作者：Roberto Annunziata ; a ; r.annunziata@dundee.ac.uk" class="auth_mail" title="E-mail the corresponding author ; Ahmad Kheirkhahc ; Shruti Aggarwalc ; Pedram Hamrahb ; c ; Emanuele Truccoa
- 关键词：Tortuosity ; Cornea ; Multiscale ; Segmentation ; Curvature ; Automated
- 刊名：Medical Image Analysis
- 出版年：2016
- 出版时间：August 2016
- 年：2016
- 卷：32
- 期：Complete
- 页码：216-232
- 全文大小：4725 K
文摘
- •
Data: 140 normal and pathological corneal images covering 4 tortuosity levels.
- •
Segmentation: combining a ridge detector with learned multi-range context filters.
- •
Multi-range context filters improve single-range context filters.
- •
Supervised, multi-scale tortuosity from highly discriminative measures.
- •
Rotation-invariant, spline-based fitting and analytic curvature estimation.