Discovery - University of Dundee - Online Publications

Library & Learning Centre

Gland segmentation in colon histology images using hand-crafted features and convolutional neural networks

Gland segmentation in colon histology images using hand-crafted features and convolutional neural networks

Research output: Chapter in Book/Report/Conference proceedingConference contribution

View graph of relations

Authors

Research units

Info

Original languageEnglish
Title of host publication2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI)
Subtitle of host publicationFrom Nano to Macro, ISBI 2016 - Proceedings
PublisherIEEE
Pages1405-1408
Number of pages4
VolumeJune-2016
ISBN (Electronic)9781479923496
ISBN (Print)9781479923502
DOIs
StatePublished - 16 Jun 2016
Event2016 IEEE 13th International Symposium on Biomedical Imaging - Prague, Czech Republic

Conference

Conference2016 IEEE 13th International Symposium on Biomedical Imaging
Abbreviated titleISBI 2016
CountryCzech Republic
CityPrague
Period13/04/1616/04/16
Internet address

Abstract

We investigate glandular structure segmentation in colon histology images as a window-based classification problem. We compare and combine methods based on fine-tuned convolutional neural networks (CNN) and hand-crafted features with support vector machines (HC-SVM). On 85 images of H&E-stained tissue, we find that fine-tuned CNN outperforms HC-SVM in gland segmentation measured by pixel-wise Jaccard and Dice indices. For HC-SVM we further observe that training a second-level window classifier on the posterior probabilities – as an output refinement – can substantially improve the segmentation performance. The final performance of HC-SVM with refinement is comparable to that of CNN. Furthermore, we show that by combining and refining the posterior probability out-puts of CNN and HC-SVM together, a further performance boost is obtained.

Download statistics

No data available

Documents

Open Access permissions

Open

Documents

DOI

Library & Learning Centre

Contact | Accessibility | Policy