Share Email Print
cover

Proceedings Paper

A method for medulloblastoma tumor differentiation based on convolutional neural networks and transfer learning
Author(s): Angel Cruz-Roa; John Arévalo; Alexander Judkins; Anant Madabhushi; Fabio González
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Convolutional neural networks (CNN) have been very successful at addressing different computer vision tasks thanks to their ability to learn image representations directly from large amounts of labeled data. Features learned from a dataset can be used to represent images from a different dataset via an approach called transfer learning. In this paper we apply transfer learning to the challenging task of medulloblastoma tumor differentiation. We compare two different CNN models which were previously trained in two different domains (natural and histopathology images). The first CNN is a state-of-the-art approach in computer vision, a large and deep CNN with 16-layers, Visual Geometry Group (VGG) CNN. The second (IBCa-CNN) is a 2-layer CNN trained for invasive breast cancer tumor classification. Both CNNs are used as visual feature extractors of histopathology image regions of anaplastic and non-anaplastic medulloblastoma tumor from digitized whole-slide images. The features from the two models are used, separately, to train a softmax classifier to discriminate between anaplastic and non-anaplastic medulloblastoma image regions. Experimental results show that the transfer learning approach produce competitive results in comparison with the state of the art approaches for IBCa detection. Results also show that features extracted from the IBCa-CNN have better performance in comparison with features extracted from the VGG-CNN. The former obtains 89.8% while the latter obtains 76.6% in terms of average accuracy.

Paper Details

Date Published: 22 December 2015
PDF: 8 pages
Proc. SPIE 9681, 11th International Symposium on Medical Information Processing and Analysis, 968103 (22 December 2015); doi: 10.1117/12.2208825
Show Author Affiliations
Angel Cruz-Roa, Univ. de los Llanos (Colombia)
Univ. Nacional de Colombia (Colombia)
John Arévalo, Univ. Nacional de Colombia (Colombia)
Alexander Judkins, Children's Hospital Los Angeles (United States)
Anant Madabhushi, Case Western Reserve Univ. (United States)
Fabio González, Univ. Nacional de Colombia (Colombia)


Published in SPIE Proceedings Vol. 9681:
11th International Symposium on Medical Information Processing and Analysis
Eduardo Romero; Natasha Lepore; Juan D. García-Arteaga; Jorge Brieva, Editor(s)

© SPIE. Terms of Use
Back to Top