Predicting Semantic Descriptions from Medical Images with Convolutional Neural Networks
- 23 June 2015
- book chapter
- Published by Springer Nature
- Vol. 24, 437-448
Abstract
Learning representative computational models from medical imaging data requires large training data sets. Often, voxel-level annotation is unfeasible for sufficient amounts of data. An alternative to manual annotation, is to use the enormous amount of knowledge encoded in imaging data and corresponding reports generated during clinical routine. Weakly supervised learning approaches can link volume-level labels to image content but suffer from the typical label distributions in medical imaging data where only a small part consists of clinically relevant abnormal structures. In this paper we propose to use a semantic representation of clinical reports as a learning target that is predicted from imaging data by a convolutional neural network. We demonstrate how we can learn accurate voxel-level classifiers based on weak volume-level semantic descriptions on a set of 157 optical coherence tomography (OCT) volumes. We specifically show how semantic information increases classification accuracy for intraretinal cystoid fluid (IRC), subretinal fluid (SRF) and normal retinal tissue, and how the learning algorithm links semantic concepts to image content and geometry.Keywords
This publication has 12 references indexed in Scilit:
- Unsupervised Pre-training Across Image Domains Improves Lung Tissue Classification2014
- Multi-fold MIL Training for Weakly Supervised Object Localization2014
- Manifold Learning of Brain MRIs by Deep Learning2013
- Multi-column deep neural networks for image classification2012
- Unsupervised learning of hierarchical representations with convolutional deep belief networksCommunications of the ACM, 2011
- Semi-Supervised Random Forests2009
- Automated 3-D Intraretinal Layer Segmentation of Macular Spectral-Domain Optical Coherence Tomography ImagesIEEE Transactions on Medical Imaging, 2009
- Region Classification with Markov Field Aspect Models2007
- Solving the multiple instance problem with axis-parallel rectanglesArtificial Intelligence, 1997
- Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in positionBiological Cybernetics, 1980