Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation
NIHR Imperial Biomedical Research Centre · Institute of Group Analysis · +2 more institutions
Abstract
We propose a dual pathway, 11-layers deep, three-dimensional Convolutional Neural Network for the challenging task of brain lesion segmentation. The devised architecture is the result of an in-depth analysis of the limitations of current networks proposed for similar applications. To overcome the computational burden of processing 3D medical scans, we have devised an efficient and effective dense training scheme which joins the processing of adjacent image patches into one pass through the network while automatically adapting to the inherent class imbalance present in the data. Further, we analyze the development of deeper, thus more discriminative 3D CNNs. In order to incorporate both local and larger…
Citation impact
- FWCI
- 162.18
- Percentile
- 100%
- References
- 119
Authors
8- KKKonstantinos KamnitsasCorresponding
NIHR Imperial Biomedical Research Centre, Institute of Group Analysis, Imperial College London
- CLChristian Ledig
Institute of Group Analysis, Imperial College London
- VNVirginia Newcombe
University of Cambridge
- JSJoanna Simpson
University of Cambridge
- ADAndrew D. Kane
University of Cambridge
Topics & keywords
- Computer science
- Conditional random field
- Discriminative model
- Artificial intelligence
- Segmentation
- Pipeline (software)
- Convolutional neural network
- Pattern recognition (psychology)
- Reduced inequalities