Deep feature learning for knee cartilage segmentation using a triplanar convolutional neural network

TY - GEN

T1 - Deep feature learning for knee cartilage segmentation using a triplanar convolutional neural network

AU - Prasoon, Adhish

AU - Petersen, Peter Kersten

AU - Igel, Christian

AU - Lauze, Francois Bernard

AU - Dam, Erik Bjørnager

AU - Nielsen, Mads

N1 - Conference code: 16

PY - 2013

Y1 - 2013

N2 - Segmentation of anatomical structures in medical images is often based on a voxel/pixel classification approach. Deep learning systems, such as convolutional neural networks (CNNs), can infer a hierarchical representation of images that fosters categorization. We propose a novel system for voxel classification integrating three 2D CNNs, which have a one-to-one association with the xy, yz and zx planes of 3D image, respectively. We applied our method to the segmentation of tibial cartilage in low field knee MRI scans and tested it on 114 unseen scans. Although our method uses only 2D features at a single scale, it performs better than a state-of-the-art method using 3D multi-scale features. In the latter approach, the features and the classifier have been carefully adapted to the problem at hand. That we were able to get better results by a deep learning architecture that autonomously learns the features from the images is the main insight of this study.

AB - Segmentation of anatomical structures in medical images is often based on a voxel/pixel classification approach. Deep learning systems, such as convolutional neural networks (CNNs), can infer a hierarchical representation of images that fosters categorization. We propose a novel system for voxel classification integrating three 2D CNNs, which have a one-to-one association with the xy, yz and zx planes of 3D image, respectively. We applied our method to the segmentation of tibial cartilage in low field knee MRI scans and tested it on 114 unseen scans. Although our method uses only 2D features at a single scale, it performs better than a state-of-the-art method using 3D multi-scale features. In the latter approach, the features and the classifier have been carefully adapted to the problem at hand. That we were able to get better results by a deep learning architecture that autonomously learns the features from the images is the main insight of this study.

U2 - 10.1007/978-3-642-40763-5_31

DO - 10.1007/978-3-642-40763-5_31

M3 - Article in proceedings

SN - 978-3-642-40762-8

T3 - Lecture notes in computer science

SP - 246

EP - 253

BT - Medical Image Computing and Computer-Assisted Intervention – MICCAI 2013

A2 - Mori, Kensaku

A2 - Sakuma, Ichiro

A2 - Sato, Yoshinobu

A2 - Barillot, Christian

A2 - Navab, Nassir

PB - Springer

T2 - International Conference - MICCAI 2013

Y2 - 22 September 2013 through 26 September 2013

ER -