Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Evaluation of Data Augmentation of MR Images for Deep Learning

Andersson, Erik and Berglund, Robin (2018) FMS820 20181
Mathematical Statistics
Abstract
Data augmentation is a process to create new artificial data by altering the available data set. It has by
many been considered a key factor for increasing robustness and performance of image classification
tasks using deep learning methods, such as the convolutional neural network. This report aims to
investigate the effect of data augmentation for the segmentation of magnetic resonance (MR) images
using neural networks. Geometric data augmentation techniques such as rotation, scaling, translation
and elastic deformation are evaluated for different data set sizes. Also, motion artifacts are simulated
in MR images as an augmentation to increase segmentation performance on images containing real
artifacts. A method for simulating... (More)
Data augmentation is a process to create new artificial data by altering the available data set. It has by
many been considered a key factor for increasing robustness and performance of image classification
tasks using deep learning methods, such as the convolutional neural network. This report aims to
investigate the effect of data augmentation for the segmentation of magnetic resonance (MR) images
using neural networks. Geometric data augmentation techniques such as rotation, scaling, translation
and elastic deformation are evaluated for different data set sizes. Also, motion artifacts are simulated
in MR images as an augmentation to increase segmentation performance on images containing real
artifacts. A method for simulating these artificial motion artifacts is developed, where the k-space of
the image is altered to simulate patient movement due to breathing.
Results showed that geometric data augmentation improved performance of the network for smaller
data sets, but lost importance as the amount of available training data grew bigger. Segmentation
performance of images containing motion artifacts significantly increased if the neural network had
been trained on a training set augmented with simulated artifacts. (Less)
Please use this url to cite or link to this publication:
author
Andersson, Erik and Berglund, Robin
supervisor
organization
course
FMS820 20181
year
type
H2 - Master's Degree (Two Years)
subject
language
English
id
8952747
date added to LUP
2018-06-25 14:21:55
date last changed
2018-06-25 14:21:55
@misc{8952747,
  abstract     = {{Data augmentation is a process to create new artificial data by altering the available data set. It has by
many been considered a key factor for increasing robustness and performance of image classification
tasks using deep learning methods, such as the convolutional neural network. This report aims to
investigate the effect of data augmentation for the segmentation of magnetic resonance (MR) images
using neural networks. Geometric data augmentation techniques such as rotation, scaling, translation
and elastic deformation are evaluated for different data set sizes. Also, motion artifacts are simulated
in MR images as an augmentation to increase segmentation performance on images containing real
artifacts. A method for simulating these artificial motion artifacts is developed, where the k-space of
the image is altered to simulate patient movement due to breathing.
Results showed that geometric data augmentation improved performance of the network for smaller
data sets, but lost importance as the amount of available training data grew bigger. Segmentation
performance of images containing motion artifacts significantly increased if the neural network had
been trained on a training set augmented with simulated artifacts.}},
  author       = {{Andersson, Erik and Berglund, Robin}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Evaluation of Data Augmentation of MR Images for Deep Learning}},
  year         = {{2018}},
}