Imperfect Segmentation Labels: How Much Do They Matter?
Published in
(2018)External Links:
TL
TL;DR
This large-scale study of model performance in the presence of varying types and degrees of error in training data shows that for each architecture, performance steadily declines with boundary-localized errors, however, U-Net was significantly more robust to jagged boundary errors than the other architectures.
Abstract
Labeled datasets for semantic segmentation are imperfect, especially in medical imaging where borders are often subtle or ill-defined. Little work has been done to analyze the effect that label errors have on the performance of segmentation methodologies. Here we present a large-scale study of model performance in the presence of varying types and degrees of error in training data. We trained U-Net, SegNet, and FCN32 several times for liver segmentation with 10 different modes of ground-truth perturbation. Our results show that for each architecture, performance steadily declines with boundary-localized errors, however, U-Net was significantly more robust to jagged boundary errors than the other architectures. We also found that each architecture was very robust to non-boundary-localized errors, suggesting that boundary-localized errors are fundamentally different and more challenging problem than random label errors in a classification setting.
References26 items
1
The potential for machine learning algorithms to improve and reduce the cost of 3-dimensional printing for surgical planning
2
An application of cascaded 3D fully convolutional networks for medical image segmentation
3
A Web-Based Platform for Distributed Annotation of Computerized Tomography Scans
4
Liver Tumor Segmentation from MR Images Using 3D Fast Marching Algorithm and Single Hidden Layer Feedforward Neural Network
5