Comparative validation of graphical models for learning tumor segmentations from noisy manual annotations

Authors:
Frederik O. Kaster;Bjoern H. Menze;Marc-André Weber;Fred A. Hamprecht
Affiliations:
Heidelberg Collaboratory for Image Processing, University of Heidelberg and German Cancer Research Center, Heidelberg, Germany;CSAIL, Massachusetts Institute of Technology, Cambridge MA and INRIA Sophia-Antipolis Méditerrannée, France;Department of Diagnostic Radiology, University of Heidelberg, Germany;Heidelberg Collaboratory for Image Processing, University of Heidelberg, Germany
Venue:
MCV'10 Proceedings of the 2010 international MICCAI conference on Medical computer vision: recognition techniques and applications in medical imaging
Year:
2010

Citing 8
Cited 0

WinBUGS – A Bayesian modelling framework: Concepts, structure, and extensibility

Statistics and Computing
Expectation Propagation for approximate Bayesian inference

UAI '01 Proceedings of the 17th Conference in Uncertainty in Artificial Intelligence
Variational Message Passing

The Journal of Machine Learning Research
Segmenting Brain Tumors using Alignment-Based Features

ICMLA '05 Proceedings of the Fourth International Conference on Machine Learning and Applications
Operations for learning with graphical models

Journal of Artificial Intelligence Research
Probabilistic Graphical Models: Principles and Techniques - Adaptive Computation and Machine Learning

Probabilistic Graphical Models: Principles and Techniques - Adaptive Computation and Machine Learning
Semi-parametric analysis of multi-rater data

Statistics and Computing
Learning From Crowds

The Journal of Machine Learning Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

Classification-based approaches for segmenting medical images commonly suffer from missing ground truth: often one has to resort to manual labelings by human experts, which may show considerable intra-rater and inter-rater variability. We experimentally evaluate several latent class and latent score models for tumor classification based on manual segmentations of different quality, using approximate variational techniques for inference. For the first time, we also study models that make use of image feature information on this specific task. Additionally, we analyze the outcome of hybrid techniques formed by combining aspects of different models. Benchmarking results on simulated MR images of brain tumors are presented: while simple baseline techniques already gave very competitive performance, significant improvements could be made by explicitly accounting for rater quality. Furthermore, we point out the transfer of these models to the task of fusing manual tumor segmentations derived from different imaging modalities on real-world data.