UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Disentangling human error from the ground truth in segmentation of medical images

Jacob, J; (2021) Disentangling human error from the ground truth in segmentation of medical images. In: Advances in Neural Information Processing Systems 33 (NeurIPS 2020). NeurIPS: Vancouver, Canada. Green open access

[thumbnail of nips_2020_final_version.pdf]
Preview
Text
nips_2020_final_version.pdf - Accepted version

Download (8MB) | Preview

Abstract

Recent years have seen increasing use of supervised learning methods for segmentation tasks. However, the predictive performance of these algorithms depends on the quality of labels. This problem is particularly pertinent in the medical image domain, where both the annotation cost and inter-observer variability are high. In a typical label acquisition process, different human experts provide their estimates of the true'' segmentation labels under the influence of their own biases and competence levels. Treating these noisy labels blindly as the ground truth limits the performance that automatic segmentation algorithms can achieve. In this work, we present a method for jointly learning, from purely noisy observations alone, the reliability of individual annotators and the true segmentation label distributions, using two coupled CNNs. The separation of the two is achieved by encouraging the estimated annotators to be maximally unreliable while achieving high fidelity with the noisy training data. We first define a toy segmentation dataset based on MNIST and study the properties of the proposed algorithm. We then demonstrate the utility of the method on three public medical imaging segmentation datasets with simulated (when necessary) and real diverse annotations: 1) MSLSC (multiple-sclerosis lesions); 2) BraTS (brain tumours); 3) LIDC-IDRI (lung abnormalities). In all cases, our method outperforms competing methods and relevant baselines particularly in cases where the number of annotations is small and the amount of disagreement is large. The experiments also show strong ability to capture the complex spatial characteristics of annotators' mistakes. Our code is available at \url{https://github.com/moucheng2017/LearnNoisyLabelsMedicalImages}.

Type: Proceedings paper
Title: Disentangling human error from the ground truth in segmentation of medical images
Event: Conference on Neural Information Processing Systems
Location: Vancouver, Canada
Open access status: An open access version is available from UCL Discovery
Publisher version: https://proceedings.neurips.cc/paper/2020/hash/b5d...
Language: English
Additional information: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.
UCL classification: UCL
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Medical Sciences
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Medical Sciences > Div of Medicine
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Medical Sciences > Div of Medicine > Respiratory Medicine
URI: https://discovery.ucl.ac.uk/id/eprint/10131612
Downloads since deposit
4Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item