Active Learning in Brain Tumor Segmentation with Uncertainty Sampling and Annotation Redundancy Restriction
Journal of Imaging Informatics in Medicine · 2024
DOI: 10.1007/s10278-024-01037-6 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Imaging Informatics in Medicine, 2024. Verfasst von Daniel D Kim und 16 weiteren Autor:innen. Beteiligte Einrichtung: Warren Alpert Medical School of Brown University, Providence, RI, USA. Hauptthemen laut MEDLINE: Brain Neoplasms, Magnetic Resonance Imaging, Bayes Theorem, Deep Learning. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 29 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: National Science Foundation of Hunan Province, International Science and Technology Innovation Joint Base of Machine Vision and Medical Image Processing in Hunan Providence, 111 project.
Abstract (Originalsprache)
Deep learning models have demonstrated great potential in medical imaging but are limited by the expensive, large volume of annotations required. To address this, we compared different active learning strategies by training models on subsets of the most informative images using real-world clinical datasets for brain tumor segmentation and proposing a framework that minimizes the data needed while maintaining performance. Then, 638 multi-institutional brain tumor magnetic resonance imaging scans were used to train three-dimensional U-net models and compare active learning strategies. Uncertainty estimation techniques including Bayesian estimation with dropout, bootstrapping, and margins sampling were compared to random query. Strategies to avoid annotating similar images were also considered. We determined the minimum data necessary to achieve performance equivalent to the model trained on the full dataset (α = 0.05). Bayesian approximation with dropout at training and testing showed results equivalent to that of the full data model (target) with around 30% of the training data needed by random query to achieve target performance (p = 0.018). Annotation redundancy restriction techniques can reduce the training data needed by random query to achieve target performance by 20%. We investigated various active learning strategies to minimize the annotation burden for three-dimensional brain tumor segmentation. Dropout uncertainty estimation achieved target performance with the least annotated data.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Warren Alpert Medical School of Brown University, Providence, RI, USA.
- Department of Diagnostic Imaging, Rhode Island Hospital, Providence, RI, USA.
- Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA.
- Department of Neurology, Second Xiangya Hospital, Central South University, Changsha, China.
- Clinical Medical Research Center for Stroke Prevention and Treatment of Hunan Province, Department of Neurology, Second Xiangya Hospital, Central South University, Changsha, China.
- Department of Radiology, Second Xiangya Hospital, Central South University, Changsha, China.
ORCID-Profile: Jian Peng
Förderung
- National Science Foundation of Hunan Province
- International Science and Technology Innovation Joint Base of Machine Vision and Medical Image Processing in Hunan Providence
- 111 project
- Hunan Province Key Areas Research and Development Program
- Huxiang High-level Talent Gathering Project-Innovation Talent
- Natural Science Foundation of China
- Natural Science Foundation of Hunan Province
- Alpert Medical School Summer Assistantship award
Zitieren
APA
Kim, D. D., Chandra, R. S., Yang, L., Wu, J., Feng, X., Atalay, M., Bettegowda, C., Jones, C., Sair, H., Liao, W. H., Zhu, C., Zou, B., Kazerooni, A. F., Nabavizadeh, A., Jiao, Z., Peng, J., & Bai, H. X. (2024). Active Learning in Brain Tumor Segmentation with Uncertainty Sampling and Annotation Redundancy Restriction. Journal of Imaging Informatics in Medicine. https://doi.org/10.1007/s10278-024-01037-6
Vancouver
Kim DD, Chandra RS, Yang L, Wu J, Feng X, Atalay M, et al. Active Learning in Brain Tumor Segmentation with Uncertainty Sampling and Annotation Redundancy Restriction. Journal of Imaging Informatics in Medicine. 2024. doi:10.1007/s10278-024-01037-6
BibTeX
@article{Kim2024,
author = {Kim, Daniel D and Chandra, Rajat S and Yang, Li and Wu, Jing and Feng, Xue and Atalay, Michael and Bettegowda, Chetan and Jones, Craig and Sair, Haris and Liao, Wei-Hua and Zhu, Chengzhang and Zou, Beiji and Kazerooni, Anahita Fathi and Nabavizadeh, Ali and Jiao, Zhicheng and Peng, Jian and Bai, Harrison X},
title = {Active Learning in Brain Tumor Segmentation with Uncertainty Sampling and Annotation Redundancy Restriction},
journal = {Journal of Imaging Informatics in Medicine},
year = {2024},
doi = {10.1007/s10278-024-01037-6}
}
EUVIMED