Learning Robust Classifier for Imbalanced Medical Image Dataset with Noisy Labels by Minimizing Invariant Risk
Lists
Only you see this
Abstract
In medical image analysis, imbalanced noisy dataset classification poses a long-standing and critical problem since clinical large-scale datasets often attain noisy labels and imbalanced distributions through annotation and collection. Current approaches addressing noisy labels and long-tailed distributions separately may negatively impact real-world practices. Additionally, the factor of class hardness hindering label noise removal remains undiscovered, causing a critical necessity for an approach to enhance the classification performance of noisy imbalanced medical datasets with various class hardness. To address this paradox, we propose a robust classifier that trains on a multi-stage noise removal framework, which jointly rectifies the adverse effects of label noise, imbalanced distribution, and class hardness. The proposed noise removal framework consists of multiple phases. Multi-Environment Risk Minimization (MER) strategy captures data-to-label causal features for noise identification, and the Rescaling Class-aware Gaussian Mixture Modeling (RCGM) learns class-invariant detection mappings for noise removal. Extensive experiments on two imbalanced noisy clinical datasets demonstrate the capability and potential of our method for boosting the performance of medical image classification.
BibTeX
@inproceedings{li2023learning,
author = {Jinpeng Li and Hanqun Cao and Jiaze Wang and Furui Liu and Qi Dou and Guangyong Chen and Pheng-Ann Heng},
title = {{Learning Robust Classifier for Imbalanced Medical Image Dataset with Noisy Labels by Minimizing Invariant Risk}},
booktitle = {International Conference on Medical Image Computing and Computer-Assisted Intervention},
year = {2023},
url = {https://openreview.net/forum?id=6n97824kSH},
}Loading related content...