Title | ||
---|---|---|
Improving Medical Images Classification With Label Noise Using Dual-Uncertainty Estimation |
Abstract | ||
---|---|---|
Deep neural networks are known to be data-driven and label noise can have a marked impact on model performance. Recent studies have shown great robustness to classic image recognition even under a high noisy rate. In medical applications, learning from datasets with label noise is more challenging since medical imaging datasets tend to have instance-dependent noise (IDN) and suffer from high observer variability. In this paper, we systematically discuss the two common types of label noise in medical images - disagreement label noise from inconsistency expert opinions and single-target label noise from biased aggregation of individual annotations. We then propose an uncertainty estimation-based framework to handle these two label noise amid the medical image classification task. We design a dual-uncertainty estimation approach to measure the
<bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">disagreement label noise</b>
and
<bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">single-target label noise</b>
via improved Direct Uncertainty Prediction and Monte-Carlo-Dropout. A boosting-based curriculum training procedure is later introduced for robust learning. We demonstrate the effectiveness of our method by conducting extensive experiments on three different diseases with synthesized and real-world label noise: skin lesions, prostate cancer, and retinal diseases. We also release a large re-engineered database that consists of annotations from more than ten ophthalmologists with an unbiased golden standard dataset for evaluation and benchmarking. The dataset is available at
<uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://mmai.group/peoples/julie/</uri>
. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1109/TMI.2022.3141425 | IEEE Transactions on Medical Imaging |
Keywords | DocType | Volume |
Diagnostic Imaging,Neural Networks, Computer,Noise,Radiography,Uncertainty | Journal | 41 |
Issue | ISSN | Citations |
6 | 0278-0062 | 0 |
PageRank | References | Authors |
0.34 | 11 | 8 |
Name | Order | Citations | PageRank |
---|---|---|---|
Lie Ju | 1 | 1 | 0.69 |
Xin Wang | 2 | 0 | 1.35 |
Lin Wang | 3 | 2 | 2.41 |
Dwarikanath Mahapatra | 4 | 312 | 33.71 |
Xin Zhao | 5 | 2 | 2.07 |
Quan Zhou | 6 | 0 | 0.34 |
Tongliang Liu | 7 | 902 | 47.13 |
zongyuan ge | 8 | 149 | 27.83 |