Sample selection with noise rate estimation in noise learning of medical image analysis
This work addresses the challenge of unreliable deep learning models due to label noise in medical image analysis, offering an incremental improvement in noise-robust learning techniques.
The paper tackles the problem of label noise in medical image datasets by introducing a sample selection method that estimates noise rates using Linear Regression on loss distributions and excludes potentially noisy samples, achieving superior performance over existing methods on five benchmark datasets and a real-life noisy medical image dataset, especially at high noise rates.
In the field of medical image analysis, deep learning models have demonstrated remarkable success in enhancing diagnostic accuracy and efficiency. However, the reliability of these models is heavily dependent on the quality of training data, and the existence of label noise (errors in dataset annotations) of medical image data presents a significant challenge. This paper introduces a new sample selection method that enhances the performance of neural networks when trained on noisy datasets. Our approach features estimating the noise rate of a dataset by analyzing the distribution of loss values using Linear Regression. Samples are then ranked according to their loss values, and potentially noisy samples are excluded from the dataset. Additionally, we employ sparse regularization to further enhance the noise robustness of our model. Our proposed method is evaluated on five benchmark datasets and a real-life noisy medical image dataset. Notably, two of these datasets contain 3D medical images. The results of our experiments show that our method outperforms existing noise-robust learning methods, particularly in scenarios with high noise rates. Key words: noise-robust learning, medical image analysis, noise rate estimation, sample selection, sparse regularization