Improved Trainable Calibration Method for Neural Networks on Medical Imaging Classification
arXiv:2009.04057
Abstract
Recent works have shown that deep neural networks can achieve super-human performance in a wide range of image classification tasks in the medical imaging domain. However, these works have primarily focused on classification accuracy, ignoring the important role of uncertainty quantification. Empirically, neural networks are often miscalibrated and overconfident in their predictions. This miscalibration could be problematic in any automatic decision-making system, but we focus on the medical field in which neural network miscalibration has the potential to lead to significant treatment errors. We propose a novel calibration approach that maintains the overall classification accuracy while significantly improving model calibration. The proposed approach is based on expected calibration error, which is a common metric for quantifying miscalibration. Our approach can be easily integrated into any classification task as an auxiliary loss term, thus not requiring an explicit training round for calibration. We show that our approach reduces calibration error significantly across various architectures and datasets.
arXiv commentAccepted to the 31th British Machine Vision Conference (BMVC 2020)
References in corpus (5)
- Distilling the Knowledge in a Neural Network
- Empirical Evaluation of Rectified Activations in Convolutional Network
- Low Dose CT Image Denoising Using a Generative Adversarial Network with Wasserstein Distance and Perceptual Loss
- Regularizing Neural Networks by Penalizing Confident Output Distributions
- Calibration tests in multi-class classification: A unifying framework
Cited by in corpus (4)
- MCUa: Multi-level Context and Uncertainty aware Dynamic Deep Ensemble for Breast Cancer Histology Image Classification
- Deep Probability Estimation
- Multi-loss ensemble deep learning for chest X-ray classification
- Optical Wavelength Guided Self-Supervised Feature Learning For Galaxy Cluster Richness Estimate