Log In Sign Up

Improved Trainable Calibration Method for Neural Networks on Medical Imaging Classification

by   Gongbo Liang, et al.

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.


page 1

page 2

page 3

page 4


Diagnostic Uncertainty Calibration: Towards Reliable Machine Predictions in Medical Domain

Label disagreement between human experts is a common issue in the medica...

Domain Generalization for Medical Imaging Classification with Linear-Dependency Regularization

Recently, we have witnessed great progress in the field of medical imagi...

Bayesian Conditional GAN for MRI Brain Image Synthesis

As a powerful technique in medical imaging, image synthesis is widely us...

LMFLOSS: A Hybrid Loss For Imbalanced Medical Image Classification

Automatic medical image classification is a very important field where t...

A Comparative Study of Confidence Calibration in Deep Learning: From Computer Vision to Medical Imaging

Although deep learning prediction models have been successful in the dis...

DBCal: Density Based Calibration of classifier predictions for uncertainty quantification

Measurement of uncertainty of predictions from machine learning methods ...