Deep Quantile Regression for Uncertainty Estimation in Unsupervised and Supervised Lesion Detection
Open Access
- 27 April 2022
- journal article
- Published by Machine Learning for Biomedical Imaging in Machine Learning for Biomedical Imaging
- Vol. 1 (IPMI 2021) , 1-23
Abstract
Despite impressive state-of-the-art performance on a wide variety of machine learning tasks in multiple applications, deep learning methods can produce over-confident predictions, particularly with limited training data. Therefore, quantifying uncertainty is particularly important in critical applications such as lesion detection and clinical diagnosis, where a realistic assessment of uncertainty is essential in determining surgical margins, disease status and appropriate treatment. In this work, we propose a novel approach that uses quantile regression for quantifying aleatoric uncertainty in both supervised and unsupervised lesion detection problems. The resulting confidence intervals can be used for lesion detection and segmentation. In the unsupervised settings, we combine quantile regression with the Variational AutoEncoder (VAE). The VAE is trained on lesion-free data, so when presented with an image with a lesion, it tends to reconstruct a lesion-free version of the image. To detect the lesion, we then compare the input (lesion) and output (lesion-free) images. Here we address the problem of quantifying uncertainty in the images that are reconstructed by the VAE as the basis for principled outlier or lesion detection. The VAE models the output as a conditionally independent Gaussian characterized by its mean and variance. Unfortunately, joint optimization of both mean and variance in the VAE leads to the well-known problem of shrinkage or underestimation of variance. Here we describe an alternative Quantile-Regression VAE (QR-VAE) that avoids this variance shrinkage problem by directly estimating conditional quantiles for the input image. Using the estimated quantiles, we compute the conditional mean and variance for input image from which we then detect outliers by thresholding at a false-discovery-rate corrected p-value. In the supervised setting, we develop binary quantile regression (BQR) for the supervised lesion segmentation task. We show how BQR can be used to capture uncertainty in lesion boundaries in a manner that characterizes expert disagreement.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- Quantile Regression for Uncertainty Estimation in VAEs with Applications to Brain Lesion Detection2021
- Unsupervised lesion detection via image restoration with a normative priorMedical Image Analysis, 2020
- Brain Lesion Detection Using A Robust Variational Autoencoder and Transfer Learning2020
- Beyond Expectation: Deep Joint Mean and Quantile Regression for Spatiotemporal ProblemsIEEE Transactions on Neural Networks and Learning Systems, 2020
- A comparison of some conformal quantile regression methodsStat, 2020
- ISLES 2015 - A public evaluation benchmark for ischemic stroke lesion segmentation from multispectral MRIMedical Image Analysis, 2017
- Transforming Research and Clinical Knowledge in Traumatic Brain Injury Pilot: Multicenter Implementation of the Common Data Elements for Traumatic Brain InjuryJournal of Neurotrauma, 2013
- The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI): A Completed Reference Database of Lung Nodules on CT ScansMedical Physics, 2011
- Smoothed binary regression quantilesJournal of Applied Econometrics, 2006
- Multiple significance tests: the Bonferroni methodBMJ, 1995