Robust Image Segmentation Quality Assessment

Deep learning based image segmentation methods have achieved great success, even having human-level accuracy in some applications. However, due to the black box nature of deep learning, the best method may fail in some situations. Thus predicting segmentation quality without ground truth would be very crucial especially in clinical practice. Recently, people proposed to train neural networks to estimate the quality score by regression. Although it can achieve promising prediction accuracy, the network suffers robustness problem, e.g. it is vulnerable to adversarial attacks. In this paper, we propose to alleviate this problem by utilizing the difference between the input image and the reconstructed image, which is conditioned on the segmentation to be assessed, to lower the chance to overfit to the undesired image features from the original input image, and thus to increase the robustness. Results on ACDC17 dataset demonstrated our method is promising.

[1]  Ben Glocker,et al.  Subject-level Prediction of Segmentation Failure using Real-Time Convolutional Neural Nets , 2018 .

[2]  Sotirios A. Tsaftaris,et al.  Factorised spatial representation learning: application in semi-supervised myocardial segmentation , 2018, MICCAI.

[3]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[4]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[5]  Joan Bruna,et al.  Intriguing properties of neural networks , 2013, ICLR.

[6]  Nassir Navab,et al.  Deep Autoencoding Models for Unsupervised Anomaly Segmentation in Brain MR Images , 2018, BrainLes@MICCAI.

[7]  Timo Kohlberger,et al.  Evaluating Segmentation Error without Ground Truth , 2012, MICCAI.

[8]  Georg Langs,et al.  Identifying and Categorizing Anomalies in Retinal Imaging Data , 2016, ArXiv.

[9]  Georg Langs,et al.  Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery , 2017, IPMI.

[10]  Qiang Yang,et al.  Cross Validation Framework to Choose amongst Models and Datasets for Transfer Learning , 2010, ECML/PKDD.

[11]  Ben Glocker,et al.  Automatic Quality Control of Cardiac MRI Segmentation in Large-Scale Population Imaging , 2017, MICCAI.

[12]  Konstantinos Kamnitsas,et al.  Unsupervised Lesion Detection in Brain CT using Bayesian Convolutional Autoencoders , 2018 .

[13]  Simon Osindero,et al.  Conditional Generative Adversarial Nets , 2014, ArXiv.

[14]  Ben Glocker,et al.  Real-time Prediction of Segmentation Quality , 2018, MICCAI.

[15]  Rama Chellappa,et al.  Visual Domain Adaptation: A survey of recent advances , 2015, IEEE Signal Processing Magazine.

[16]  Samy Bengio,et al.  Adversarial examples in the physical world , 2016, ICLR.

[17]  Ender Konukoglu,et al.  Unsupervised Detection of Lesions in Brain MRI using constrained adversarial auto-encoders , 2018, ArXiv.

[18]  Carole Lartizien,et al.  Regularized siamese neural network for unsupervised outlier detection on brain multiparametric magnetic resonance imaging: Application to epilepsy lesion screening , 2019, Medical Image Anal..

[19]  Konstantinos Kamnitsas,et al.  Reverse Classification Accuracy: Predicting Segmentation Performance in the Absence of Ground Truth , 2017, IEEE Transactions on Medical Imaging.