Unsupervised Lesion Detection with Locally Gaussian Approximation

Generative models have recently been applied to unsupervised lesion detection, where a distribution of normal data, i.e. the normative distribution, is learned and lesions are detected as out-of-distribu-tion regions. However, directly calculating the probability for the lesion region using the normative distribution is intractable. In this work, we address this issue by approximating the normative distribution with local Gaussian approximation and evaluating the probability of test samples in an iterative manner. We show that the local Gaussian approximator can be applied to several auto-encoding models to perform image restoration and unsupervised lesion detection. The proposed method is evaluated on the BraTS Challenge dataset, where the proposed method shows improved detection and achieves state-of-the-art results.

[1]  Alan C. Evans,et al.  A fully automatic and robust brain MRI tissue classification method , 2003, Medical Image Anal..

[2]  Koenraad Van Leemput,et al.  Automated segmentation of multiple sclerosis lesions by model outlier detection , 2001, IEEE Transactions on Medical Imaging.

[3]  Cam-CAN Group,et al.  The Cambridge Centre for Ageing and Neuroscience (Cam-CAN) data repository: Structural and functional MRI, MEG, and cognitive data from a cross-sectional adult lifespan sample , 2017, NeuroImage.

[4]  Ben Glocker,et al.  Reconstructing subject-specific effect maps , 2017, NeuroImage.

[5]  Konstantinos Kamnitsas,et al.  Efficient multi‐scale 3D CNN with fully connected CRF for accurate brain lesion segmentation , 2016, Medical Image Anal..

[6]  Ben Glocker,et al.  Decision Forests for Tissue-Specific Segmentation of High-Grade Gliomas in Multi-channel MR , 2012, MICCAI.

[7]  Max Welling,et al.  Auto-Encoding Variational Bayes , 2013, ICLR.

[8]  Guido Gerig,et al.  A brain tumor segmentation framework based on outlier detection , 2004, Medical Image Anal..

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

[10]  Simon Andermatt,et al.  Pathology Segmentation using Distributional Differences to Images of Healthy Origin , 2018, BrainLes@MICCAI.

[11]  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..

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

[13]  Christos Davatzikos,et al.  Advancing The Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features , 2017, Scientific Data.

[14]  Ender Konukoglu,et al.  Unsupervised Lesion Detection via Image Restoration with a Normative Prior , 2018, MIDL.

[15]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[16]  Christos Davatzikos,et al.  Abnormality Detection via Iterative Deformable Registration and Basis-Pursuit Decomposition , 2016, IEEE Transactions on Medical Imaging.

[17]  Klaus H. Maier-Hein,et al.  Context-encoding Variational Autoencoder for Unsupervised Anomaly Detection , 2018, ArXiv.

[18]  Ben Glocker,et al.  Deep Generative Models in the Real-World: An Open Challenge from Medical Imaging , 2018, ArXiv.