Anomaly Detection for Skin Disease Images Using Variational Autoencoder

Abstract

In this paper, we demonstrate the potential of applying Variational Autoencoder (VAE) [10] for anomaly detection in skin disease images. VAE is a class of deep generative models which is trained by maximizing the evidence lower bound of data distribution [10]. When trained on only normal data, the resulting model is able to perform efficient inference and to determine if a test image is normal or not. We perform experiments on ISIC2018 Challenge Disease Classification dataset (Task 3) and compare different methods to use VAE to detect anomaly. The model is able to detect all diseases with 0.779 AUCROC. If we focus on specific diseases, the model is able to detect melanoma with 0.864 AUCROC and detect actinic keratosis with 0.872 AUCROC, even if it only sees the images of nevus. To the best of our knowledge, this is the first applied work of deep generative models for anomaly detection in dermatology.

8 pages, 2 figures, submitted to ISIC Skin Image Analysis Workshop and Challenge at MICCAI 2018

References in corpus (2)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery

Cited by in corpus (8)

• Deep Learning for Anomaly Detection: A Survey
• Patch SVDD: Patch-level SVDD for Anomaly Detection and Segmentation
• Anomaly Detection based on Zero-Shot Outlier Synthesis and Hierarchical Feature Distillation
• Anomaly Detection in Images
• Evaluation of pseudo-healthy image reconstruction for anomaly detection with deep generative models: Application to brain FDG PET
• Out-of-Distribution Detection for Skin Lesion Images with Deep Isolation Forest
• Autoencoding Under Normalization Constraints
• Anomaly Detection for Solder Joints Using $\beta$-VAE