Data Augmentation for Skin Lesion Analysis

Abstract

Deep learning models show remarkable results in automated skin lesion analysis. However, these models demand considerable amounts of data, while the availability of annotated skin lesion images is often limited. Data augmentation can expand the training dataset by transforming input images. In this work, we investigate the impact of 13 data augmentation scenarios for melanoma classification trained on three CNNs (Inception-v4, ResNet, and DenseNet). Scenarios include traditional color and geometric transforms, and more unusual augmentations such as elastic transforms, random erasing and a novel augmentation that mixes different lesions. We also explore the use of data augmentation at test-time and the impact of data augmentation on various dataset sizes. Our results confirm the importance of data augmentation in both training and testing and show that it can lead to more performance gains than obtaining new images. The best scenario results in an AUC of 0.882 for melanoma classification without using external data, outperforming the top-ranked submission (0.874) for the ISIC Challenge 2017, which was trained with additional data.

8 pages, 3 figures, to be presented on ISIC Skin Image Analysis Workshop

References in corpus (9)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Random Erasing Data Augmentation
• Automatic Skin Lesion Analysis using Large-scale Dermoscopy Images and Deep Residual Networks
• Knowledge Transfer for Melanoma Screening with Deep Learning
• Image Classification of Melanoma, Nevus and Seborrheic Keratosis by Deep Neural Network Ensemble
• Skin Lesion Synthesis with Generative Adversarial Networks
• RECOD Titans at ISIC Challenge 2017
• Skin Lesion Analysis Toward Melanoma Detection: A Challenge at the 2017 International Symposium on Biomedical Imaging (ISBI), Hosted by the International Skin Imaging Collaboration (ISIC)
• Towards Automated Melanoma Screening: Proper Computer Vision & Reliable Results

Cited by in corpus (16)

• The impact of patient clinical information on automated skin cancer detection
• Skin Lesion Synthesis with Generative Adversarial Networks
• Melanoma Detection using Adversarial Training and Deep Transfer Learning
• Exploring convolutional neural networks with transfer learning for diagnosing Lyme disease from skin lesion images
• Attention-based Fully Gated CNN-BGRU for Russian Handwritten Text
• Handling Inter-Annotator Agreement for Automated Skin Lesion Segmentation
• Data, Depth, and Design: Learning Reliable Models for Skin Lesion Analysis
• A Smartphone based Application for Skin Cancer Classification Using Deep Learning with Clinical Images and Lesion Information
• Deep-Learning Ensembles for Skin-Lesion Segmentation, Analysis, Classification: RECOD Titans at ISIC Challenge 2018
• Recent advances in deep learning applied to skin cancer detection
• Artificial Intelligence-Based Image Classification for Diagnosis of Skin Cancer: Challenges and Opportunities
• (De)Constructing Bias on Skin Lesion Datasets
• Data Augmentation for Skin Lesion using Self-Attention based Progressive Generative Adversarial Network
• Learning Test-time Augmentation for Content-based Image Retrieval
• When Medical Imaging Met Self-Attention: A Love Story That Didn't Quite Work Out
• Class-dependent Compression of Deep Neural Networks