Deep image mining for diabetic retinopathy screening

Abstract

Deep learning is quickly becoming the leading methodology for medical image analysis. Given a large medical archive, where each image is associated with a diagnosis, efficient pathology detectors or classifiers can be trained with virtually no expert knowledge about the target pathologies. However, deep learning algorithms, including the popular ConvNets, are black boxes: little is known about the local patterns analyzed by ConvNets to make a decision at the image level. A solution is proposed in this paper to create heatmaps showing which pixels in images play a role in the image-level predictions. In other words, a ConvNet trained for image-level classification can be used to detect lesions as well. A generalization of the backpropagation method is proposed in order to train ConvNets that produce high-quality heatmaps. The proposed solution is applied to diabetic retinopathy (DR) screening in a dataset of almost 90,000 fundus photographs from the 2015 Kaggle Diabetic Retinopathy competition and a private dataset of almost 110,000 photographs (e-ophtha). For the task of detecting referable DR, very good detection performance was achieved: $A_z=0.954$ in Kaggle's dataset and $A_z=0.949$ in e-ophtha. Performance was also evaluated at the image level and at the lesion level in the DiaretDB1 dataset, where four types of lesions are manually segmented: microaneurysms, hemorrhages, exudates and cotton-wool spots. The proposed detector outperforms recent algorithms trained to detect those lesions specifically, as well as competing heatmap generation algorithms for ConvNets. This detector is part of the Messidor system for mobile eye pathology screening. Because it does not rely on expert knowledge or manual segmentation for detecting relevant patterns, the proposed solution is a promising image mining tool, which has the potential to discover new biomarkers in images.

Accepted for publication in Medical Image Analysis

References in corpus (6)

• Adam: A Method for Stochastic Optimization
• Improving neural networks by preventing co-adaptation of feature detectors
• Understanding Neural Networks Through Deep Visualization
• Fractional Max-Pooling
• Improved Microaneurysm Detection using Deep Neural Networks
• Ensemble of Deep Convolutional Neural Networks for Learning to Detect Retinal Vessels in Fundus Images

Cited by in corpus (20)

• Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications
• Application of Deep Learning in Fundus Image Processing for Ophthalmic Diagnosis -- A Review
• DR$|$GRADUATE: uncertainty-aware deep learning-based diabetic retinopathy grading in eye fundus images
• ExplAIn: Explanatory Artificial Intelligence for Diabetic Retinopathy Diagnosis
• Transfer Learning based Detection of Diabetic Retinopathy from Small Dataset
• An Interpretable Multiple-Instance Approach for the Detection of referable Diabetic Retinopathy from Fundus Images
• Iterative Augmentation of Visual Evidence for Weakly-Supervised Lesion Localization in Deep Interpretability Frameworks: Application to Color Fundus Images
• Improving Lesion Segmentation for Diabetic Retinopathy using Adversarial Learning
• DISCOVER: 2-D Multiview Summarization of Optical Coherence Tomography Angiography for Automatic Diabetic Retinopathy Diagnosis
• Robust deep learning for eye fundus images: Bridging real and synthetic data for enhancing generalization
• Certifiably Robust Interpretation in Deep Learning
• Combining Fine- and Coarse-Grained Classifiers for Diabetic Retinopathy Detection
• Automated Smartphone based System for Diagnosis of Diabetic Retinopathy
• Instant automatic diagnosis of diabetic retinopathy
• Detailed Investigation of Deep Features with Sparse Representation and Dimensionality Reduction in CBIR: A Comparative Study
• A comprehensive survey on computer-aided diagnostic systems in diabetic retinopathy screening
• Institutionally Distributed Deep Learning Networks
• Point-of-Care Diabetic Retinopathy Diagnosis: A Standalone Mobile Application Approach
• Automatic detection of rare pathologies in fundus photographs using few-shot learning
• An Efficient Framework for Automated Screening of Clinically Significant Macular Edema