BiRA-Net: Bilinear Attention Net for Diabetic Retinopathy Grading
TL
TL;DR
A new deep learning architecture, called BiRA-Net, is proposed, which combines the attention model for feature extraction and bilinear model for fine-grained classification, and a new loss function, called grading loss, which leads to improved training convergence of the proposed approach.
Abstract
Diabetic retinopathy (DR) is a common retinal disease that leads to blindness. For diagnosis purposes, DR image grading aims to provide automatic DR grade classification, which is not addressed in conventional research methods of binary DR image classification. Small objects in the eye images, like lesions and microaneurysms, are essential to DR grading in medical imaging, but they could easily be influenced by other objects. To address these challenges, we propose a new deep learning architecture, called BiRA-Net, which combines the attention model for feature extraction and bilinear model for fine-grained classification. Furthermore, in considering the distance between different grades of different DR categories, we propose a new loss function, called grading loss, which leads to improved training convergence of the proposed approach. Experimental results are provided to demonstrate the superior performance of the proposed approach.
References19 items
1
Semi-Supervised Self-Taught Deep Learning for Finger Bones Segmentation
2
Bayesian VoxDRN: A Probabilistic Deep Voxelwise Dilated Residual Network for Whole Heart Segmentation from 3D MR Images
3
Multi-Cell Multi-Task Convolutional Neural Networks for Diabetic Retinopathy Grading
4
Survey on diagnosis of diseases from retinal images
5