An ensemble deep learning based approach for red lesion detection in fundus images

  title={An ensemble deep learning based approach for red lesion detection in fundus images},
  author={Jos{\'e} Ignacio Orlando and Elena Prokofyeva and Mariana del Fresno and Matthew B. Blaschko},
  journal={Computer methods and programs in biomedicine},

A Deep Learning-Based Unified Framework for Red Lesions Detection on Retinal Fundus Images

A two-stream red lesions detection system dealing simultaneously with small and large red lesions on fundus images, based on blood vessel segmentation and morphological operations, and reduces the computational complexity, and enhances the detection accuracy by generating a small number of potential candidates.

Automated Detection and Diagnosis of Diabetic Retinopathy: A Comprehensive Survey

This review covers the literature dealing with AI approaches to DR such as ML and DL in classification and segmentation that have been published in the open literature within six years (2016–2021) and a comprehensive list of available DR datasets is reported.

A Transfer Learning and U-Net-based automatic detection of diabetic retinopathy from fundus images

This study proposed a novel two-stage framework for automatic DR classification using two distinct U-Net models for optic disc and blood vessel segmentation during the preprocessing and achieved state-of-the-art performance.

Deep Learning based Computer-Aided Diagnosis Systems for Diabetic Retinopathy: A Survey

Diabetic Retinopathy Pathological Signs Detection using Image Enhancement Technique and Deep Learning

An automated machine learning algorithm for detecting diabetic retinopathy in fundus images is introduced and a slight improvement in classification accuracy is revealed, compared to those original images with no enhancement for both datasets.

A deep learning model for classification of diabetic retinopathy in eye fundus images based on retinal lesion detection

The proposed method follows the same workflow as the clinicians, providing information that can be interpreted clinically to support the prediction, and has an area-under-the-curve, sensitivity, and specificity of 0.948, 0.886, and 0.875, respectively.

Robust Localization of Retinal Lesions via Weakly-supervised Learning

This paper presents a new approach to discriminate the location of various lesions based on image-level labels via weakly learning, which leverages the multilevel feature maps and classification score to cope with both bright and red lesions in fundus images.



Beyond Lesion-Based Diabetic Retinopathy: A Direct Approach for Referral

The proposed technique for direct referral is promising, achieving an area under the curve of 96.4%, thus, reducing the classification error by almost 40% over the current state of the art, held by lesion-based techniques.

An Incremental Feature Extraction Framework for Referable Diabetic Retinopathy Detection

  • Jay NandyW. HsuM. Lee
  • Computer Science
    2016 IEEE 28th International Conference on Tools with Artificial Intelligence (ICTAI)
  • 2016
This paper first learns a universal Gaussian mixture model (GMM) from a small set of annotated images and applies this universal GMM as the prior belief to learn an adaptive GMM for individual images to capture the characteristics of referable versus non-referable images.

Red Lesion Detection Using Dynamic Shape Features for Diabetic Retinopathy Screening

A novel method for automatic detection of both microaneurysms and hemorrhages in color fundus images is described and validated and it proves to be robust with respect to variability in image resolution, quality and acquisition system.

Validation of microaneurysm-based diabetic retinopathy screening across retina fundus datasets

A straightforward pipeline from microaneurysm (an early sign of DR) detection to automatic classification of DR without employing any additional features is proposed and the generalisation ability of the MA detection method is quantified by employing synthetic examples.

Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs.

An algorithm based on deep machine learning had high sensitivity and specificity for detecting referable diabetic retinopathy and diabetic macular edema in retinal fundus photographs from adults with diabetes.

Automatic detection of red lesions in digital color fundus photographs

A novel red lesion detection method is presented based on a hybrid approach, combining prior works by Spencer et al. (1996) and Frame (1998) with two important new contributions, including a new red lesions candidate detection system based on pixel classification.

Convolutional neural network transfer for automated glaucoma identification

Results on the Drishti-GS1 dataset suggests the viability of using CNNs that are pre-trained from non-medical data for automated glaucoma detection and offer significant evidence of the importance of well-chosen image pre-processing for transfer learning when the amount of data is not sufficient for fine-tuning the network.

Automatic detection of microaneurysms in retinal fundus images

New Deep Neural Nets for Fine-Grained Diabetic Retinopathy Recognition on Hybrid Color Space

Two deep convolutional neural networks - Combined Kernels with Multiple Losses Network and VGGNet with Extra Kernel and VNXK are proposed, which are an improvement upon GoogLeNet and V GGNet in context of DR tasks and a hybrid color space, LGI, is proposed for DR recognition via proposed nets.