Year-end Sale % OFF 
Abstract
BackgroundConvolution neural networks have been considered for automatic analysis of fundus images to detect signs of diabetic retinopathy but suffer from low sensitivity.MethodsThis study has proposed an alternate method using probabilistic output from Convolution neural network to automatically and simultaneously detect exudates, hemorrhages and microaneurysms. The method was evaluated using two approaches: patch and image-based analysis of the fundus images on two public databases: DIARETDB1 and e-Ophtha. The novelty of the proposed method is that the images were analyzed using probability maps generated by score values of the softmax layer instead of the use of the binary output.ResultsThe sensitivity of the proposed approach was 0.96, 0.84 and 0.85 for detection of exudates, hemorrhages and microaneurysms, respectively when considering patch-based analysis. The results show overall accuracy for DIARETDB1 was 97.3% and 86.6% for e-Ophtha. The error rate for image-based analysis was also significantly reduced when compared with other works.ConclusionThe proposed method provides the framework for convolution neural network-based analysis of fundus images to identify exudates, hemorrhages, and microaneurysms. It obtained accuracy and sensitivity which were significantly better than the reported studies and makes it suitable for automatic diabetic retinopathy signs detection.
Highlights
Convolution neural networks have been considered for automatic analysis of fundus images to detect signs of diabetic retinopathy but suffer from low sensitivity
For the patch-based evaluation, the mean results of ten repetitions for the training are described in Table 5 and Fig. 10 shows the Receiver Operating Characteristic (ROC) curve for the Convolutional Neural Network (CNN) performance
It is observed that for DIARETDB1, the proposed method achieved the accuracy of 0.96, 0.98 and 0.97 and error rate of 3.9%, 2.1% and 2.04% for segmentation of exudate, hemorrhage and microaneurysm, respectively which shows that this technique outperforms techniques reported in literature
Summary
Convolution neural networks have been considered for automatic analysis of fundus images to detect signs of diabetic retinopathy but suffer from low sensitivity. Vision impairment due to DR can be significantly reduced if it is diagnosed in the early stages. It is diagnosed by visual examination of retinal images to detect three most common pathological signs i.e. (i) exudate (ii) hemorrhage and (iii) microaneurysm [6]. This is a manual time-consuming procedure and outcomes are subjective and dependent on expertise, there is potential bias of the examiner.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
