Comparative Study of Fine-Tuning of Pre-Trained Convolutional Neural Networks for Diabetic Retinopathy Screening

2017 ◽  
Author(s):  
Saboora Mohammadian ◽  
Ali Karsaz ◽  
Yaser M. Roshan
Keyword(s):  
Neural Networks ◽  
Diabetic Retinopathy ◽  
Comparative Study ◽  
Convolutional Neural Networks ◽  
Fine Tuning ◽  
Diabetic Retinopathy Screening ◽  
Retinopathy Screening

scholarly journals Digital imaging in diabetic retinopathy screening?a comparative study of two digital cameras

2002 ◽  
Vol 19 (9) ◽  
pp. 287-289 ◽  
Author(s):  
A Basu ◽  
AD Kamal ◽  
W Illahi ◽  
M Khan ◽  
P Stavrou ◽  
...  
Keyword(s):  
Diabetic Retinopathy ◽  
Comparative Study ◽  
Digital Imaging ◽  
Digital Cameras ◽  
Diabetic Retinopathy Screening ◽  
Retinopathy Screening

scholarly journals Performance of deep neural network-based artificial intelligence method in diabetic retinopathy screening: a systematic review and meta-analysis of diagnostic test accuracy

2020 ◽  
Vol 183 (1) ◽  
pp. 41-49 ◽  
Author(s):  
Shirui Wang ◽  
Yuelun Zhang ◽  
Shubin Lei ◽  
Huijuan Zhu ◽  
Jianqiang Li ◽  
...  
Keyword(s):  
Neural Networks ◽  
Diabetic Retinopathy ◽  
Diagnostic Accuracy ◽  
Meta Analysis ◽  
Cochrane Library ◽  
Test Accuracy ◽  
Sample Sizes ◽  
Diabetic Retinopathy Screening ◽  
Retinopathy Screening ◽  
Training Sets

Objective Automatic diabetic retinopathy screening system based on neural networks has been used to detect diabetic retinopathy (DR). However, there is no quantitative synthesis of performance of these methods. We aimed to estimate the sensitivity and specificity of neural networks in DR grading. Methods Medline, Embase, IEEE Xplore, and Cochrane Library were searched up to 23 July 2019. Studies that evaluated performance of neural networks in detection of moderate or worse DR or diabetic macular edema using retinal fundus images with ophthalmologists’ judgment as reference standard were included. Two reviewers extracted data independently. Risk of bias of eligible studies was assessed using QUDAS-2 tool. Results Twenty-four studies involving 235 235 subjects were included. Quantitative random-effects meta-analysis using the Rutter and Gatsonis hierarchical summary receiver operating characteristics (HSROC) model revealed a pooled sensitivity of 91.9% (95% CI: 89.6% to 94.3%) and specificity of 91.3% (95% CI: 89.0% to 93.5%). Subgroup analyses and meta-regression did not provide any statistically significant findings for the heterogeneous diagnostic accuracy in studies with different image resolutions, sample sizes of training sets, architecture of convolutional neural networks, or diagnostic criteria. Conclusions State-of-the-art neural networks could effectively detect clinical significant DR. To further improve diagnostic accuracy of neural networks, researchers might need to develop new algorithms rather than simply enlarge sample sizes of training sets or optimize image quality.

scholarly journals Convolutional neural networks for mild diabetic retinopathy detection: an experimental study

2019 ◽  
Author(s):  
Rubina Sarki ◽  
Sandra Michalska ◽  
Khandakar Ahmed ◽  
Hua Wang ◽  
Yanchun Zhang
Keyword(s):  
Neural Networks ◽  
Diabetic Retinopathy ◽  
Performance Improvement ◽  
Convolutional Neural Networks ◽  
Large Scale ◽  
Data Augmentation ◽  
Comprehensive Evaluation ◽  
Optimal Solution ◽  
Fine Tuning ◽  
Fundus Photography

AbstractCurrently, Diabetes and the associated Diabetic Retinopathy (DR) instances are increasing at an alarming rate. Numerous previous research has focused on automated DR detection from fundus photography. The classification of severe cases of pathological indications in the eye has achieved over 90% accuracy. Still, the mild cases are challenging to detect due to CNN inability to identify the subtle features, discrimnative of disease. The data used (i.e. annotated fundus photographies) was obtained from 2 publicly available sources – Messidor and Kaggle. The experiments were conducted with 13 Convolutional Neural Networks architectures, pre-trained on large-scale ImageNet database using the concept of Transfer Learning. Several performance improvement techniques were applied, such as: (i) fine-tuning, (ii) data augmentation, and (iii) volume increase. The results were measured against the standard Accuracy metric on the testing dataset. After the extensive experimentation, the maximum Accuracy of 86% on No DR/Mild DR classification task was obtained for ResNet50 model with fine-tuning (un-freeze and re-train the layers from 100 onwards), and RMSProp Optimiser trained on the combined Messidor + Kaggle (aug) datasets. Despite promising results, Deep learning continues to be an empirical approach that requires extensive experimentation in order to arrive at the most optimal solution. The comprehensive evaluation of numerous CNN architectures was conducted in order to facilitate an early DR detection. Furthermore, several performance improvement techniques were assessed to address the CNN limitation in subtle eye lesions identification. The model also included various levels of image quality (low/high resolution, under/over-exposure, out-of-focus etc.), in order to prove its robustness and ability to adapt to real-world conditions.

Diabetic Retinopathy Screening: A Systematic Review of Qualitative Literature

2021 ◽  
Author(s):  
Oluwaseun Egunsola ◽  
Laura E. Dowsett ◽  
Ruth Diaz ◽  
Michael Brent ◽  
Valeria Rac ◽  
...  
Keyword(s):  
Systematic Review ◽  
Diabetic Retinopathy ◽  
Diabetic Retinopathy Screening ◽  
Qualitative Literature ◽  
Retinopathy Screening

Digital Ophthalmology in the UK – Diabetic Retinopathy Screening and Virtual Glaucoma Clinics in the National Health Service

2020 ◽  
Vol 237 (12) ◽  
pp. 1400-1408
Author(s):  
Heinrich Heimann ◽  
Deborah Broadbent ◽  
Robert Cheeseman
Keyword(s):  
Diabetic Retinopathy ◽  
National Health Service ◽  
Health Service ◽  
National Health ◽  
Daily Practice ◽  
Diabetic Retinopathy Screening ◽  
Technological Advances ◽  
Imaging Results ◽  
The Uk ◽  
Retinopathy Screening

AbstractThe customary doctor and patient interactions are currently undergoing significant changes through technological advances in imaging and data processing and the need for reducing person-to person contacts during the COVID-19 crisis. There is a trend away from face-to-face examinations to virtual assessments and decision making. Ophthalmology is particularly amenable to such changes, as a high proportion of clinical decisions are based on routine tests and imaging results, which can be assessed remotely. The uptake of digital ophthalmology varies significantly between countries. Due to financial constraints within the National Health Service, specialized ophthalmology units in the UK have been early adopters of digital technology. For more than a decade, patients have been managed remotely in the diabetic retinopathy screening service and virtual glaucoma clinics. We describe the day-to-day running of such services and the doctor and patient experiences with digital ophthalmology in daily practice.

Sign in / Sign up

Export Citation Format

Share Document