Enhanced Detection of Referable Diabetic Retinopathy via DCNNs and Transfer Learning
| Autor: | Yuchen Xie, Mong Li Lee, Nguyen Duc Quang, Tien Yin Wong, Haslina Hamzah, Xin Qi Lee, Wynne Hsu, Jinyi Ho, Valentina Bellemo, Gilbert Lim, Michelle Y.T. Yip, Zhan Wei Lim, Daniel Shu Wei Ting |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Computer science
business.industry Deep learning Diabetic retinopathy medicine.disease Convolutional neural network Residual neural network 030218 nuclear medicine & medical imaging Screening programme 03 medical and health sciences 0302 clinical medicine 030221 ophthalmology & optometry medicine Sensitivity (control systems) Artificial intelligence business Transfer of learning |
| Zdroj: | Computer Vision – ACCV 2018 Workshops ISBN: 9783030210731 ACCV Workshops |
| DOI: | 10.1007/978-3-030-21074-8_23 |
| Popis: | A clinically acceptable deep learning system (DLS) has been developed for the detection of diabetic retinopathy by the Singapore Eye Research Institute. For its utility in a national screening programme, further enhancement was needed. With newer deep convolutional neural networks (DCNNs) being introduced and technological methodology such as transfer learning gaining recognition for better performance, this paper compared the performance of the DCNN used in the original DLS, VGGNet, with newer DCNNs, ResNet and Ensemble, with transfer learning. The DLS performance improved with higher AUC, sensitivity and specificity with the adoption of the newer DCNNs and transfer learning. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|