Explainable Diabetic Retinopathy using EfficientNET*

Diabetic retinopathy (DR) is a medical condition due to diabetes mellitus that can damage the patient retina and cause blood leaks. This condition can cause different symptoms from mild vision problems to complete blindness if it is not timely treated. In this work, we propose the use of a deep learning architecture based on a recent convolutional neural network called EfficientNet to detect referable diabetic retinopathy (RDR) and vision-threatening DR. Tests were conducted on two public datasets, EyePACS and APTOS 2019. The obtained results achieve state-of-the-art performance and show that the proposed network leads to higher classification rates, achieving an Area Under Curve (AUC) of 0.984 for RDR and 0.990 for vision-threatening DR on EyePACS dataset. Similar performances are obtained for APTOS 2019 dataset with an AUC of 0.966 and 0.998 for referable and vision-threatening DR, respectively. An explainability algorithm was also developed and shows the efficiency of the proposed approach in detecting DR signs.

[1]  Enhua Wu,et al.  Squeeze-and-Excitation Networks , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  E. Finkelstein,et al.  Development and Validation of a Deep Learning System for Diabetic Retinopathy and Related Eye Diseases Using Retinal Images From Multiethnic Populations With Diabetes , 2017, JAMA.

[3]  Ramprasaath R. Selvaraju,et al.  Grad-CAM: Why did you say that? Visual Explanations from Deep Networks via Gradient-based Localization , 2016 .

[4]  Matthew D. Davis,et al.  Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales. , 2003, Ophthalmology.

[5]  Sheikh Muhammad Saiful Islam,et al.  Deep Learning based Early Detection and Grading of Diabetic Retinopathy Using Retinal Fundus Images , 2018, ArXiv.

[6]  Mark Sandler,et al.  MobileNetV2: Inverted Residuals and Linear Bottlenecks , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[7]  Tien Yin Wong,et al.  Artificial intelligence using deep learning to screen for referable and vision-threatening diabetic retinopathy in Africa: a clinical validation study. , 2019, The Lancet. Digital health.

[8]  Laude,et al.  FEEDBACK ON A PUBLICLY DISTRIBUTED IMAGE DATABASE: THE MESSIDOR DATABASE , 2014 .

[9]  Kimmo Kaski,et al.  Deep Learning Fundus Image Analysis for Diabetic Retinopathy and Macular Edema Grading , 2019, Scientific Reports.

[10]  Subhashini Venugopalan,et al.  Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. , 2016, JAMA.

[11]  Quoc V. Le,et al.  GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism , 2018, ArXiv.

[12]  Quoc V. Le,et al.  EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks , 2019, ICML.

[13]  Bo Chen,et al.  MnasNet: Platform-Aware Neural Architecture Search for Mobile , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  A. Rakhlin.  Diabetic Retinopathy detection through integration of Deep Learning classification framework , 2017, bioRxiv.

[15]  Huazhu Fu,et al.  Evaluation of Retinal Image Quality Assessment Networks in Different Color-spaces , 2019, MICCAI.