Improving Robustness using Joint Attention Network For Detecting Retinal Degeneration From Optical Coherence Tomography Images

16 May 2020  ·  Sharif Amit Kamran, Alireza Tavakkoli, Stewart Lee Zuckerbrod ·

Noisy data and the similarity in the ocular appearances caused by different ophthalmic pathologies pose significant challenges for an automated expert system to accurately detect retinal diseases. In addition, the lack of knowledge transferability and the need for unreasonably large datasets limit clinical application of current machine learning systems. To increase robustness, a better understanding of how the retinal subspace deformations lead to various levels of disease severity needs to be utilized for prioritizing disease-specific model details. In this paper we propose the use of disease-specific feature representation as a novel architecture comprised of two joint networks -- one for supervised encoding of disease model and the other for producing attention maps in an unsupervised manner to retain disease specific spatial information. Our experimental results on publicly available datasets show the proposed joint-network significantly improves the accuracy and robustness of state-of-the-art retinal disease classification networks on unseen datasets.

PDF Abstract

Datasets


  Add Datasets introduced or used in this paper
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Retinal OCT Disease Classification OCT2017 Joint-Attention-Network MobileNet-v2 Acc 95.6 # 10
Retinal OCT Disease Classification OCT2017 Joint-Attention-Network ResNet50-v1 Acc 92.4 # 13
Retinal OCT Disease Classification OCT2017 Joint-Attention-Network OpticNet-71 Acc 77.4 # 14
Retinal OCT Disease Classification Srinivasan2014 Joint-Attention-Network ResNet50-v1 Acc 100 # 1
Retinal OCT Disease Classification Srinivasan2014 Joint-Attention-Network MobileNet-v2 Acc 99.36 # 4
Retinal OCT Disease Classification Srinivasan2014 Joint-Attention-Network OpticNet-71 Acc 99.68 # 3

Methods


No methods listed for this paper. Add relevant methods here