Optimized Deep Convolutional Neural Networks for Identification of Macular Diseases from Optical Coherence Tomography Images

Citations Over TimeTop 10% of 2019 papers

Abstract

Finetuning pre-trained deep neural networks (DNN) delicately designed for large-scale natural images may not be suitable for medical images due to the intrinsic difference between the datasets. We propose a strategy to modify DNNs, which improves their performance on retinal optical coherence tomography (OCT) images. Deep features of pre-trained DNN are high-level features of natural images. These features harm the training of transfer learning. Our strategy is to remove some deep convolutional layers of the state-of-the-art pre-trained networks: GoogLeNet, ResNet and DenseNet. We try to find the optimized deep neural networks on small-scale and large-scale OCT datasets, respectively, in our experiments. Results show that optimized deep neural networks not only reduce computational burden, but also improve classification accuracy.

Related Papers

• → Comparative Assessment among Different Convolutional Neural Network Architectures for Alzheimer’s Disease Detection(2021)18 cited
• → Development of Breast Cancer Detection Model Using Transfer Learning of Residual Neural Network (ResNet-50)(2020)13 cited
• → Noninvasive Three-Dimensional Analysis of In Vivo Human Skin Using the Next-Generation Optical Coherence Tomography (OCT), Fourier Domain OCT(2008)1 cited
• Residual Network 모델의 특징에 관한 실증비교연구(2016)
• → Study for Performance of Un-Pretrained and Pre-trained Models based on CNN(2023)