Mixed-decomposed convolutional network: A lightweight yet efficient convolutional neural network for ocular disease recognition

Eye health has become a global health concern and attracted broad attention. Over the years, researchers have proposed many state-of-the-art convolutional neural networks (CNNs) to assist ophthalmologists in diagnosing ocular diseases efficiently and precisely. However, most existing methods were dedicated to constructing sophisticated CNNs, inevitably ignoring the trade-off between performance and model complexity. To alleviate this paradox, this paper proposes a lightweight yet efficient network architecture, mixed-decomposed convolutional network (MDNet), to recognise ocular diseases. In MDNet, we introduce a novel mixed-decomposed depthwise convolution method, which takes advantage of depthwise convolution and depthwise dilated convolution operations to capture low-resolution and high-resolution patterns by using fewer computations and fewer parameters. We conduct extensive experiments on the clinical anterior segment optical coherence tomography (AS-OCT), LAG, University of California San Diego, and CIFAR-100 datasets. The results show our MDNet achieves a better trade-off between the performance and model complexity than efficient CNNs including MobileNets and MixNets. Specifically, our MDNet outperforms MobileNets by 2.5% of accuracy by using 22% fewer parameters and 30% fewer computations on the AS-OCT dataset.

Automated summary

Eye health is a global concern and researchers have proposed many advanced convolutional neural networks to aid in the diagnosis of ocular diseases. However, most of these methods focus on constructing complex networks, neglecting the trade-off between performance and model complexity. To address this, this paper introduces a lightweight yet efficient network architecture, MDNet, which utilizes a novel mixed-decomposed depthwise convolution method. Experimental results demonstrate that MDNet achieves a better trade-off between performance and model complexity compared to existing efficient CNNs.