基于加速复用卷积网络的视网膜疾病检测
Detection of retinal diseases using an accelerated reused convolutional network
AI总结：
本研究针对现有眼部疾病检测神经网络计算复杂度高的问题，通过设计优化ArConv卷积层构建轻量通用模型，仅130万参数，在RfMiD数据集上准确率优于同条件下的MobileNetV2，适配移动端部署。
摘要：
卷积神经网络持续演进，部分研究致力于提升准确率，部分聚焦于提高速度，还有部分旨在增强可部署性。提升可部署性能够拓宽神经网络在更多任务中的应用范围，包括眼部疾病检测。眼部疾病的早期诊断与眼科医生问诊可预防诸多视力障碍。鉴于该问题的重要性，研究人员已从角膜采集了各类数据集，以推动神经网络模型的构建流程。但过往提出的多数方法计算复杂度较高。本研究尝试提升深度神经网络模型的可部署性，我们从最基础层面着手，具体通过重新设计并优化卷积层来实现。借此我们构建了一种全新的通用模型，该模型融入了我们提出的新型卷积层——ArConv层（加速复用卷积层）。得益于这种新型层的高效性能，该模型复杂度适配手机端使用，且能以高准确率完成疾病存在性诊断任务。我们提出的最终模型仅包含130万个参数。与拥有220万个参数的MobileNetV2模型相比，在相同条件下于RfMiD数据集上进行训练和评估时，我们的模型展现出更优的准确率：在RfMiD测试集上准确率达0.9328，而MobileNetV2为0.9266。
英文摘要：
Convolutional neural networks are continually evolving, with some efforts aimed at improving accuracy, others at increasing speed, and some at enhancing accessibility. Improving accessibility broadens the application of neural networks across a wider range of tasks, including the detection of eye diseases. Early diagnosis of eye diseases and consulting an ophthalmologist can prevent many vision disorders. Given the importance of this issue, various datasets have been collected from the cornea to facilitate the process of making neural network models. However, most of the methods introduced in the past are computationally complex. In this study, we tried to increase the accessibility of deep neural network models. We did this at the most fundamental level, specifically by redesigning and optimizing the convolutional layers. By doing so, we created a new general model that incorporates our novel convolutional layer named ArConv layers. Thanks to the efficient performance of this new layer, the model has suitable complexity for use in mobile phones and can perform the task of diagnosing the presence of disease with high accuracy. The final model we present contains only 1.3 million parameters. In comparison to the MobileNetV2 model, which has 2.2 million parameters, our model demonstrated better accuracy when trained and evaluated on the RfMiD dataset under identical conditions, achieving an accuracy of 0.9328 versus 0.9266 on the RfMiD test set.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
