摘要
本文提出一种名为E2E-DRNet的模型,旨在解决当前人工糖尿病视网膜病变(diabetic retinopathy,DR)诊断中分类性能差、耗时费力以及视网膜图像等级差异小、病灶不明显等问题.该模型基于EfficientNetV2,并结合了有效通道注意力模块.通过对DR数据集进行处理和优化,引入了Focal Loss损失函数以解决样本不均衡问题,并将模型分为两个阶段以实现DR分类的细分.实验结果表明,该方法在公开数据集和临床数据集上表现良好,提高了对眼底病变区域的可解释性,有助于提高DR病变的筛查效率,克服了人工诊断的局限性.
This study proposes a model called E2E-DRNet to address issues in manual diabetic retinopathy(DR)diagnosis,including poor classification performance,laborious processes,minimal differences in grades of retinal images,and inconspicuous lesions.This model is based on EfficientNetV2 and incorporates the efficient channel attention(ECA)module.By processing and optimizing a DR dataset,the Focal Loss function is introduced to address sample imbalance.The model achieves refined DR classification through two stages.Experimental results demonstrate that the proposed model performs well on both public and clinical datasets.Additionally,it enhances the interpretability of lesion regions in fundus images,thereby improving the efficiency of DR lesion screening and overcoming the limitations of manual diagnosis.
作者
刘圆圆
陈麓
鲁峰
叶阳
安禹潼
金明慧
邢开原
曾光
LIU Yuan-Yuan;CHEN Lu;LU Feng;YE Yang;AN Yu-Tong;JIN Ming-Hui;XING Kai-Yuan;ZENG Guang(School of Medical Informatics,Harbin Medical University(Daqing),Daqing 163319,China;Department of Ophthalmology,Daqing Oilfield General Hospital,Daqing 163319,China)
出处
《计算机系统应用》
2024年第12期248-255,共8页
Computer Systems & Applications
基金
黑龙江省省属高等学校基本科研业务费基础研究项目(JFQN202303)。
