Ahmed Faisal, Bhuiyan Mohammad Alfrad Nobel, Coskunuzer Baris
Department of Data Science and Mathematics, Embry-Riddle Aeronautical University, 3700 Willow Creek Rd, 86301, Prescott, AZ, USA.
Department of Medicine, Louisiana State University Health Sciences Center at Shreveport, 1501 Kings Highway, 71103, Shreveport, LA, USA.
J Imaging Inform Med. 2025 Jun 25. doi: 10.1007/s10278-025-01575-7.
The analysis of fundus images is vital for early detection of retinal diseases such as diabetic retinopathy (DR), glaucoma, and age-related macular degeneration (AMD), but traditional methods are resource-intensive. We propose an automated and interpretable diagnostic framework that leverages novel feature representations to improve performance. Our main contribution is a topological feature extraction technique based on Topological Data Analysis (TDA), which captures geometric and structural patterns in fundus images. These features are computationally efficient and interpretable. We integrate them with pretrained CNN features (e.g., ResNet-50) into a hybrid deep model, Topo-CNN, combining global image context with topological structure. We evaluate Topo-CNN on three benchmarks: APTOS (binary and five-class DR), ORIGA (Glaucoma), and IChallenge-AMD. Our model achieves 98.7% accuracy/98.9 AUC on binary DR, 95.5 AUC on five-class DR, 93.8% accuracy/93.6 AUC on AMD, and 82.3% accuracy/95.8 specificity on glaucoma. Ablation studies confirm the added value of topological features, and our Topo-CNN consistently outperforms existing methods across tasks.
眼底图像分析对于诸如糖尿病视网膜病变（DR）、青光眼和年龄相关性黄斑变性（AMD）等视网膜疾病的早期检测至关重要，但传统方法资源消耗大。我们提出了一种自动化且可解释的诊断框架，该框架利用新颖的特征表示来提高性能。我们的主要贡献是一种基于拓扑数据分析（TDA）的拓扑特征提取技术，它能捕捉眼底图像中的几何和结构模式。这些特征计算效率高且可解释。我们将它们与预训练的卷积神经网络（CNN）特征（如ResNet - 50）集成到一个混合深度模型Topo - CNN中，将全局图像上下文与拓扑结构相结合。我们在三个基准数据集上评估Topo - CNN：APTOS（二元和五类DR）、ORIGA（青光眼）和IChallenge - AMD。我们的模型在二元DR上达到了98.7%的准确率/98.9的AUC，在五类DR上达到了95.5的AUC，在AMD上达到了93.8%的准确率/93.6的AUC，在青光眼上达到了82.3%的准确率/95.8的特异性。消融研究证实了拓扑特征的附加价值，并且我们的Topo - CNN在各项任务中始终优于现有方法。
