Ejaz Sara, Zia Hafiz U, Majeed Fiaz, Shafique Umair, Altamiranda Stefania Carvajal, Lipari Vivian, Ashraf Imran
Department of Information Technology, University of Gujrat, Gujrat, Pakistan.
Universidad Europea del Atlantico, Santander, Spain.
Digit Health. 2025 Mar 28;11:20552076251328120. doi: 10.1177/20552076251328120. eCollection 2025 Jan-Dec.
Deep learning models assist ophthalmologists in early detection of diseases from retinal images and timely treatment.
Owing to robust and accurate results from deep learning models, we aim to use convolutional neural network (CNN) to provide a non-invasive method for early detection of eye diseases.
We used a hybridized CNN with deep learning (DL) based on two separate CNN blocks, to identify multiple Optic Disc Cupping, Diabetic Retinopathy, Media Haze, and Healthy images. We used the RFMiD dataset, which contains various categories of fundus images representing different eye diseases. Data augmenting, resizing, coping, and one-hot encoding are used among other preprocessing techniques to improve the performance of the proposed model. Color fundus images have been analyzed by CNNs to extract relevant features. Two CCN models that extract deep features are trained in parallel. To obtain more noticeable features, the gathered features are further fused utilizing the Canonical Correlation Analysis fusion approach. To assess the effectiveness, we employed eight classification algorithms: Gradient boosting, support vector machines, voting ensemble, medium KNN, Naive Bayes, COARSE- KNN, random forest, and fine KNN.
With the greatest accuracy of 93.39%, the ensemble learning performed better than the other algorithms.
The accuracy rates suggest that the deep learning model has learned to distinguish between different eye disease categories and healthy images effectively. It contributes to the field of eye disease detection through the analysis of color fundus images by providing a reliable and efficient diagnostic system.
深度学习模型可协助眼科医生从视网膜图像中早期发现疾病并及时进行治疗。
鉴于深度学习模型能产生强大且准确的结果，我们旨在使用卷积神经网络（CNN）提供一种用于早期检测眼部疾病的非侵入性方法。
我们基于两个独立的CNN模块使用了一种结合深度学习（DL）的混合CNN，以识别多种视盘杯状凹陷、糖尿病视网膜病变、介质混浊和健康图像。我们使用了RFMiD数据集，该数据集包含代表不同眼部疾病的各类眼底图像。在其他预处理技术中使用了数据增强、调整大小、复制和独热编码，以提高所提出模型的性能。彩色眼底图像已通过CNN进行分析以提取相关特征。两个提取深度特征的CCN模型并行训练。为了获得更显著的特征，利用典型相关分析融合方法对收集到的特征进行进一步融合。为了评估有效性，我们采用了八种分类算法：梯度提升、支持向量机、投票集成、中等K近邻、朴素贝叶斯、粗糙K近邻、随机森林和精细K近邻。
集成学习的准确率最高，达到93.39%，其表现优于其他算法。
准确率表明深度学习模型已学会有效区分不同的眼部疾病类别和健康图像。通过分析彩色眼底图像，它为眼部疾病检测领域做出了贡献，提供了一个可靠且高效的诊断系统。
