Abstract

This cross-sectional study aims to detect Diabetic Retinopathy (DR) in patients who have had retinal scans and ophthalmological exams. The research makes use of tailored retinal images together with the OPF (Optimum-Path Forest) and RBM (Restricted Boltzmann Machine) models to categorize images according to the presence or absence of DR. In this work, features were extracted from the retinal images using both the RBM and OPF models. In particular, after a thorough system training phase, RBM was able to extract between 500 and 1000 features from the images. The study included fifteen distinct trial series, each with thirty cycles of repetition. The research comprised 122 eyes, or 73 diabetic patients, with a gender distribution that was reasonably balanced and an average age of 59.7 years. Remarkably, the RBM-1000 model stood out as the top performer, with the highest overall accuracy of 89.47% in diagnosis. In terms of specificity, the RBM-1000 and OPF-1000 models surpassed the competition, correctly categorizing all images free of DR symptoms. These findings highlight the potential of machine learning, particularly the RBM model, for self-identifying illnesses. The potential of machine learning models—in particular, RBM and OPF—to automate the diagnosis of diabetic retinopathy is demonstrated by this work. The results show how well the RBM model diagnoses, how sensitive it is, and how well it can be applied for efficient DR screening and diagnosis. This information may be used to improve the effectiveness of systems that identify retinal illnesses.

CONCLUSION

In conclusion, employing VGG16 and VGG19 architectures for diabetes detection harnesses their robust feature extraction capabilities, derived from their deep and hierarchical structures. Leveraging transfer learning by pre-training on large image datasets such as ImageNet and fine-tuning for diabetic-specific data proves advantageous, enhancing model performance with limited labeled samples. However, the trade-off between model complexity and computational efficiency must be considered, as these architectures are characterized by a substantial number of parameters. The success of diabetes detection using VGG16 and VGG19 hinges on the quality and quantity of training data, demanding diverse and representative datasets. Real-world deployment considerations, including real-time processing and system integration, add practical challenges to the implementation of these models. Model interpretability remains an ongoing concern due to the inherent “black-box” nature of deep learning models, prompting the exploration of interpretability tools for better understanding. Lastly, the dynamic nature of deep learning research implies a need for continuous monitoring of advancements that may introduce new architectures or techniques, potentially refining and enhancing the effectiveness of VGG16 and VGG19 in diabetic detection systems.

Authors: Arepalli Gopi, Sudha L.R, Iwin Thanakumar Joseph,  Ewin Thanakumar Joseph S.

Department of Computer Science & Engineering, Annamalai University, Chidambaram, Tamilnadu, India

View Full Abstract:  https://iridology-research.com/pdf/AGS_2025_197.pdf