A Comprehensive Approach to Glaucoma Detection in Retinal Fundus Images with GlaucVGL-Net

In the evolving field of secure medical image processing, ensuring reliability and accuracy plays a vital role. Glaucoma, one of the leading causes of blindness worldwide, remains difficult to detect in its early stages due to its slow progression and complex nature. Traditional methods often miss early signs, which is why better detection tools are so crucial. GlaucVGL-Net is a novel approach to automatically detect glaucoma using retinal fundus images. This model combines the strength of two advanced learning techniques: VGG-19, which is really good at extracting important details from images, and LSTM, which excels at understanding sequences of data. This integrated approach helps the process to clearly understand the features that have been obtained after preprocessing of retinal fundus images, which can be used for the detection of glaucoma. The proposed GlaucVGL-Net was evaluated on the DRISHTI dataset comprising 400 retinal fundus images, achieving an accuracy of 98.3%, sensitivity of 99.6%, and specificity of 99.42%. These results demonstrate the model's superior performance compared to existing methods, confirming its potential for reliable glaucoma detection.