Enhancement of retinal images through modified anisotropic diffusion

Background: Diabetic retinopathy (DR) occurs in the patient having past history of diabetes for around 5 to 10 years. DR is a significant contributor to blindness among individuals aged 20 to 64 years, particularly in those who had diabetes for an extended duration. This study presents a robust method for improving the quality of retinal images in computer-aided diagnosis (CAD) systems, with a primary focus on enhancing contrast for better visual perception in assessing retinal diseases like DR and hypertensive retinopathy. Methods: The proposed approach integrates a modified anisotropic diffusion technique and various filters to elevate image quality. The process involves converting RGB to HSV, extracting the V channel, and applying a 5×5 window modified anisotropic diffusion filter. Further enhancements are achieved through the application of Wiener and unsharp masking filters. Results: The results obtained were highly promising, with an average peak signal-to-noise ratio (PSNR) of 45.31±40.78, an average structural similarity index measure (SSIM) of 0.97, and an average entropy value of 7.18±6.17. These quantitative evaluation metrics clearly indicate the effectiveness of the proposed approach. Conclusions: The results demonstrate a substantial improvement in image quality, characterized by enhanced contrast and improved visual clarity. In conclusion, this research contributes valuable insights to the realm of retinal image enhancement for CAD, making notable strides in the field of medical imaging.