Clinical application of enhanced retinal vasculature visualization using hemoglobin absorbance

Purpose To develop a digital filter for better visualization of retinal blood vessels and verify its clinical usefulness Methods Four‐hundred 24‐bit color fundus images were analyzed and properties of red, green, and blue channels were extracted. Then, using hemoglobin absorption coefficients, the relevant weights for gray‐scale conversion that emphasizes retinal vessels were calculated. To evaluate images, edges were detected via convolutional 2D Laplacian kernel from the processed images, and the number of edges, number of effective edges, and sum of intensities of edges were evaluated. The values of weights for red, green, and blue channels were calculated to be ‐0.0572, 0.7335, and 2.2079, respectively. The number of edges, effective edges, and sum of intensities of edges were all found to be significantly higher in the images processed with the new filter (P < 10‐16). To evaluate the clinical usefulness of the digital filter, 2 individual observers graded 100 photographs from subjects with diabetes mellitus before and after application of newly developed digital filter. Grading was compared with FA findings. Results Correspondence of diabetic retinopathy grading from 2 independent observers were 0.36 and 0.29 before application of digital filter and 0.61 and 0.65 after application of digital filter(kappa value, respectively). Correspondence of presence of diabetic retinopathy from 2 independent observers were 0.69 and 0.69 before application of digital filter and 0.86 and 0.84 after application of digital filter(kappa value, respectively). Conclusions The hemoglobin absorbance reinforced the edges of retinal blood vessels, verifying that the new RGB filter can enhance the visualization of retinal vasculature and increase the accuracy of diagnosis.