A new approach for sensitivity improvement of retinal blood vessel segmentation in high-resolution fundus images based on phase stretch transform

• Main Authors: Firdausy, Kartika, Wahyunggoro, Oyas, Nugroho, Hanung Adi, Sasongko, Muhammad Bayu
• Format: Article PeerReviewed
• Language: English
• Published: Universitas Ahmad Dahlan 2022
• Subjects: Electrical and Electronic Engineering; Engineering
• Online Access: https://repository.ugm.ac.id/278619/1/Wahyunggoro_TK.pdf; https://repository.ugm.ac.id/278619/; http://ijain.org/index.php/IJAIN/article/view/914; https://doi.org/10.26555/ijain.v8i3.914
• Institution: Universitas Gadjah Mada

The eye-fundus photograph is widely used for eye examinations. Accurate identification of retinal blood vessels could reveal information that is helpful for clinical diagnoses of many health disorders. Although many researchers have proposed different methods to segment images of retinal blood vessels, the sensitivity is plausible to be improved. The algorithm's sensitivity refers to the algorithm's ability to identify retinal vessel pixels correctly. Furthermore, the resolution and quality of retinal images are improving rapidly. Consequently, new segmentation methods are in demand to overcome issues from high-resolution images. This study presented improved performance of retinal vessel segmentation using a novel edge detection scheme based on the phase stretch transform (PST) function as its kernel. Before applying the edge detection stage, the input retinal images were preprocessed. During the preprocessing step, non-local means filtering on the green channel image, followed by contrast limited adaptive histogram equalization (CLAHE) and median filtering, were applied to enhance the retinal image. After applying the edge detection stage, the post-processing steps (CLAHE, median filtering, thresholding, and morphological operations) were implemented to obtain the segmented image. The proposed method was evaluated using images from the high-resolution fundus (HRF) public database and yielded promising results for sensitivity improvement of retinal blood vessel detection. The proposed approach contributes to a better segmentation performance with an average sensitivity of 0.813, representing a clear improvement compared to several benchmark techniques.