Multi-level deep neural network for efficient segmentation of blood vessels in fundus images

• Published in: Electronics letters Vol. 53; no. 16; pp. 1096 - 1098
• Main Authors: Ngo, L, Han, J.-H
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 03.08.2017
• Subjects: age‐related macular degeneration; biomedical optical imaging; Biomedical technology; blood vessel segmentation; blood vessels; diabetic retinopathy; eye; image segmentation; medical image processing; multilevel deep neural network; neural nets; resizing technique; retinal fundus camera images; eye; retinal fundus camera images; resizing technique; image segmentation; blood vessels; multilevel deep neural network; diabetic retinopathy; age-related macular degeneration; neural nets; medical image processing; blood vessel segmentation; biomedical optical imaging
• ISSN: 0013-5194; 1350-911X; 1350-911X
• DOI: 10.1049/el.2017.2066

The exact blood vessel trees segmented from fundus images provide important information required for screening and following-up of diabetic retinopathy and age-related macular degeneration. The trained deep neural network presents an automated prediction of the blood vessels in retinal fundus camera images in the publicly DRIVE database with accuracy up to 0.9533 and area under the receiver operating characteristic curve up to 0.9752, which is better than manual recognition by expert human eyes. A resizing technique is introduced and applied to the multi-level network combining dropout and spatial-dropout layers to obtain more generalised training. The proposed model has the potential for the classification of other types of images.