Abstract:
Hypertension and diabetes are diagnosed and monitored using retinal vasculature clues. Such conditions involve the microvascular system, which can only be seen in the retina. Recent advances in retinal imaging and computer vision have renewed interest in the objective assessment of retinal vessels as a surrogate biomarker for systemic vascular diseases. We present RAVIR, a novel dataset for semantic segmentation of Retinal Arteries and Veins in IR imaging. It allows deep learning-based models to identify extracted vessel type without post-processing. SegRAVIR, a deep learning-based method, semantically segments retinal arteries and veins and quantitatively measures their widths. SegRAVIR outperforms leading models in our extensive experiments. We propose a knowledge distillation framework to domain adapt RAVIR pretrained networks on color images. Our pretraining yields new state-of-the-art benchmarks on the DRIVE, STARE, and CHASE_DB1 datasets.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
