Radon transform technique for linear structures detection: Application to vessel detection in fluorescein angiography fundus images

In medical images linear patterns such as blood vessels are important structures for computer-aided diagnosis and follow-up of many diseases. Moreover, image processing techniques are required to extract suitable information about vascular tree and its alteration. Analyzing of retinal blood vessel is critical work for the investigation of some diseases. In this study, we present an automated method for detecting retinal vasculatures based upon Radon transform. In preprocessing, we used top-hat transformation and averaging filter. Our main processing was included applying Radon transform, vessel certifying, and vessel refinement. Comparing the results of our method with gold standard showed that our results have more than 93% for true positive rate. In conclusion, it is possible to use Radon transform for vessel segmentation in fluorescein angiography fundus images, with acceptable sensitivity and specificity, as a necessary step in some diagnostic algorithm for retinal pathology.