Active Contour Level Set Segmentation for Leakage Detection in Retinal Diseases
The detection and assessment of leakage in retinal fluorescein angiogram images is important for the management of a wide range of retinal diseases. This paper is concerned with a new saliency-based method for detection of leakage in FA image. After saliency detection, we detect the leakage detected region in FA image. We have developed a framework that can automatically detect leakage . This framework comprises two step: saliency detection and leakage detection. First superpixel segmentation (SLIC) done, divided images into meaningful patches at various levels. We have done superpixel segmentation based on two cues-namely; intensity and compactness. Saliency maps are finally fused based on intensity and compactness. Finally, leaking regions are detected using graph-cut segmentation. The proposed method validated on two retinal disease like malarial retinopathy and diabetic retinopathy. Leaking regions finally segmented using active –contour level set segmentation. Finally performance parameters like accuracy, sensitivity, specificity, AUC and DC are calculated for proposed method.
Keywords - AUC,DC,SLIC and FA image