Optical Coherence Tomography (OCT) is an imaging modality that is used extensively for ophthalmic diagnosis, near-histological visualization and quantification of retinal abnormalities such as cysts, exudates, retinal layer disorganization, etc. Intra-retinal cysts (IRCs) occur in several macular disorders such as, diabetic macular edema, retinal vascular disorders, age-related macular degeneration, and inflammatory disorders. Automated segmentation of IRCs poses challenges owing to variations in the acquisition system scan intensities, speckle noise, and imaging artifacts. Several segmentation methods have been proposed in the literature for IRC segmentation on vendor-specific OCT images that lack generalizability across imaging systems. In this work, we propose a fully convolutional network (FCN) model for vendor-independent IRC segmentation. The proposed method counteracts image noise variabilities and trains FCN models on OCT sub-images from the OPTIMA cyst segmentation challenge dataset (with four different vendor-specific images, namely, Cirrus, Nidek, Spectralis, and Topcon). Further, optimal data augmentation and model hyper-parametrization is shown to prevent over-fitting for IRC area segmentation. The proposed method is evaluated on the test data set with a recall/precision rate of 0.66/0.79 across imaging vendors. The Dice correlation coefficient of the proposed method outperforms that of the published algorithms in the OPTIMA cyst segmentation challenge with a Dice rate of 0.71 across the vendors.

No abstract

Anti-islanding protection schemes currently enforce the DGs to disconnect immediately for grid faults through loss of grid (LOG) protection system. This greatly reduces the benefits of DG deployment. For preventing disconnection of DGs during LOG, several islanding protection schemes are being developed. Their main objectives are to detect LOG and disconnect the DGs from the utility. This allows the DGs to operate as power islands suitable for maintaining uninterruptible power supply to critical loads. A major challenge for the islanding protection schemes is the protection co-ordination of distribution systems with bidirectional fault current flows. This is unlike the conventional overcurrent protection for radial systems with unidirectional fault current flow. This paper presents a comprehensive survey of various islanding protection schemes that are being developed, tested and validated through extensive research activities across the globe. Chui Fen Ten received the BEng degree from Universiti Teknologi Malaysia, Skudai, Malaysia, in 2005 and the MSc degree from University of Manchester (formerly UMIST), Manchester, UK, in 2006. Currently, she is working towards her PhD degree in the School of Electrical and Electronic Engineering of the University of Manchester (formerly UMIST), Manchester, UK. Prof. P.A.Crossley is Professor of Electrical Engineering at the University of Manchester. He graduated with a B.Sc degree from UMIST, United Kingdom in 1977 and a Ph.D. degree from the University of Cambridge, United Kingdom in 1983. He had been involved in the design and application of digital protection relays and systems for more than 25 years, first with GEC, then with ALSTOM and UMIST and the Queen's University of Belfast, United Kingdom and currently with the University of Manchester. Currently he is the Director of Joule Centre for Energy Research, University of Manchester, UK. He is an active member of various CIGRE, IEEE and IET committees on protection.