Topology-Aware Retinal Artery–Vein Classification via Deep Vascular Connectivity Prediction

Retinal artery–vein (AV) classification is a prerequisite for quantitative analysis of retinal vessels, which provides a biomarker for neurologic, cardiac, and systemic diseases, as well as ocular diseases. Although convolutional neural networks have presented remarkable performance on AV classification, it often comes with a topological error, like an abrupt class flipping on the same vessel segment or a weakness for thin vessels due to their indistinct appearances. In this paper, we present a new method for AV classification where the underlying vessel topology is estimated to give consistent prediction along the actual vessel structure. We cast the vessel topology estimation as iterative vascular connectivity prediction, which is implemented as deep-learning-based pairwise classification. In consequence, a whole vessel graph is separated into sub-trees, and each of them is classified as an artery or vein in whole via a voting scheme. The effectiveness and efficiency of the proposed method is validated by conducting experiments on two retinal image datasets acquired using different imaging techniques called DRIVE and IOSTAR.

Development of Quality Control Tools for ARCMAP using Python

Topological error free data in the database is very essential for Geographic Information System (GIS) analysis. To minimize some of the most common GIS errors, five Quality Assessment/Quality Control tools has indigenously discussed in this paper, namely Auto registration tool, Dangles correction tool, Irrelevant points removal Tool, Delete polygon less than 4 vertices and Connect and Split using the programming language Python. Tools are developed for ArcMap environment for more effective data cleaning and validation. ArcMap is one of the main components in Architecture Geographical Information system (ArcGIS) suite. These tools will be helpful to identify the possible errors in a particular sheet and will navigate user to the error positions. These tools not only finds the errors they also provides several solutions to solve those errors in automatic or semi-automatic mode, hence making the data best suited for use in GIS. This paper describes the brief details of all the 5 developed tools, and also demonstrates the benefits of applying these tools.

Graphical algorithms and threshold error rates for the 2d color code

Recent work on fault-tolerant quantum computation making use of topological error correction shows great potential, with the 2d surface code possessing a threshold error rate approaching 1\%. However, the 2d surface code requires the use of a complex state distillation procedure to achieve universal quantum computation. The color code of is a related scheme partially solving the problem, providing a means to perform all Clifford group gates transversally. We review the color code and its error correcting methodology, discussing one approximate technique based on graph matching. We derive an analytic lower bound to the threshold error rate of 6.25\% under error-free syndrome extraction, while numerical simulations indicate it may be as high as 13.3\%. Inclusion of faulty syndrome extraction circuits drops the threshold to approximately 0.10 \pm 0.01\%.