François Anton

Towards Web-based representation and processing of health information

BackgroundThere is great concern within health surveillance, on how to grapple with environmental degradation, rapid urbanization, population mobility and growth. The Internet has emerged as an efficient way to share health information, enabling users to access and understand data at their fingertips. Increasingly complex problems in the health field require increasingly sophisticated computer software, distributed computing power, and standardized data sharing. To address this need, Web-based mapping is now emerging as an important tool to enable health practitioners, policy makers, and the public to understand spatial health risks, population health trends and vulnerabilities. Today several web-based health applications generate dynamic maps; however, for people to fully interpret the maps they need data source description and the method used in the data analysis or statistical modeling. For the representation of health information through Web-mapping applications, there still lacks a standard format to accommodate all fixed (such as location) and variable (such as age, gender, health outcome, etc) indicators in the representation of health information. Furthermore, net-centric computing has not been adequately applied to support flexible health data processing and mapping online.ResultsThe authors of this study designed a HEalth Representation XML (HERXML) schema that consists of the semantic (e.g., health activity description, the data sources description, the statistical methodology used for analysis), geometric, and cartographical representations of health data. A case study has been carried on the development of web application and services within the Canadian Geospatial Data Infrastructure (CGDI) framework for community health programs of the New Brunswick Lung Association. This study facilitated the online processing, mapping and sharing of health information, with the use of HERXML and Open Geospatial Consortium (OGC) services. It brought a new solution in better health data representation and initial exploration of the Web-based processing of health information.ConclusionThe designed HERXML has been proven to be an appropriate solution in supporting the Web representation of health information. It can be used by health practitioners, policy makers, and the public in disease etiology, health planning, health resource management, health promotion and health education. The utilization of Web-based processing services in this study provides a flexible way for users to select and use certain processing functions for health data processing and mapping via the Web. This research provides easy access to geospatial and health data in understanding the trends of diseases, and promotes the growth and enrichment of the CGDI in the public health sector.

Optimization-based Fluid Simulation on Unstructured Meshes

We present a novel approach to fluid simulation, allowing us to take into account the surface energy in a precise manner. This new approach combines a novel, topology-adaptive approach to deformable interface tracking, called the deformable simplicial complexes method (DSC) with an optimization-based, linear finite element method for solving the incompressible Euler equations. The deformable simplicial complexes track the surface of the fluid: the fluid-air interface is represented explicitly as a piecewise linear surface which is a subset of tetrahedralization of the space, such that the interface can be also represented implicitly as a set of faces separating tetrahedra marked as inside from the ones marked as outside. This representation introduces insignificant and controllable numerical diffusion, allows robust topological adaptivity and provides both a volumetric finite element mesh for solving the fluid dynamics equations as well as direct access to the interface geometry data, making inclusion of a new surface energy term feasible. Furthermore, using an unstructured mesh makes it straightforward to handle curved solid boundaries and gives us a possibility to explore several fluid-solid interaction scenarios.

Tetrahedral Mesh Improvement Using Multi-face Retriangulation

In this paper we propose a simple technique for tetrahedral mesh improvement without inserting Steiner vertices, concentrating mainly on boundary conforming meshes. The algorithm makes local changes to the mesh to remove tetrahedra which are poor according to some quality criterion. While the algorithm is completely general with regard to quality criterion, we target improvement of the dihedral angle. The central idea in our algorithm is the introduction of a new local operation called multi-face retriangulation (MFRT) which supplements other known local operations. Like in many previous papers on tetrahedral mesh improvement, our algorithm makes local changes to the mesh to reduce an energy measure which reflects the quality criterion. The addition of our new local operation allows us to advance the mesh to a lower energy state in cases where no other local change would lead to a reduction. We also make use of the edge collapse operation in order to reduce the size of the mesh while improving its quality. With these operations, we demonstrate that it is possible to obtain a significantly greater improvement to the worst dihedral angles than using the operations from the previous works, while keeping the mesh complexity as low as possible.

LTE RF subsystem power consumption modeling

This paper presents a new power consumption emulation model, for all possible scenarios of the RF subsystem, when transmitting a LTE signal. The model takes the logical interface parameters, Tx power, carrier frequency and bandwidth between the baseband and RF subsystem as inputs to compute the power consumption. An analysis of modeling approaches was conducted and the modeling approach with the least sum of squared errors is used to compute the emulation model. The neural networks applying the Pseudo-Gauss Newton algorithm for optimization proved to have the least sum of squared errors. This approach was validated against a real life scenario with a relative error of 5.77%.

Multi-domain, higher order level set scheme for 3D image segmentation on the GPU

Level set method based segmentation provides an efficient tool for topological and geometrical shape handling. Conventional level set surfaces are only C0 continuous since the level set evolution involves linear interpolation to compute derivatives. Bajaj et al. present a higher order method to evaluate level set surfaces that are C2 continuous, but are slow due to high computational burden. In this paper, we provide a higher order GPU based solver for fast and efficient segmentation of large volumetric images. We also extend the higher order method to multi-domain segmentation. Our streaming solver is efficient in memory usage.

Decision support for flood event prediction and monitoring

In this paper the development of Web GIS based decision support system for flood events is presented. To improve flood prediction we developed the decision support system for flood prediction and monitoring that integrates hydrological modelling and CARIS GIS. We present the methodology for data integration, floodplain delineation, and online map interfaces. Our Web-based GIS model can dynamically display observed and predicted flood extents for decision makers and the general public. The users can access Web-based GIS that models current flood events and displays satellite imagery and digital elevation model integrated with flood plain area. The system can show how the flooding prediction based on the output from hydrological modeling for the next 48 hours along the lower Saint John River Valley.

Crisp Clustering Algorithm for 3D Geospatial Vector Data Quantization

In the next few years, 3D data is expected to be an intrinsic part of geospatial data. However, issues on 3D spatial data management are still in the research stage. One of the issues is performance deterioration during 3D data retrieval. Thus, a practical 3D index structure is required for efficient data constellation. Due to its reputation and simplicity, R-Tree has been received increasing attention for 3D geospatial database management. However, the transition of its structure from 2D to 3D had caused a serious overlapping among nodes. Overlapping nodes also occur during splitting operation of the overflown node N of M + 1 entry. Splitting operation is the most critical process of 3D R-Tree. The produced tree should satisfy the condition of minimal overlap and minimal volume coverage in addition with preserving a minimal tree height. Based on these concerns, in this paper, we proposed a crisp clustering algorithm for the construction of a 3D R-Tree. Several datasets are tested using the proposed method and the percentage of the overlapping parallelepipeds and volume coverage are computed and compared with the original R-Tree and other practical approaches. The experiments demonstrated in this research substantiated that the proposed crisp clustering is capable to preserve minimal overlap, coverage and tree height, which is advantageous for 3D geospatial data implementations. Another advantage of this approach is that the properties of this crisp clustering algorithm are analogous to the original R-Tree splitting procedure, which makes the implementation of this approach straightforward.

Spatial access method for urban geospatial database management: An efficient approach of 3D vector data clustering technique

In the last few years, 3D urban data and its information are rapidly increased due to the growth of urban area and urbanization phenomenon. These datasets are then maintain and manage in 3D spatial database system. However, performance deterioration is likely to happen due to the massiveness of 3D datasets. As a solution, 3D spatial index structure is used as a booster to increase the performance of data retrieval. In commercial database, commonly and widely used index structure for 3D spatial database is 3D R-Tree. This is due to its simplicity and promising method in handling spatial data. However, 3D R-Tree produces serious overlapping among nodes. The overlapping factor is important for an efficient 3D R-Tree to avoid replicated data entry in a different node. Thus, an efficient and reliable method is required to reduce the overlapping nodes in 3D R-Tree nodes. In this paper, we proposed a 3D geospatial data clustering to be used in the construction of 3D R-Tree and respectively could reduce the overlapping among nodes. The proposed method is tested on 3D urban dataset for the application of urban infill development. By using several cases of data updating operations such as building infill, building demolition and building modification, the proposed method indicates that the percentage of overlapping coverage among nodes is reduced compared with other existing approaches.

Fast streaming 3D level set segmentation on the GPU for smooth multi-phase segmentation

Level set method based segmentation provides an efficient tool for topological and geometrical shape handling, but it is slow due to high computational burden. In this work, we provide a framework for streaming computations on large volumetric images on the GPU. A streaming computational model allows processing large amounts of data with small memory footprint. Efficient transfer of data to and from the graphics hardware is performed via a memory manager. We show volumetric segmentation using a higher order, multi-phase level set method with speedups of the order of 5 times.

Homotopic Object Reconstruction Using Natural Neighbor Barycentric Coordinates

One of the challenging problems in computer vision is object reconstruction from cross sections. In this paper, we address the problem of 2D object reconstruction from arbitrary linear cross sections. This problem has not been much discussed in the literature, but holds great importance since it lifts the requirement of order within the cross sections in a reconstruction problem, consequently making the reconstruction problem harder. Our approach to the reconstruction is via continuous deformations of line intersections in the plane. We define Voronoi diagram based barycentric coordinates on the edges of n-sided convex polygons as the area stolen by any point inside a polygon from the Voronoi regions of each open oriented line segment bounding the polygon. These allow us to formulate homotopies on edges of the polygons from which the underlying object can be reconstructed. We provide results of the reconstruction including the necessary derivation of the gradient at polygon edges.