Sada Narayanappa

A Mobile Data Analysis Framework for Environmental Health Decision Support

Relations between negative health effects like asthma and lung cancer and elevated levels of the environmental factors, such as air pollution, tobacco smoke and humidity, have been detected in several large scale exposure studies. Thus, public health care and service systems require the ability to track, monitor, store, and analyze individual moving trajectories along with several environmental conditions the individual is exposed to in order to identify meaningful relationships among theses data and derive conclusions for environmental health decision support. With continued advances in information technology, patients can be monitored with numerous intelligent devices. Sensors can be integrated into their mobile devices such as smart phones for continuous health assistance and disease attack prevention. However, researchers must overcome many challenges, such as data acquisition, data scales and data uncertainty, in order to develop a real-time health monitoring system. In this paper, we propose a system framework for modeling and analyzing individual exposure to environmental triggers of asthma attacks. The proposed system can provide a tool to develop more accurate asthma prevention and care plans enabled by real-time patient monitoring and communication through alerts for potential environmental triggers.

Web data retrieval: solving spatial range queries using k-nearest neighbor searches

As Geographic Information Systems (GIS) technologies have evolved, more and more GIS applications and geospatial data are available on the web. Spatial objects in a given query range can be retrieved using spatial range query − one of the most widely used query types in GIS and spatial databases. However, it can be challenging to retrieve these data from various web applications where access to the data is only possible through restrictive web interfaces that support certain types of queries. A typical scenario is the existence of numerous business web sites that provide their branch locations through a limited “nearest location” web interface. For example, a chain restaurant’s web site such as McDonalds can be queried to find some of the closest locations of its branches to the user’s home address. However, even though the site has the location data of all restaurants in, for example, the state of California, it is difficult to retrieve the entire data set efficiently due to its restrictive web interface. Considering that k-Nearest Neighbor (k-NN) search is one of the most popular web interfaces in accessing spatial data on the web, this paper investigates the problem of retrieving geospatial data from the web for a given spatial range query using only k-NN searches. Based on the classification of k-NN interfaces on the web, we propose a set of range query algorithms to completely cover the rectangular shape of the query range (completeness) while minimizing the number of k-NN searches as possible (efficiency). We evaluated the efficiency of the proposed algorithms through statistical analysis and empirical experiments using both synthetic and real data sets.

Optimizing map labeling of point features based on an onion peeling approach

Map labeling of point features is the problem of placing text labels to correspond- ing point features on a map in a way that minimizes overlaps while satisfying basic rules for the quality. This is a critical problem in the application of cartography and geographical information systems (GIS). In this paper we study the fundamental issues related to map labeling of point features and develop a new genetic algorithm to solve this problem. We adopt a method called convex onion peeling and utilize it in our proposed convex onion peeling genetic algorithm (COPGA )t o efficiently manage map labels of point features. The proposed algorithm takes advantage of a convex onion peeling structure to achieve better map label initialization and to enhance the evolutionary process. The performance of the proposed algorithm was evaluated through extensive experiments on both synthetic and real datasets. In experiments with an implementation of our algorithm using OpenMap, the results show that our genetic algorithm, based on convex onion peeling, is an efficient, robust, and extensible algorithm for automated map labeling of point features.

MobiS: a distributed paradigm of mobile sensor data analytics for evaluating environmental exposures

Continued advances and cost reduction in personal mobile devices such as smart phones made them widely used in daily-life practices. Mobile devices can be integrated with a growing set of cheap powerful embedded sensors that enable the emergence of mobile sensing applications, including healthcare, environmental monitoring and transportation. As the size of the sensor data continuously grows, managing the data becomes increasingly difficult using traditional database systems. This paper proposes a new framework for large-scale continuously changing mobile sensor data analysis. We discuss the emerging environmental sensing paradigms and opportunities to apply HBase and MapReduce for managing multiple sensor data in the environmental exposome domain. Moreover, we provide an architectural framework and present a concrete use case with a set of data models, spatio-temporal queries, and MapReduce functions.

An interactive framework for spatial joins: a statistical approach to data analysis in GIS

Many Geographic Information Systems (GIS) handle a large volume of geospatial data. Spatial joins over two or more geospatial datasets are very common operations in GIS for data analysis and decision support. However, evaluating spatial joins can be very time intensive due to the size of datasets. In this paper, we propose an interactive framework that provides faster approximate answers of spatial joins. The proposed framework utilizes two statistical methods: probabilistic join and sampling based join. The probabilistic join method provides speedup of two orders of magnitude with no correctness guarantee, while the sampling based method provides an order of magnitude improvement over the full indexing tree joins of datasets and also provides running confidence intervals. The framework allows users to trade-off speed versus bounded accuracy, hence it provides truly interactive data exploration. The two methods are evaluated empirically with real and synthetic datasets.

Convex onion peeling genetic algorithm: an efficient solution to map labeling of point-feature

Map labeling of point-feature is the problem of placing text labels to corresponding point features on a map in a way that minimizes overlaps while satisfying basic rules for the quality. This problem is a critical problem in the applications of cartography and Geographical Information Systems (GIS). In this paper we study the fundamental issues related to map labeling of point-feature and develop a new genetic algorithm to solve this problem. We adopt a data structure called convex onion peeling and utilize it in our proposed Convex Onion Peeling Genetic Algorithm (COPGA) to efficiently manage point features. We evaluated the performance of the proposed algorithm through extensive experiments on both synthetic and real datasets. The experimental results show that our genetic algorithm based on the convex onion peeling structure is an efficient, robust and extensible algorithm for automated map labeling of point-feature.

Ultra wideband indoor positioning system in support of emergency evacuation

There has been a research focus recently on developing indoor positioning systems (IPS) that can support tracking and evacuation in case of emergency. Real-time positioning, target tracking and mobile agent communication are important operations in evacuation planning and management in non-open environments such as indoor, urban and underground areas. However, these are challenging tasks since the performance of GPS and other conventional sensing technologies for positioning and target tracking are either limited or even non-existent due to serious effects of radio signal fading and multiple path interference. This paper investigates and proposes an effective methodology using ultra wideband (UWB) technology to develop a robust and self-organized mobile ad hoc sensor network. Such network is able to provide integrated capability of real-time accurate position tracking and voice/data communications in support of emergency rescue operations in critical complex environments. We present a prototype of UWB IPS for real-time positioning, target tracking and communication support. The system design and installation are presented in detail and the performance of the system is evaluated through test-beds based on some scenarios.

Voronoi maps: an approach to individual-based environmental exposure estimation

Estimating an individuals environmental exposure is a complicated problem that depends on the amount of time of the individuals exposure, the uncertain location of the individual, and the uncertainty in the levels of environmental factors based on available localized measurements. This problem is critical in the applications of environmental science and public health. In this paper we study the fundamental issues related to spatio-temporal uncertainty of human trajectories and environmental measurements and define a model of exposure uncertainty. We adopt a geometric data structure called the Voronoi diagram to interpolate environmental data, and utilize it in our proposed method to efficiently solve this problem. We evaluate the performance of the proposed method through experiments on both synthetic and real road networks. The experimental results show that our solution based on probabilistic routing aggregation is an efficient and extensible method for environmental exposure time estimation.

SCHAS: A Visual Evaluation Framework for Mobile Data Analysis of Individual Exposure to Environmental Risk Factors

Exposure to environmental risk factors as well as weather conditions are known to have negative effects on health. Until recently, there was little a society could do for an individual at risk, other than provide general warnings when the concentration of pollutants or weather conditions deviate from the norm. Similarly, the assessment of individuals’ exposure over time has been confined to population and geographic averages, rather than individualized estimates. Recent advances in sensors and mobile technology have enabled real-time measurements of environmental variables and, at the same time, provided information about the spatio-temporal behavior of individuals. This can dramatically change the way health and wellness are assessed as well as how care and treatment are delivered. This paper presents a system framework called “Smart and Connected Health Alert System (SCHAS)” for individual-level environmental exposure in an attempt to better understand the relationships among exposures, symptoms and human health conditions. We demonstrate user interface, data acquisition and visual evaluation tools for large mobile sensor data analysis.

Steel threads: Software engineering constructs for defining, designing and developing software system architecture

A steel thread is a software engineering construct that identifies the most important execution paths, including software and hardware elements, through a computer system, while meeting business objectives and demonstrating executable architecture. Steel threads are often used in the context of defining software system architecture. Although there have been references to steel threads in software engineering literature, it is hard to find clear definitions and usage of steel threads in the software industry or among the research community. This paper provides an overview of steel threads in software architecture design and development as well as presenting the contexts of steel threads. In addition, we show how to identify important scenarios and execution paths to construct steel threads and discuss the contexts under which steel threads are applicable in the software development life cycle. We also discuss the roles of steel threads in system development as well as their usability and applicability. Finally, a case study of the use of steel threads in a software system is presented.