University of Virginia
Ellen J. Bass

Research

Current Sponsored Research

Title: Collaborative Adaptive Sensing of the Atmosphere (CASA) (D. McLaughlin, PI; Bass UVa PI: 9/01/04-8/31/13)
Sponsor: National Science Foundation (subcontract from UMass Amherst)

The Center for Collaborative Adaptive Sensing of the Atmosphere (CASA) is developing networks of low-power, low-cost radars that adaptively collect, process and visualize high resolution data in the lowest portion of the atmosphere. These technological advances could enhance the warning process by providing higher resolution radar data that sense closer to the ground, have greater spatial resolution, and have faster update rates. We are investigating the impact of increased spatio-temporal resolution weather radar data on forecaster accuracy of predictions of weather hazards. We are also assessing the impact of these radar data on forecaster confidence and on warning decisions. We are also working with emergency managers in order to ensure that the network concept is designed with emergency managers’ needs in mind. Interviews, surveys, analysis of product usage logs, and simulated scenarios are being used to solicit EM input. Results indicate the need for products for both high and low bandwidth end users, visualizations for velocity products that are more easily interpreted, and enhanced training. We are developing interventions to address these needs.

Title: National Library of Medicine T-15 Training Grant: A Systems Engineering Focus on Medical Informatics (S. Guerlain, PI)
Sponsor: National Library of Medicine

Please go to the training grant website for more information.

Title: Supporting Hand-Off of Care: Sign-out Support Tools, Process and Evaluation (E. Bass, PI; 2/1/09-1/13/11)
Sponsor: Institute for Quality for Patient Safety

Hand-off of care is the transfer of information, responsibility, and/or authority from one set of caregivers to another. The proposed work aims to improve the hand-off of care process across the health center by extending the current sign-out process and tool development efforts beyond Pediatrics. By collaborating with Medicine and Surgery, this proposed work will lead to the development of enhanced sign-out tools and procedures by focusing on 8 interacting goals:

Title: Sign-out Training Development and Evaluation (L. Waggoner-Fountain, PI; 11/1/08-10/31/10)
Sponsor: Graduate Medical Education

The ability to give and receive sign-out is generally learned informally by more senior providers giving sign-out themselves. Despite the critical importance of resident sign-out in assuring safe and effective patient care, very little research has examined the content, communication, and procedures of resident sign-out. Moreover, even less is known about how sign-out related interventions (such as tools and training) improve the quality of sign-out, or how these interventions influence patient safety and the quality of patient care. This is particularly true in cognitive-based medical specialties where physicians spend much time examining and treating patients as opposed to performing procedures. Few training programs formally teach resident physicians effective communication strategies for sign-out (or the information that should be communicated) and even fewer programs assess residents’ abilities to apply these strategies during sign-out. Current measures of assessment have focused primarily on residents’ opinion and general observation, and have not involved assessment of learning or behavioral changes that result from training specific sign-out strategies. This research involves developing and evaluating educational programs to train residents how to communicate effectively at sign-out.

Title: Formal Methods to Inform Human-Computer Interface Design for Ground-based Spacecraft Logistical Planning Automation (E. Bass, PI; 9/01/06-8/31/10)
Sponsor: NASA Langley Research Center

Modern, safety-critical systems are inherently complex, with multiple sub-systems and people interacting with each other in order to achieve multiple, and sometimes conflicting, goals. While the majority of the sub-systems (including the interfaces human’s use to interact with the system) are well engineered, large scale systems failures still occur. Such failures are difficult to predict (and therefore design against) because they are emergent features of the complex interactions that occur within the system. We are currently investigating the use of techniques from human-operator modeling and formal methods in order to predict sequences of human error capable of producing complex system failure. By understanding when such errors are capable of occurring, engineers will be able to guard against them in their designs.

Title: Event Prediction and Mitigation (E. Bass, PI; 5/01/09-4/30/10)
Sponsor: Center for Operator Performance at Wright State University

Despite advances in safety interlock systems and instrumentation, situations still arise that operators are unable to accurately assess and result in unplanned equipment outages, lost productivity, and unsafe conditions. What is needed are better techniques to illustrate the state of the equipment and process and to provide advanced warning of pending events to the operator. The effort will determine how to identify key indicators, how to present these indicators to operators, and how to incorporate decision aids to assist the operator.

Title:Multi-scale Tools for Airspace Modeling and Design (S. Patek, PI; 3/01/08-5/28/10)
Sponsor: NASA Langley Research Center

Congestion around airports in the National Airspace System (NAS) contributed substantially to the billions of hours of delay experienced by domestic U.S. air passengers, and the Federal Aviation Administration (FAA) projects that the number of congested airports within the NAS will only grow. Strategic long-term measures, such as slot controls and congestion pricing, have been employed at NAS airports to reduce or prevent congestion in advance of operation, but their use has not been widespread nor has there been a consensus about their effectiveness. In the operations research literature, network flow models have been used largely to model and support decision-making for short-term congestion management in the NAS. This research employs such a network flow model for modeling and evaluating strategic, long-term management of air traffic and congestion, specifically the use of slot controls, and seeks to determine the utility of these models towards that purpose.

Title: Celebratory Drinking Prevention Program (S. Bruce, PI; 8/1/08-6/30/10)
Sponsor: Virginia Department of Alcoholic Beverage Control

Celebratory drinking among college students often occurs at predictable times in the academic calendar. Research on student drinking at the University of Virginia (U.Va.), indicates there are three distinct, annual events during which a greater than average percentage of students engage in hazardous drinking. These events are Halloween, the weekend of the last home football game, and the spring Foxfield horse races. This research seeks to use collaborative peer leadership to enhance existing educational efforts to reduce hazardous drinking and the accompanying adverse consequences experienced by students during these events, including the second hand effects of drinking.

Prior Sponsored Research

Title: Simplify Dynamic Tailoring (D. Brown, PI; 6/01/08-8/30/09)
Sponsor: National Technical Alliance

Today’s world of vizualization engines, coupled with customized and visually-based end-user applications, have been used by industry to create rich situational awareness enviromnents. These environments support decisonmakers who must comprehend complex dynamic relationships in their mission space and who must respond with rapid, effective decision making to shape that mission space and assure their mission success. An unfortunate reality is that most fielded and developmental systems provide complex graphical user interface (GUI) mechanisms corresponding to the complexity of the modeled world. These systems are highly specialized and require users to gain expertise in a very narrow field of endeavor. We are developing a scalable research prototype system based upon Google Earth that allows users to access multiple data streams, including the substrate (mapping data) and one or more overlays of KML-based object data. The users can then save these views as profiles and share them with others users. These users can then easily adapt the profiles to explore and use similar data streams. These profiles will act as customized applications reducing the overall cost in specialized software and task specific training. Additionally this system will enhance analyst tradecraft by allowing users to shaire their tools and techniques.

Title: Resident Sign-Out - A Precarious Exchange of Critical Information in a Fast Paced World (S. Borowitz, PI: 6/1/05-5/30/07)
Sponsor: Graduate Medical Education

Sign-out is a mechanism of transferring information, responsibility, and/or authority from one set of caregivers to another. The primary objective of any patient sign-out is the accurate transfer of information about a patient’s state and plan of care. This transfer of clinical information during sign-out is crucial for patient safety and successful care. This research will improve the gathering of essential and accurate information about patients. This research will investigate the use of information technology to manage information, support resident self-education (web based daily note) and facilitate the learning of students and other health care professionals (cooperative development of daily progress note using computer-based software). This research will promote working effectively with others as a member or the health care team (both sign out lists and timely, useful daily progress notes). This research will promote the practice of cost-effective health care and resource allocation that does not compromise quality of care (accurate information exchange may decrease duplicative studies), advocate high quality patient care with system complexities (accurate and up to date patient information will be conveyed in meaningful progress notes) and facilitate cooperation with other health care providers and managers to assess, coordinate and improve health care for patients (accurate daily note).

Title: Terrain awareness in synthetic vision displays (E. Bass, PI: 6/17/04-9/30/06)
Sponsor: NASA Langley Research Center

Controlled flight into terrain (CFIT), a condition where a functional aircraft is flown into the ground, constitutes to fatal accidents in global commercial aviation. It is generally associated with a loss of situation awareness in low level flight and low visibility conditions. A Synthetic Vision System (SVS) is a cockpit display technology that combats this problem by creating a synthetic, clear-day view of the world surrounding the aircraft. This research introduces new measures of spatial awareness to determine how SVS display parameters (Field of View (FOV) and Texture) and the relative location of terrain affect a pilot’s ability to perceive the relative spatial and temporal location of a terrain point.

Title: Wake Encounters Investigation (E. Bass, PI: 9/07/06-9/25/08)
Sponsor: NASA Langley Research Center

Redefining the current aircraft separation standards could increase airspace capacity while maintaining safety by defining an acceptable wake vortex encounter based on aircraft controllability and pilot acceptance of aircraft response. This work implements an aircraft wake encounter analysis framework that can run on a standard workstation. Two analyses are conducted as a proof-of-concept using a Boeing 757-200 follower aircraft. The first analysis examines parameters that influence follower aircraft response. The second examines the impact of the weight of the generator on the follower aircraft response. The Burnham-Hallock wake flow-field model as adapted by Tatnall defines the wake of a generator aircraft. For the follower aircraft’s response, strip theory calculates the wake-induced forces and moments acting on the aircraft. For the follower aircraft-centered wake encounter hazard metrics, this research considers maximum bank angle upset, maximum roll acceleration, and vertical and horizontal flightpath deviations. This work demonstrates the flexibility of the framework’s software implementation as its ultimate purpose is to aid researchers studying wake vortex encounters.

Title: Evaluating measures of human-automation interaction in pilot self-separation tasks (E. Bass, PI; 11/03/03-8/30/05)
Sponsor: NASA Langley Research Center

A variety of decision support tools have been designed, not only to call a person’s attention to a problem, but also to make a recommendation on a preferred course of action. In aviation, human-automation interaction problems frequently surface in what have been called “nonconformance” problems and errors as pilots do not slavishly follow the recommendations of their tools. This research involves developing a simulated pilot model that can be integrated with traffic conflict prediction automation in order to investigate non-conformance issues.

Title: Flight dispatcher icing training (E. Bass, PI; 11/03/03-5/31/06)
Sponsor: NASA Glenn Research Center

In the airline industry, flight dispatchers are responsible for planning aircraft routes safely and economically and are co-responsible with the pilot in command for the safe operation of each flight they are assigned. One aspect of weather that is of particular concern to flight dispatchers is atmospheric icing. Training is essential to help flight dispatchers understand when and where icing may occur and what to do when it does. Currently, icing training products developed specifically for flight dispatchers do not exist; thus instructors try to adapt pilot training materials for flight dispatchers. While airlines could design completely separate icing training programs for each role including flight dispatchers, this practice would not be efficient. Therefore, airlines would benefit from the knowledge of how to create icing training that is beneficial across organizational roles without compromising role-specific applicability. The goal of this project is to create and empirically study methods for the development of re-usable training.

Title: Human-automation interaction methodologies for personal air vehicle systems flown by pilots with varying skills (E. Bass, PI; 2/01/03-6/30/05)
Sponsor: National Institute of Aerospace

Systems whose function includes making assessments of the environment through observation are becoming increasingly capable and common. Trust is a key feature in a successful relationship between a human operator and such information analysis automation. The relationship between trust and automation usage is not clearly understood. Many researchers use subjective measures to quantify trust. Behavioral measures of trust along with subjective measures would aid in the design of automation. This research aims to develop behavioral measures of trust and to compare them to subjective measures in the hopes that either type of measure could be used to help evaluate human-automation interaction.

Title: Network modeling to support operational concept development for on-demand air transportation (E. Bass, PI; 11/1/03-6/30/05)
Sponsor: NASA Langley Research Center

As a potential addition to scheduled air carrier, hub-and-spoke service, the notion of on-demand flight service by small aircraft to small community airports presents an attractive method to increase air transportation capacity. The development of a safe, efficient, reliable and affordable, on-demand air transportation capability requires the ability to not only serve the demand but also provide a profit to the on-demand service suppliers (i.e., the operators). To gain a more dynamic and therefore more realistic view of the situation, the long term goal is to combine agent-based simulation and dynamic programming techniques to better understand and therefore validate the efficacy of such on-demand capabilities. Our near-term goal is to work toward developing the modeling framework and validating network models that can capture airspace utilization issues in broad strokes and facilitate the design and optimization of economic mechanisms for allocating access to airspace for the on-demand system. Such models can be used to gain a basic understanding of when and where the demand for airspace will emerge and how the demand will be accommodated (or fail to be accommodate) based on the resources that are available.

Title: Cell Migration Consortium (W. Pearson, PI; E. Bass participation: 1/30/03-12/30/04)
Sponsor: National Institute of General Medical Science

Protein domain databases are indispensable tools in modern biology. They are used to investigate uncharacterized proteins via comparison to related proteins with known structural and functional information. The different domain resources employ their own identification methodologies, each with its own strengths and weaknesses. Domain databases vary in the number and breadth of entries, the statistical accuracy of their entries, and the frequency with which updates are released. The goal of this project is to develop a single protein domain tool that combines existing protein domain resources and depicts to the user the different types of uncertainty associated with any search results.

Title: Comparison of Protein Sequences (W. Pearson, PI; E. Bass participation: 1/30/03-6/30/05)
Sponsor: National Library of Medicine

Unlike tools in traditional engineering domains, those supporting scientific discovery not aim to help find a single, best solution, or test a particular hypothesis. Such tools enhance understanding of a problem domain even when the underlying models are flawed and imperfect. Such tools must therefore integrate human judgment and expertise with algorithmic methods in novel ways that improve domain understanding. In this research, we explore how visualizations of near-optimal protein sequence alignments from detailed and overview perspectives can be extended by adding detail and context information to assist in comprehension of the solution space. By providing alternative perspectives on sets of solutions we hope to provide insight into the problem domain that was heretofore unavailable.

Title: Project to Understand Biologic Systems, Learn to Integrate, Interpret and Communicate Their Findings (PUBLIIC). (W. Knaus, PI; E. Bass participation: 6/30/03-6/30/04)
Sponsor: Fund for Excellence in Science and Technology

Title: Tobacco Digital Patient Record- Human Gene Expression project. (W. Knaus, PI; E. Bass participation: 10/1/02-6/30/04)
Sponsor: Virginia Tobacco Settlement Fund