Achievable networking performance of applications in a supercomputer depends on the exact combination of the communication patterns of the applications and the routing algorithms used by the supercomputer. In order to achieve the highest networking performance for the applications, the routing algorithms need to be designed optimally for those communication patterns. However, while communication patterns usually vary from application to application and even from phase to phase in an application, routing algorithms have limited variation and usually are optimized for typical communication patterns. This results in high networking performance for some communication patterns. In this paper, we present approaches for improving communication performance by using multiple paths and re-balancing load on physical links on the Blue Gene/Q supercomputer. We realize our approaches in a framework called OPTIQ and demonstrate the efficacy of our framework via a set of benchmarks. Our results show that we can achieve 43 -- 67% higher throughput on average from 91 experiments, and can achieve higher throughput than default MPI_Alltoallv used for certain communication patterns.

Improving Communication Throughput by Multipath Load Balancing on Blue Gene/Q

The amount of data that we deal with, on a daily basis, is ever increasing and coupled with the fact that display technology is getting cheaper by the day, this has led to the rise of large/scalable resolution display systems. With more and more people adopting large/scalable resolution display systems in their work places, these display systems are being used for data vizualization as well as collaboration. 
The complexity of data being visualized and the nature of collaborative tasks being carried out using large/scalable resolution displays are changing the way we use these displays. It is no longer enough to passively look at data on these displays. There is a growing need to actively engage and interact with the content. As such, large/scalable resolution display systems that handle large amounts of data need to provide adequate functionality to interact with this data. SAGE2 comes as an answer to this. It allows scalable displays to effectively act as platforms for data visualization and at the same time gives these systems the power of multiuser interaction. I propose two textual annotation tools to annotate documents and images in SAGE2. These annotation tools will further the capabilities of SAGE2 in terms of allowing users to collaborate. Through this thesis, I will evaluate these tools in the hopes of gaining insight about their usability, affordances, and limitations.

SAGE2 annotation with marker and note highlighted with pointer color.

Textual Annotation Tools for SAGE2

Every day, travelers use Public Transport Systems to reach their destinations. Public Transport Authorities generally use passive way-finding devices, including ambient displays, to provide useful information for travelers, such as the number of vehicles in the vicinity of a stop and their estimated times of arrival. However, due to both the complexity of the transport networks and the lack of sophistication in the design of these displays, the information provided by these devices is limited. We present TransitTrace, a visualization that exploits interaction-free ambient displays to provide travelers with more detailed information and ultimately to help them to navigate the city using a Public Transport System. Specifically, the proposed design makes use of a novel animation strategy to aid travelers in route planning tasks. In this paper, we describe details about the system and visualization design of TransitTrace, as well as its initial implementation using transportation data provided by the City of Chicago. 
Categories and Subject Descriptors 
H.2.8 [Database Applications]: Spatial databases and GIS; H.5.1 [Multimedia Information Systems]: Animations 
Keywords 
Public Transport Systems, Geotemporal Visualization, Ambient Displays

The animation begins by showing the current position of all vehicles that will be arriving in the users&rsquo; vicinity within a given window of time.

TransitTrace: Route Planning using Ambient Displays

Accessibility is an important element in urban transportation planning. Accessibility measures combine mobility and land use measures to provide a more complete picture of the transportation-land use nexus than either of these measures alone. By providing insights into the varying degrees to which different areas of a region are connected to opportunities by the transportation system, accessibility analysis helps urban planners to understand the relationship between transportation and land use, and provides reference for them to improve the equality of the residents. Calculating accurate accessibility values and visualizing them in an efficient way is a complex and challenging process. In this paper, we present a web-based system that visualizes multimodal
accessibility to multiple land uses of Chicago metropolitan area, as the first step of an effort to build an integrated platform for accessibility analysis tasks. We also discuss some use cases of this system, and show its effectiveness by providing experts feedback of this prototype. 
Categories and Subject Descriptors 
I.3.8 [Computer Graphics]: Applications; H.5.2 [Information Interfaces and Presentation]: User Interfaces|graphical user interfaces 
General Terms 
Design 
Keywords 
accessibility analysis, geographic visualization, urban transportation planning

Travel Time View. Users can investigate the travel time from any block group to other block groups showed as an isochrone map.

Understanding Transportation Accessibility of Metropolitan Chicago Through Interactive Visualization

A natural language interface to visual data exploration would allow a user to directly specify questions through speech, allowing the user to focus on higher-order tasks, such as hypothesis generation and question formulation. 
However, visual data exploration involves repeated cycles of visualization construction and interaction, as well as reasoning across many visualizations generated over the course of an exploratory session. A &rsquo;conversational interface&rsquo;, which maintains a dialog with the user through natural language and gestures, could support these complex tasks. 
We conducted an observational, exploratory study to observe the interaction between a subject and a remote data analysis expert (DAE) who assists the subject in an exploratory data analysis task.

Research subject involved in an exploratory data analysis task using the EVL Cyber-Commons tiled display wall.

&ldquo;Show me data&rdquo; Observational study of a conversational interface in visual data exploration

We present the design and implementation of an immersive visual mining and analysis tool for cosmological data. The tool consists of an immersive linked multiview display which allows domain experts to interact with visual representations of spatial and nonspatial cosmology data. Nonspatial data is represented as time-aligned merger trees, and through a pixel-based heatmap. Spatial data is represented through GPU-accelerated point clouds and geometric primitives. The user can select a halo and visualize a 3D representation of the raw particles, as well as of the halos at the particular time stamp. We have demonstrated the tool to a senior staff member of the Adler Planetarium and report their feedback. The tool can assist researchers in the interaction navigation and mining of large scale cosmological simulation data. 
Keywords: Three-dimensional displays; Data visualization; Data mining; Spatial databases 
doi: 10.1109/SciVis.2015.7429497

Collaboratively examining several halos and their substructures in the immersive environment.

Cavern Halos: Exploring Spatial and Nonspatial Cosmological Data in an Immersive Virtual Environment

The use of electronic health records (EHRs) in clinical environments provides new opportunities for nurses to integrate data analyses into their practice. While having access to these records has many benefits, the act of recording, retrieving, and analyzing this data can nonetheless introduce communication issues, as navigating and interpreting large amounts of heterogeneous data can be difficult, and conclusions can be hard to validate. In this paper, we describe a series of integrated visual interfaces to help nurses document and reason about patient data and about clinicians&rsquo; understanding of patient data. The interfaces present the output of a predictive algorithm that makes use of historical EHR data, patient vital signs, and nurse handoff reports in order to classify a patient in terms of their likelihood of experiencing clinical events. Furthermore, the interfaces enable the nurses to quickly explore the original data and to examine other nurses&rsquo; interpretation of patient activity during previous shifts. We present a series of usage scenarios that introduce our interactive visualization tools in the context of real-world healthcare situations. 
CCS Concepts 
Applied computing -> Health care information systems; Health informatics; Human-centered computing -> Visual analytics; Information visualization

Screenshot of the augmented  ow sheet timeline.

Augmenting EHR Interfaces for Enhanced Nurse Communication and Decision Making

This paper introduces a novel web application that visualizes dark matter halo merger trees and their evolution through space and
time. Our application enables users to interact with individual halos within these trees in order to perform a range of visual analysis
tasks, including: identifying the substructure and superstructure of the halos; observing the movement of halos across a specified range of time steps; and comparing the branching characteristics of multiple trees. Central to our application is the ability to navigate the halos by interactively &ldquo:jumping&rdquo; from tree to tree. By clearly marking halos that have &ldquo;tributaries&rdquo; - that is, halos that split off into multiple halos or that merge with one or more halos - we make it easy for the user to traverse the complex structure of the universe. Our application is publicly available online and runs at interactive rates on the browser.

A screen capture of our web application showing a complete dark matter halo merger tree.

Halos in a Dark Sky: Interactively Exploring the Structure of Dark Matter Halo Merger Trees

University of Illinois at Chicago&rsquo;s CAVE2 is a large-scale, 320-degree radial visualization environment with a 360-degree 20.2 channel radial speaker system. The purpose of our research is to develop solutions for spatially accurate playback of audio within a virtual reality environment, reconciling differences between the circular speaker array, the location of a user in the physical space, and the location of virtual sound objects within CAVE2&rsquo;s OmegaLib virtual reality software, all in real time. Previous research presented at AES 137 detailed our work on object geometry, dynamically mapping a virtual object’s width and distance to the speaker array with volume and delay compensation. Our recent work improves virtual width perception using dynamic decorrelation with transient fidelity, implemented via Supercollider on the CAVE2 sound server.

Decorrelated Audio Imaging in Radial Virtual Reality Environments

This paper describes novel methods for constructing the intrinsic geometry of the human brain connectome using dimensionality-reduction techniques. We posit that the high-dimensional, complex geometry that represents this intrinsic topology can be mathematically embedded into lower dimensions using coupling patterns encoded in the corresponding brain connectivity graphs.
We tested both linear and nonlinear dimensionality-reduction techniques using the diffusion-weighted structural connectome data acquired from a sample of healthy subjects. Results supported the nonlinearity of brain connectivity data, as linear reduction techniques such as the multidimensional scaling yielded inferior lower-dimensional embeddings. To further validate our results, we
demonstrated that for tractography-derived structural connectome more influential regions such as rich-club members of the brain are more centrally mapped or embedded. Further, abnormal brain connectivity can be visually understood by inspecting the altered geometry of these three-dimensional (3D) embeddings that represent the topology of the human brain, as illustrated using simulated lesion studies of both targeted and random removal. Last, in order to visualize brain&rsquo;s intrinsic topology we have developed software that is compatible with virtual reality technologies, thus allowing researchers to collaboratively and interactively explore and manipulate brain connectome data. 
Keywords: Diffusion MRI, Tractography, Dimensionality reduction, Virtual reality, Connectomics

The anatomy of a human brain shown in BRAINtrinsic. Inset shows the Oculus Rift technology being used o visualize brain connectome data using the proposed approach.

The intrinsic geometry of the human brain connectome

Visualizing and analyzing the relationships between taxonomic entities represented in multiple input classifications is both challenging and required due to recurrent new discoveries and inferences of taxa and their phylogenetic relationships. Despite the availability of numerous visualization techniques, the large size of hierarchical classifications and complex relations between taxonomic entities generated during a multi-taxonomy alignment process requires new visualizations. This paper introduces ProvenanceMatrix, a novel tool allowing end users (taxonomists, ecologists, phylogeneticists) to explore and comprehend the outcomes of taxonomic alignments. We illustrate the use of ProvenanceMatrix through examples using taxonomic classifications of various sizes, from a few to hundreds of taxonomic entities and hundreds of thousands of relationships. 
Keywords: Taxonomic classification, multi-taxonomy alignment, phylogenetic relationship, matrix representation, glyph-based visualization

Abstract toolkit input and output example rendered with GraphViz.

ProvenanceMatrix: A Visualization Tool for Multi-Taxonomy Alignments

A common strategy for encoding multidimensional data for visual analysis is to use dimensionality reduction techniques that project data from higher dimensions onto a lower-dimensional space. This article examines the use of motion to retain an accurate representation of the point density of clusters that might otherwise be lost when a multidimensional dataset is projected into a two-dimensional space. Specifically, we consider different types of density-based motion, where the magnitude of the motion is directly related to the density of the clusters. We investigate how users interpret motion in two-dimensional scatterplots and whether or not they are able to effectively interpret the point density of the clusters through motion. We conducted a series of user studies with both synthetic and real-world datasets to explore how motion can help users in completing various multidimensional data analysis tasks. Our findings indicate that for some tasks, motion outperforms the static scatterplots; circular path motions in particular give significantly better results compared to the other motions. We also found that users were easily able to distinguish clusters with different densities as long the magnitudes of motion were above a particular threshold. Our results indicate that incorporating density-based motion into visualization analytics systems effectively enables the exploration and analysis of multidimensional datasets. 
Keywords: Visual analytics, visual clustering, user study, spatial layout, scatterplots, multidimensional visualization,
perception

Examples from our user study in which a user is asked to detect a cluster with a different magnitude of motion.

Density-based motion

Our long-term goal is to ensure nurse clinical decision support (CDS) works as intended before full deployment in clinical practice. As part of a broader effort, this pilot explores factors influencing acceptance/non-acceptance of 8 CDS suggestions displayed through selecting a blinking red button in an electronic health record (EHR) based nursing plan of care software prototype. A diverse sample of 21 nurses participated in this high fidelity clinical simulation experience and completed a questionnaire to assess reasons for accepting/not accepting the CDS suggestions. Of 168 total suggestions displayed during the experiment (8 for each of the 21 nurses), 123 (73.2%) were accepted and 45 (26.8%) were not accepted. The mode number of acceptances by nurses was 7 of 8 with only 2 of 21 nurses accepting all. The main reason for CDS acceptance was the nurse’s belief that the suggestions were good for the patient (n=100%) with other features being secondarily reinforcing. Reasons for non-acceptance were less clear, with under half of the subjects indicating low confidence in the evidence. This study provides preliminary evidence that high-quality simulation and targeted questionnaires about specific CDS selections offer a cost-effective means for testing before full deployment in clinical practice. 
Keywords:</strrong> electronic health records; nursing care plans; standardization; clinical decision support systems; simulation lab

Nursing care plan examples within the three different HANDS prototypes.

Use of Simulation to Study Nurses&rsquo; Acceptance and Nonacceptance of Clinical Decision Support Suggestions

Imagining Macondo is a public artwork that commemorates Nobel Prize-winning author Gabriel García Marquez. It was first showcased at the Bogota International Book Fair in April 2015 to an audience of more than 300,000 over the course of two weeks. The project involved extensive collaboration between an international team of artists, designers, and programmers. This article explores the historical and artistic contexts for the creation of the work, discusses the audience reception to the work, and describes the significant software and production requirements necessary to create an installation with thousands of participants and hundreds of thousands of viewers.

Users interacting with the Imagining Macondo installation during the 2015 Bogota International Book Fair.

Imagining Macondo: Interacting with García Marquez&rsquo;s Literary Landscape

This paper explores the specialized nature of research-oriented web applications that enable interactions with and the visual analysis of &ldquo;Big Data,&rdquo; i.e., large, heterogeneous scientific datasets. We introduce a pragmatic methodology for the design and evaluation of scientific workflows in research-oriented web applications. Through an in-depth usability study of the CoGe web application, a system that provides a rich set of tools for exploring genomic datasets, we demonstrate: how to identify bottlenecks in multi-step tasks; how to analyze these bottlenecks in order to provide effective solutions for improving user experience; and how these solutions may more generally apply to similar research-oriented websites in other scientific domains that also enable scientific workflows. Specifically, we provide details regarding: our user interviews, the visualization system we created to analyze complex tasks associated with scientific workflows, and how this analysis directly leads to suggestions for improvements in the current implementation of the CoGe web application. A follow-up study was carried out which indicates that our suggestions improved the ability of CoGe users to navigate and complete custom workflows, leading us to believe that our approach could also be applied to other research-oriented web applications that utilize scientific datasets. 
Keywords: Scientific workflows, Big Data, user evaluation, visual analytics.

Comparison of typical workflow for user completion of Task.

Designing and Evaluating Scientific Workflows for Big Data Interactions

Achievable networking performance of applications in a supercomputer depends on the exact combination of the communication patterns of the applications and the routing algorithms used by the supercomputer. In order to achieve the highest networking performance for the applications the routing algorithms need to be designed optimally for those communication patterns. However, while communication patterns usually have a wide variation from application to application and even from phase to phase in an application, routing algorithms have a limited variation and usually are optimized for typical communication patterns. This results in high networking performance for favored communication patterns but low networking performance for others. In this paper we present approaches for improving networking performance by rebalancing load on physical links on the Blue Gene Q supercomputer. We realize our approaches in a framework called OPTIQ and demonstrate the efficacy of our framework via a set of benchmarks. Our results show that we can achieve 30% higher throughput on experiment with data and patterns from a real application. The improvement can be up to several times higher throughput than default MPI_Alltoallv used in the Blue Gene Q supercomputer for certain communication patterns. 
Keywords: multi-path data movement, BG/Q, optimization, heuristic, interconnection network, communication patterns, network load balancing

Multipath Load Balancing for M x N Communication Patterns on the Blue Gene⁄Q Supercomputer Interconnection Network

Thanks to advances in non-invasive technologies such as functional Magnetic Resonance Imaging (fMRI) and Diffusion Tensor Imaging (DTI), highly-detailed maps of brain structure and function can now be collected. In this context, brain connectomics have emerged as a fast growing field that aims at understanding these comprehensive maps of brain connectivity using sophisticated computational models. In this paper we present BRAINtrinsic, an innovative web-based 3D visual analytics tool that allows users to intuitively and iteratively interact with connectome data. Moreover, BRAINtrinsic implements a novel visualization platform that reconstructs connectomes&rsquo; intrinsic geometry, i.e., the topological space as informed by brain connectivity, via dimensionality reduction. BRAINtrinsic is implemented with virtual reality in mind and is fully compatible with the Oculus Rift technology. Last, we demonstrate its effectiveness through a series of case studies involving both structural and resting-state MR imaging data. 
Keywords: connectomics, connectome datasets, intrinsic geometry, neuroimaging

Intrinsic geometry of the tractography-derived structural and the resting-state fMRI connectome, as well as the locations of rich-club regions in these spaces.

BRAINtrinsic: A Virtual Reality-Compatible Tool for Exploring Intrinsic Topologies of the Human Brain Connectome

The study presented in this paper focuses on a dimensional theory to augment agent nonverbal behavior including emotional facial expression and head gestures to evaluate subtle differences in fine-grained conditions in the context of emotional storytelling. The result of a user study in which participants rated perceived naturalness for seven different conditions showed significantly higher preference for the augmented facial expression whereas the head gesture model received mixed ratings: significant preference in high arousal cases (happy) but not significant in low arousal cases (sad). 
Keywords: Virtual humans, Intelligent agents, Dimensional theory, PAD

Emotionally Augmented Storytelling Agent - The Effects of Dimensional Emotion Modeling for Agent Behavior Control

This paper describes an artwork created in response to a question about the role of the artist in communicating climate change issues. The artwork, titled Turbulent World, incorporates turbulence and surprise as a means to visualize the potential instability of our culture and the environment due to climatic changes indicated by increased worldwide temperatures. The artwork makes use of a custom fluid engine that can represent any amount of turbulence and energy. A dataset encoding a simulation of rising surface-air temperatures over the next century is mapped to the turbulence system; and the visualization is updated as the months and years flow by, based on the projected temperatures at different areas of the world. That is, the increased turbulence of the system causes a representation of a map of the world to become distorted in different ways. A secondary view is overlaid, showing numerical data and providing a more dispassionate display of the inexorable increase in world temperature. 
Keywords: climate change, art-science, software art

Screen capture from the artwork as it shows the average surface-air temperature across the world for September 2067, using the projected values generated by the CM2.1 model developed by the Geophysical Fluid Dynamics Laboratory.

TurbulentWorld: An Artwork Indicating the Impact of Climate Change

&ldquo;Hearts and Minds: The Interrogations Project&rdquo; is an interactive installation made for the CAVE2&trade; large-scale 320-degree panoramic virtual reality environment that visualizes stories of violence and the post-traumatic stress experienced by ordinary American soldiers who became torturers in the course of serving their country. During the American-led counterinsurgency and counterterrorism campaigns in Iraq in the years after September 11, 2001, the torture and abuse of detainees was a commonplace tactic. The project is based on interviews of American soldiers and attempts to extend and make accessible difficult narratives based on the actual testimonies involved. By bridging together different stories and environments, the project uses visualization to provide conditions for stories to unfold—stories that connect the homes that soldiers come from and return to, with distant experiences of war. The immersion of the CAVE2 virtual reality theater allows for a different type of affective experience of the narrative, activated through the visceral immersion in the visual and auditory environment. The project represents a complex contemporary issue and provides a platform for discussion of military interrogation methods and their effects on detainees, soldiers, and society. The project was developed through a unique international collaboration between artists, scientists, and researchers from five different universities. 
Keywords: Virtual reality, interactive installation, performance, immersive environments, visualization, art, storytelling, torture, digital humanities.

Hearts and Minds virtual environment and interaction with the immersive scene.

Hearts and Minds: The Residue of War

This paper discusses the conflicting expectations for media artists taking part in art-science collaborations. Despite the increasing opportunity to participate in these interdisciplinary projects, it can be unclear how media arts activities are best articulated, or even if they need to be defined at all. Additionally, this paper examines a methodological framework widely used in the visualization community for identifying different visualization tasks within research activities. Inspired by its success, this paper proposes a new methodological framework for media arts activities in art-science contexts. This framework splits media arts activities into overlapping areas: generation, augmentation, provocation and mediation, providing a useful way to articulate the broader importance of media arts in interdisciplinary collaboration.

Articulating Media Arts Activities in Art-Science Contexts

We present findings of a comparative study of numeracy and graph literacy in a representative group of 60 practicing nurses. The paper focuses on a fundamental concern related to the effectiveness of numeric information displayed in various features in the electronic health record during clinical workflow. Our findings suggest the need to consider numeracy and graph literacy when presenting numerical information as well as the potential for tailoring numeric display types to individual&rsquo;s cognitive strengths. 
Keywords: Clinical decision making; electronic health records; graphical literacy; numeracy; nurses

Numeracy and Graph Literacy Means and Standard Deviations by Nurse Characteristics

Nurses&rsquo; Numeracy and Graphical Literacy: Informing Studies of Clinical Decision Support Interfaces

Background: The volume of complete bacterial genome sequence data available to comparative genomics researchers is rapidly increasing. However, visualizations in comparative genomics--which aim to enable analysis tasks across collections of genomes--suffer from visual scalability issues. While large, multi-tiled and high-resolution displays have the potential to address scalability issues, new approaches are needed to take advantage of such environments, in order to enable the effective visual analysis of large genomics datasets. 
Results: In this paper, we present Bacterial Gene Neighborhood Investigation Environment, or BactoGeNIE, a novel and visually scalable design for comparative gene neighborhood analysis on large display environments. We evaluate BactoGeNIE through a case study on close to 700 draftEscherichia coli genomes, and present lessons learned from our design process. 
Conclusions: BactoGeNIE accommodates comparative tasks over substantially larger collections of neighborhoods than existing tools and explicitly addresses visual scalability. Given current trends in data generation, scalable designs of this type may inform visualization design for large-scale comparative research problems in genomics.

BactoGeNIE running on EVL's Cyber-Commons tiled display.

BactoGeNIE: A Large-Scale Comparative Genome Visualization for Big Displays

Background: Molecular and systems biologists are tasked with the comprehension and analysis of incredibly complex networks of biochemical interactions, called pathways, that occur within a cell. Through interviews with domain experts, we identified four common tasks that require an understanding of the causality within pathways, that is, the downstream and upstream relationships
between proteins and biochemical reactions, including: visualizing downstream consequences of perturbing a protein; finding the shortest path between two proteins; detecting feedback loops within the pathway; and identifying common downstream elements from two or more proteins. 
Results: We introduce ReactionFlow, a visual analytics application for pathway analysis that emphasizes the structural and causal relationships amongst proteins, complexes, and biochemical reactions within a given pathway. To support the identified causality analysis tasks, user interactions allow an analyst to filter, cluster, and select pathway components across linked views. Animation is used to highlight the flow of activity through a pathway. 
Conclusions: We evaluated ReactionFlow by providing our application to two domain experts who have significant experience with biomolecular pathways, after which we conducted a series of in-depth interviews focused on each of the four causality analysis tasks. Their feedback leads us to believe that our techniques could be useful to researchers who must be able to understand and analyze the complex nature of biological pathways. 
ReactionFlow is available at <a href="https://github.com/CreativeCodingLab/ReactionFlow">https://github.com/CreativeCodingLab/ReactionFlow</a>. 
Keywords: Pathway visualization; Biological networks; Causality analysis; Topological ordering

Downstream consequences if a protein/complex is &dlquo;knocking out&drquo; in the defects in vitamin and cofactor metabolism pathway: (a) GIF protein (b) TCll:Cbl complex (c) MTRR:MTR complex.

ReactionFlow: An Interactive Visualization Tool for Causality Analysis in Biological Pathways

Background: Molecular activation pathways are inherently complex, and understanding relations across many biochemical reactions and reaction types is difficult. Visualizing and analyzing a pathway is a challenge due to the network size and the diversity of relations between proteins and molecules. 
Results: In this paper, we introduce PathwayMatrix, a visualization tool that presents the binary relations between proteins in the pathway via the use of an interactive adjacency matrix. We provide filtering, lensing, clustering, and brushing and linking capabilities in order to present relevant details about proteins within a pathway. 
Conclusions: We evaluated PathwayMatrix by conducting a series of in-depth interviews with domain experts who provided positive feedback, leading us to believe that our visualization technique could be helpful for the larger community of researchers utilizing pathway visualizations. 
PathwayMatrix is freely available at <a href="https://github.com/CreativeCodingLab/PathwayMatrix">https://github.com/CreativeCodingLab/PathwayMatrix</a>. 
Keywords: Pathway visualization; Biological networks; Matrix visualization; Binary Relationships

Visualizing protein binary relationships of the HIV life cycle by different tools: BioFabric (top panel), Cytoscape (middle panel), Hive Plot (bottom left panel), and PathwayMatrix (bottom right panel).

PathwayMatrix: Visualizing Binary Relationships between Proteins in Biological Pathways

Background: Biologists make use of pathway visualization tools for a range of tasks, including investigating inter-pathway connectivity and retrieving details about biological entities and interactions. Some of these tasks require an understanding of the hierarchical nature of elements within the pathway or the ability to make comparisons between multiple pathways. We introduce a technique inspired by LineSets that enables biologists to fulfill these tasks more effectively. 
Results: We introduce a novel technique, Extended LineSets, to facilitate new explorations of biological pathways. Our technique incorporates intuitive graphical representations of different levels of information and includes a well designed set of user interactions for selecting, filtering, and organizing biological pathway data gathered from multiple databases. 
Conclusions: Based on interviews with domain experts and an analysis of two use cases, we show that our technique provides functionality not currently enabled by current techniques, and moreover that it helps biologists to better understand both inter-pathway connectivity and the hierarchical structure of biological elements within the pathways.

Comparison of the same dataset realized with (a) LineSets, showing set information and network information simultaneously leads to visual clutter; (b) Extended LineSets, mitigates visual clutter via a more effective depiction of category information.

Extended LineSets: A visualization technique for the interactive inspection of biological pathways

The purpose of this concurrent mixed-methods study was to 1) examine the factors pediatric intensive care unit nurses consider when assessing and intervening for children who report severe pain and to 2) determine the effect of child behavior and diagnosis on the nurses' pain ratings and intervention choices for written and virtual human vignettes. Quantitative and qualitative results substantiated that despite recommendations to use self-report, many PICU nurses use behavior as the primary indicator to assess and treat pain, even when a child is old enough to articulate pain intensity and there is sufficient cause for pain to be present. 
Keywords: Acute pain; Hospitalized child; Knowledge use in pain care; Patient simulation; Pediatric intensive care; Pediatric nurse.

Univariate Differences in Pain Ratings and Morphine Doses Between Behaviors

PICU Nurses&rsquo; Pain Assessments And Intervention Choices for Virtual Human And Written Vignettes

This paper introduces a technique that enables the creative reshaping of one or more video signals based on granular synthesis techniques, normally applied only to audio signals. We demonstrate that a wide range of novel video processing effects can be generated through conceptualizing a video signal as being composed of a large number of video grains. These grains can be manipulated and maneuvered in a variety of ways, and a new video signal can then be created through the resynthesis of these altered grains; effects include cloning, rotating, and resizing the video grains, as well as repositioning them in space and time. These effects have been used successfully in a series of interactive multimedia performances, leading us to believe that our approach has significant artistic potential. 
Categories and Subject Descriptors (according to ACM CCS): I.3.6 [Computer Graphics]: Methodology and Techniques-Graphics data structures and data types; I.3.7 [Computer Graphics]: Three-Dimensional Graphics
and Realism-Animation.

This video portrait captures the subject as she is singing, and the audio features (pitch and amplitude) change aspects of the temporal shuffling of the grain, including grain size and delay.

Video Granular Synthesis

Collaborative visual analytics that feature mixtures of spatial and nonspatial data occur across disciplines, and are particularly common in bioinformatics, neuroscience and geospatial analysis. In this work we analyze, from a human-centric perspective, data collected from the design and evaluation of three successful visual analysis tools, spanning seven case studies. We focus on the importance of the users&rsquo; background to the design process, and we discuss the importance of visual scaffolding to such collaborative, integrated spatial and nonspatial visual analysis tools. Scaffolding is a psychology concept which denotes the support given during a learning process. We further present evidence that spatial and nonspatial coordinated views can serve as a form of visual scaffolding for expert-level, collaborative visual analyses. 
Categories and Subject Descriptors (according to ACM CCS): I.3.3 [Computer Graphics]: Visual analysis processes
and workflows - Cognitive and perceptual aspects of visual analytics - Collaborative visual analytics

Two visual analyses in a visual framework for the integration, comparison, and exploration of correlations in spatial and nonspatial geriatric data.

Visual Scaffolding in Integrated Spatial and Nonspatial Analysis

Algorithmically decreasing speech tempo, or &ldquo;audio dilation,&rdquo; can improve speech perception in attentionally demanding tasks [Gygi & Shafiro (2014), Hear. Res. 310, pp. 76-86]. On-line audio dilation [Novak, et al. (2013), Interspeech 1869-1871] is a recently developed technique that decreases the tempo of audio signals as they are generated. This pilot study investigated effects of on-line audio dilation on performance in interactive problem solving tasks. We used a Diapix task [Baker & Hazan (2011), Behav. Res. Methods 43(3), 761-770] to elicit and record spontaneous speech from pairs of participants under various dilation conditions: participants, seated in different rooms, were asked to find ten differences on two similar pictures, while their speech was either transmitted as spoken or dilated. Conditions tested include stretching one, both, or neither audio signal by 40%. Subsequent analysis shows that the technique, even using this substantial increase, did not interfere with interactive problem solving tasks, and did not lead to changes in speech production rate, measured as number of syllables per second. The lack of negative effects of on-line speech dilation provides a preliminary basis for further assessment of this method in speech perception tasks with high attentional and memory load.

Diapix Image Test

Audio dilation in real time speech communication

Dynamic network visualization has been a challenging topic for dynamic networks analysis, especially for spatially embedded networks like brain networks.  In this paper, we present an animated interactive visualization design that combines enhanced node-link diagrams and distance matrix layouts to assist neuroscientists in their exploration of dynamic brain networks and that enables them to understand how functional connections relate to the spatial structure of the brain.  Our visualization also provides the ability to observe the evolution of a network, the change in the community identities, and node behavior over time.

Screenshot of our visualization tool showing the coordinated dual-representation of dynamic network data.

Visualizing Dynamic Brain Networks Using an Animated Dual-Representation

Large high-resolution displays are becoming increasingly common in research settings, providing data scientists with visual interfaces for the analysis of large datasets. Numerous studies have demonstrated unique perceptual and cognitive benefits afforded by these displays in visual analytics and information visualization tasks. However, the effects of these displays on knowledge discovery in exploratory visual analysis are still poorly understood. We present the results of a small-scale study to better understand how display size and resolution affect insight. Analyzing participants&rsquo; verbal statements, we find preliminary evidence that larger displays with more pixels can significantly increase the number of discoveries reported during visual exploration, while yielding broader, more integrative insights. Furthermore, we find important differences in how participants performed the same visual exploration task using displays of varying sizes. We tie these results to extant work and propose explanations by considering the cognitive and interaction costs associated with visual exploration.

An overview map (left) and a compound view showing one years worth of criminal activity (a) for a particular type of crime (b). The heatmap shows the aggregate spatial distribution of crime over the year.

Effects of Display Size and Resolution on User Behavior and Insight Acquisition in Visual Exploration

The popularity of large-scale, high-resolution display walls, as visualization endpoints in eScience infrastructure, is rapidly growing. These displays can be connected to distributed computing resources over high-speed network, providing effective means for researchers to visualize, interact with, and understand large volumes of datasets. Typically large display walls are built by tiling multiple physical displays together and running a tiled display wall required a cluster of computers. With the advent of advanced graphics hardware, a single computer can now drive over a dozen displays, thereby greatly reducing the cost of ownership and maintenance of a tiled display wall system. This in turn enables a broader user base to take advantage of such technologies. Since tiled display walls are also well suited to collaborative work, users tend to launch and operate multiple applications simultaneously. To ensure that applications maintain a high degree of responsiveness to the users even under heavy use loads, the display wall must now ensure that the limited system resources are prioritized to maximize interactivity rather than thread-level fair sharing or overall job-completion throughput. In this paper, we present a new resource scheduling scheme that is specifically designed to prioritize responsiveness in collaborative large display wall environments where multiple users can interact with multiple applications simultaneously. We evaluate our scheduling scheme with a user study involving groups of users interacting simultaneously on a tiled display wall with multiple applications. Results show that our scheduling framework provided a higher frame rate for applications, which led to a significantly higher user performance (approx. 25%) in a target acquisition test when compared against the traditional operating system scheduling scheme. 
Keywords: Scheduling, Human factors, Algorithms, Interactive systems, Distributed graphics

A single machine-driven tiled-display wall running SAGE at the Electronic Visualization Laboratory in the University of Illinois at Chicago. The 20′ by 6′ display wall is made up of 18 LCD panels with a total resolution approximately 17 megapixels.

Multiuser-Centered Resource Scheduling for Collaborative Display Wall Environments

When investigating multidimensional data sets with very large numbers of objects and/or a very large number of dimensions, a variety of visualization methods can be employed in order to represent the data effectively and to enable the user to explore the data at different levels of detail. A common strategy for encoding multidimensional data for visual analysis is to use dimensionality reduction techniques that project data from higher dimensions onto a lower-dimensional space. In this paper, we focus on projection techniques that output 2D or 3D scatterplots which can then be used for a range of data analysis tasks. Existing taxonomies for multidimensional data projections focus primarily on tasks in order to evaluate the human perception of class or cluster separation and/or preservation. However, real-world data analysis of complex data sets often includes other tasks besides cluster separation, such as: cluster identification, similarity seeking, cluster ranking, comparisons, counting objects, etc. A contribution of this paper is the identification of subtasks grouped into four main categories of data analysis tasks. We believe that this user-centric task categorization can be used to guide the organization of multidimensional data projection layouts. Moreover, this taxonomy can be used as a guideline for visualization designers when faced with complex data sets requiring dimensionality reduction. Our taxonomy aims to help designers evaluate the effectiveness of a visualization system by providing an expanded range of relevant tasks. These tasks are gathered from an extensive study of visual analytics projects across real-world application domains, all of which involve multidimensional projection. In addition to our survey of tasks and the creation of the task taxonomy, we also explore in more detail specific examples of how to represent data sets effectively for particular tasks. These case studies, while not exhaustive, provide a framework for how specifically to reason about tasks and to decide on visualization methods. That is, we believe that this taxonomy will help visualization designers to determine which visualization methods are appropriate for specific multidimensional data projection tasks. 
Keywords: Multidimensional data analysis, task taxonomy, multidimensional data projection, user-centric evaluation.

Non-convex hulls computed from enclosing surfaces isodistant to cluster using LSP projection.

A user-centric taxonomy for multidimensional data projection tasks

SAGE, the Scalable Adaptive Graphics Environment, is the defacto operating system for managing Big Data content on scalable-resolution tiled display walls, providing the scientific community with persistent visualization and collaboration services for global cyberinfrastructure. SAGE2, the next-generation SAGE, is a wholly new implementation of SAGE using state-of-the-art web and cloud technologies. SAGE2 better addresses the needs of contemporary collaborative projects, and has the potential to expand the SAGE user community because of its enhanced support for development and integration of multi-user applications and lower barrier to entry.

SAGE2 at the University of Sunshine Coast, Mechanical Engineering, CAVE2 Australia

SAGEnext: Next Generation Integrated Persistent Visualization and Collaboration Services for Global Cyberinfrastructure

Our paper presents new perceptual metrics for evaluating time-coherent animations based on different matching strategies. We introduce new terminology to more accurately describe perceptual aspects of moving images. These terms, &ldquo;smoothness,&rdquo; &ldquo;cohesion,&rdquo; and &ldquo;accuracy,&rdquo; are used to describe the quality of the matching algorithms in comparison to a ground-truth reference animation. Smoothness is a desirable quality of the animation that implies that the image elements should change slowly with time. Abrupt changes on position, size, shape, etc., are distracting and destroy the feeling of continuity. Cohesion is the spatial equivalent of smoothness. In a cohesive animation, features that are close to each other should behave in a similar way. Accuracy is an indication of how well the recreated image resembles the original analyzed picture. We introduce a set of metrics that quantify these terms and that can be applied to parametric spaces of any finite dimension. We conducted a user study with 30 students in order to find out which matching algorithms are highly related to these perceptual metrics. We discuss how these matching strategies could be used effectively when developing novel techniques for the non-photorealistic rendering of videos, or stylized animation. 
Keywords: Motion perception, interactivity, animation, temporal coherence

The test sequence created as a ground truth for the matching of the inner circles.

Evaluating the perception of different matching strategies for time-coherent animations

The density of points within multidimensional clusters can impact the effective representation of distances and groups when projecting data from higher dimensions onto a lower dimensional space. This paper examines the use of motion to retain an accurate representation of the point density of clusters that might otherwise be lost when a multidimensional dataset is projected into a 2D space. We investigate how users interpret motion in 2D scatterplots and whether or not they are able to effectively interpret the point density of the clusters through motion. Specifically, we consider different types of density-based motion, where the magnitude of the motion is directly related to the density of the clusters. We conducted a series of user studies with synthetic datasets to explore how motion can help users in various multidimensional data analyses, including cluster identification, similarity seeking, and cluster ranking tasks. In a first user study, we evaluated the motions in terms of task success, task completion times, and subject confidence. Our findings indicate that, for some tasks, motion outperforms the static scatterplots; circular path motions in particularly give significantly better results compared to the other motions. In a second user study, we found that users were easily able to distinguish clusters with different densities as long the magnitudes of motion were above a particular threshold. Our results indicate that it may be effective to incorporate motion into visualization systems that enable the exploration and analysis of multidimensional data.

Density-based motion for multidimensional data projection

Introduction: As virtual experiences are increasingly used in health care training and research, it is important that adequate processes are applied for developing valid scenarios. We describe the development and validation of virtual human (VH) vignettes, computer-generated scenarios with animated patients and clinical information, for a mixed-methods study regarding nurses&rsquo; assessment and intervention choices for critically ill children&rsquo;s pain. 
Methods: We followed the case development and review process for high-fidelity simulation case scenarios, including the use of validated written vignettes and content experts. Forty nurses described their pain assessment and intervention choices for the newly derived VH vignettes and completed a pain questionnaire. Nurses&rsquo; reports of VH vignette consistency with their professional experience and recognition of VH facial expressions were evaluated to establish face validity. Their pain ratings for the VH and written (questionnaire) vignettes were evaluated for convergent validity. Qualitative content analysis, descriptive statistics, correlations, and paired t-tests were used. 
Results: Most nurses (68.4%) supported vignette consistency with their professional experience. Facial expression recognition was 98.4%. Smiling children&rsquo;s pain was rated significantly lower than grimacing children in both VH and written vignettes. Pain was rated significantly lower for grimacing children in the VH vignettes than the written vignettes. Virtual human vignette pain ratings were strongly correlated with their written counterparts. 
Conclusions: This process was effective for developing VH vignettes that demonstrated good face validity with participants and convergent validity with written vignettes. Virtual human vignettes may be useful in studying the influence of facial actions on nurses' choices for children&rsquo;s pain assessment and treatment.

Formation of base head models. Examples of photographs of two of the four children (left, top and bottom) used to develop the base head models for the virtual human vignettes.

Development and Validation of a Virtual Human Vignette to Compare Nurses Assessment and Intervention Choices for Pain in Critically Ill Children

U.S. energy consumption is dominated by the burning of fossil fuels such as coal, natural gas and petroleum. Production of electricity and propulsion systems are two primary reasons for this energy use. Both processes, at their core, involve turbulent combustion. Turbulent combustion modeling is an important area of research driven by the effort to improve the efficiency of these processes, reduce fuel consumption and reduce pollution. 
In order of decreasing fidelity and costs, three computational approaches to turbulence combustion are direct numerical simulation (DNS), large eddy simulation (LES) and Reynolds-averaged Navier-Stokes (RANS). Specific tensor quantities such as stress, strain, and turbulent stress play in these computational approaches, as part of the computational modeling process; these quantities are discussed in detail in our previous work. DNS requires directly capturing the wide range of lengthand time-scales. This severely limits the approach to simulations of relatively simple, canonical configurations. The complex nature of turbulent reacting flows can be attributed to the non-linear convection terms and scalar transport terms appearing in the coupled set of governing equations. In DNS these terms are accounted for without modeling. However, as discussed in more detail in our previous work. in both LES and RANS these terms create a closure problem and require modeling. As such, an important aspect of turbulence modeling and model validation involves comparison of the subgrid scale stress tensor (for LES) or the Reynolds stress tensor (for RANS) with available DNS data.

Publications - 2015

Improving Communication Throughput by Multipath Load Balancing on Blue Gene/Q

Textual Annotation Tools for SAGE2

TransitTrace: Route Planning using Ambient Displays

Understanding Transportation Accessibility of Metropolitan Chicago Through Interactive Visualization

“Show me data” Observational study of a conversational interface in visual data exploration

Cavern Halos: Exploring Spatial and Nonspatial Cosmological Data in an Immersive Virtual Environment

Contact