A5083540777- Data Visualization and Analytics
- Head and Neck Cancer Studies
- Radiomics and Machine Learning in Medical Imaging
- Scientific Computing and Data Management
- Data Management and Algorithms
- Data Analysis with R
- Medical Imaging and Analysis
- Image Retrieval and Classification Techniques
- Generative Adversarial Networks and Image Synthesis
- Cell Image Analysis Techniques
- Computer Graphics and Visualization Techniques
- Esophageal Cancer Research and Treatment
- Bioinformatics and Genomic Networks
- Gene Regulatory Network Analysis
- Protein Structure and Dynamics
- Species Distribution and Climate Change
- Video Analysis and Summarization
- Distributed and Parallel Computing Systems
- Advanced Clustering Algorithms Research
- Combustion and flame dynamics
- Advanced Data Storage Technologies
- Advanced Combustion Engine Technologies
- Topological and Geometric Data Analysis
- Advanced Vision and Imaging
- Remote Sensing in Agriculture
University of Illinois Chicago
2017-2021
University of Illinois Urbana-Champaign
2020
University of Illinois System
2018
Brown University
2014-2015
University of Pittsburgh
2011-2014
Visualization research often seeks designs that first establish an overview of the data, in accordance to information seeking mantra: "Overview first, zoom and filter, then details on demand". However, computational fluid dynamics (CFD), as well other domains, there are many situations where such a spatial is not relevant or practical for users, example when experts already have good mental analysis large overall structure may be related specific, information-driven tasks users. Using...
We describe a visual computing approach to radiation therapy (RT) planning, based on spatial similarity within patient cohort. In radiotherapy for head and neck cancer treatment, dosage organs at risk surrounding tumor is large cause of treatment toxicity. Along with the availability repositories, this situation has lead clinician interest in understanding predicting RT outcomes previously treated similar patients. To enable type analysis, we introduce novel topology-based measure, T-SSIM,...
This demo presents AstroShelf, our on-going effort to enable astrophysicists collaboratively investigate celestial objects using data originating from multiple sky surveys, hosted at different sites. The AstroShelf platform combines database and stream, workflow visualization technologies provide a means for querying displaying telescope images (in Google Sky manner), visualizations of spectrum data, managing annotations. In addition the user interface, supports programmatic interface...
We introduce a web-based computing infrastructure to assist the visual integration, mining and interactive navigation of large-scale astronomy observations. Following an analysis application domain, we design client-server architecture fetch distributed image data partition local into spatial index structure that allows prefix-matching objects. In conjunction with hardware-accelerated pixel-based overlays online cross-registration pipeline, this approach fetching, displaying, panning zooming...
Precision medicine seeks to tailor therapy the individual patient, based on statistical correlates from patients who are similar one under consideration. These can and should go beyond genetics, in general, tabular or array data that be easily represented computationally compared. For example, many types of cancer, cancer treatment toxicity depend large measure spatial disease spread-e.g., metastasizes regional lymph nodes head neck cancer. However, there is currently a lack methodology for...
Visualizing the complex probability landscape of stochastic gene regulatory networks can further biologists' understanding phenotypic behavior associated with specific genes.We present PRODIGEN (PRObability DIstribution GEne Networks), a web-based visual analysis tool for systematic exploration distributions over simulation time and state space in such networks. was designed collaboration bioinformaticians who research The combines novel way existing, expanded, new encodings to capture...
Simulation and modeling of turbulent flow, reacting flow in particular, involve solving for analyzing time-dependent spatially dense tensor quantities, such as stress tensors. The interactive visual exploration these quantities can effectively steer the computational combustion systems. In this article, authors analyze challenges symmetric-tensor visualization applied to calculation; most notable among are dataset size density. They analyze, together with domain experts, feasibility using...
Knowledge of the 3D structure and functionality proteins can lead to insight into associated cellular processes, speed up creation pharmaceutical products, develop drugs that are more effective in combating disease. We present design implementation a visual mining analysis tool help identify protein mutations across family structural models discover effect these on function. integrate sequence information common interface; multiple linked views computational backbone allow comparison at...
Simulation and modeling of turbulent flow, reacting flow in particular, involve solving for analyzing time-dependent spatially dense tensor quantities, such as stress tensors. The interactive visual exploration these quantities can effectively steer the computational combustion systems. In this article, authors analyze challenges symmetric-tensor visualization applied to calculation; most notable among are dataset size density. They analyze, together with domain experts, feasibility using...
Abstract Purpose Using a cohort of 582 head and neck cancer patients with nodal disease, we employed clustering over novel graph-based geometrical representation lymph node spread in order to identify groups similar patients. We show that these are significantly correlated radiation-associated dysphagia (RAD), predictive late aspiration feeding tube toxicity. Materials methods All received radiotherapy for oropharyngeal (OPC) had non-metastatic affected nodes the neck. Affected were...
We introduce a web-based, client-server computing infrastructure to assist the interactive navigation of large-scale astronomy observations. Large image datasets are partitioned into spatial index structure that allows prefix-matching objects. In conjunction with pixel-based overlays, this approach fetching, displaying, panning and zooming gigabit panoramas sky in real time. Images from three surveys (SDSS, FIRST simulated LSST results) cross-registered integrated as allowing cross-spectrum...
These preliminary results show that pixel-based overlays have the potential to generate scalable, graphical representations of astronomy data. This approach may allow us overcome bandwidth and screen-space current limitations in database visualization by following a WebGL - PHP client-server architecture. The advantages this are its versatility visual scalability (to pixel level), enabling analysis large datasets. resulting allows for flexible control over client-side scripts. Accessing...
Our Web-based image analysis tool for the VAST 2017 Mini-Challenge 3 combines small multiple views of satellite images, linked semantic zooming and intensity histograms, along with filter controls. The resulting allow users to interactively analyze spatio-temporal changes in preserve area.
Mini Challenge 2 of the VAST 2017 focused on a small industrial area south fictional Mistford preserve, specifically around four manufacturing factories. Our main goal was to develop visual analytics tool explore spatio-temporal chemical readings and wind data. Specifically, we wanted determine which factories were responsible for emitting chemicals performance nine sensors in area. In order help achieve this goal, developed web-based application that utilizes interactive visualizations path...