Deep Umbra is a novel computational framework that enables the quantification of sunlight access and shadows at a global scale. Our framework is based on a conditional generative adversarial network that considers the physical form of cities to compute high-resolution spatial information of accumulated shadow cast for the different seasons of the year. 
Deep Umbra&rsquo;s primary motivation is the impact that shadow management can have on people&rsquo;s quality of life, since it can affect levels of comfort, heat distribution, public parks, etc. We also present the Global Shadow Dataset, a comprehensive dataset with the accumulated shadow information for over 100 cities on 6 continents. 
A <a href="https://www.evl.uic.edu/shadows/map">webviewer</a> has been developed to visualize accumulated shadow cast over a specific geographic area. The map shows the accumulated shadow for three different days of the year - Summer, Spring/Fall, and Winter.

Deep Umbra: A Global-Scale Generative Adversarial Approach for Sunlight Access and Shadow Accumulation in Urban Spaces

The COVID-19 pandemic has upended the world over the past year, in a health crisis not seen since the 1918 flu pandemic. As of July 2021, the disease has caused the death of more than 4 million people worldwide, 600,000 in the US alone. While waiting for an
effective vaccine amidst the surging cases worldwide, governments adopted measures to inhibit the spread of the disease. Despite these efforts, waves of COVID-19 infection with varying characteristics, continued to ravage communities [24], highlighting the necessity to understand the spatio-temporal complexity of the problem. Forecast models are among the tools available to public health experts and policymakers to predict likely pandemic outcomes (i.e., number of cases, deaths) and prepare for different scenarios. In the past year, several models have been proposed by a myriad of experts, and using vastly different assumptions, methodologies, parameters and data sources. As a result, models produce a range of sometimes radically different forecasts, especially at higher spatial resolutions, limiting their use by decision makers and the public in general. In order to address this problem, more recently, ensemble models have been proposed for COVID-19 forecasts, with promising results. Our research in this area has resulted in COVID-19 EnsembleVis, a web-based interactive visual interface that allows the assessment of the errors of ensembles and individual models by enabling users to effortlessly navigate through and compare the outputs of models considering their space and time dimensions. COVID-19 EnsembleVis enables a more detailed understanding of uncertainty and the range of forecasts generated by individual models.

Enabling the visual analysis of COVID-19 ensemble forecast models with COVID-19 EnsembleVis.

COVID-19 EnsembleVis: Visual Analysis of County-level Ensemble Forecast Models

DASS is a visual computing system designed to help scientists build machine learning models that classify head and neck cancer patients into groups, based on their risk of developing permanent side effects from radiation therapy. Its visual interface helps data scientists to build a model for each side-effect by visualizing the patients&rsquo anatomical and other clinical data, and to figure out what parameters the model should consider, such as which organs may be damaged when side effects occur, what amount of radiation dose causes the damage to the organs, and how many risk groups to create. DASS also has features for explaining to clinicians the finished models and examining whether their predictions perform well, and the reasons behind why a given patient belongs in a certain risk group. DASS was created by visual computing researchers at EVL, in collaboration with clinicians at the MD Anderson Cancer Center and data mining specialists at University of Iowa, and is supported by funding from the US National Institutes of Health (NIH awards NCI-R01-CA258827 and NLM-R01-LM012527), and US National Science Foundation (NSF awards CDSE-1854815 and CNS-1828265).

Stylized scatterplot. Patients are represented by a custom glyph that encodes the outcome of interest as marks extending radially. Markers are colored by cluster membership, and a contour is shown around the currently selected cluster.

DASS

Cancers that depend on the spatial location of the disease affect all ethnicities, and are diagnosed each year in large numbers in the United States, leading to large, rich repositories of patient data. This project develops a longitudinal decision support tool to assist clinicians in selecting treatments that balance survival and side-effects responses to cancer therapy. The project will lead to better insights into the biological and therapeutic factors related to cancer survival and toxicity, allowing personalization of care at the individual level. 
This research is sponsored by the National Institutes of Health: National Cancer Institute, and led by EVL faculty and Computer Science associate professor Liz Marai (PI) and co-PI Xinhua Zhang, along with collaborators Guadalupe Canahuate at University of Iowa, and Dave Fuller at University of Texas, MD Anderson Cancer Center.

Longitudinal Spatial-Nonspatial Decision Support for Competing Outcomes in Head and Neck Cancer Therapy

The UIC School of Public Health (UIC-SPH) Emergency Management and Resiliency Planning (EMRP) program, in collaboration with the Electronic Visualization Laboratory (EVL) and content experts, is presenting a series of Midwest Comprehensive Visualization Dashboard(s) (MCVD) focusing on environmental health and justice issues in this region. A MCVD has two components: the visualization interface(s), i.e., link to interactive maps, and the end-user documentation (i.e., background information, data sources, methodology, findings, etc.). 
<a href="https://univofillinois.maps.arcgis.com/apps/instant/nearby/index.html?appid=486f6e438ecf4b048deebc8dafd9f2c1&sliderDistance=1">Link to the interactive map on the MCVD dashboard.</a> 
MCVD has been created by Joel Flax-Hatch, Michael D. Cailas (School of Public Health), Fabio Miranda (EVL), and Apostolis Sambanis (Division of Environmental and Occupational Health Sciences - EOHS)) with the participation of Michael Siciliano (College of Urban Planning and Public Affairs), Sybil Derrible (Civil, Materials, and Environmental Engineering), Yuan Shao (EOHS), and Greg Arling (Purdue University). 
<a href="https://ago-item-storage.s3.us-east-1.amazonaws.com/0e1922e480134854918ed0f3f97319fe/MCVD_EJ.3.V1.1.pdf?X-Amz-Security-Token=IQoJb3JpZ2luX2VjEKb%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEaCXVzLWVhc3QtMSJHMEUCID3FXdMyOfYIb%2Fafu1eOW49fXssLx10%2Bs%2FOh29tg7m7yAiEA1YqdmEA8DmO%2FPfoparcE3Mr%2BOmRA89eoTEPh8ZoxOPcqgwQI3v%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FARAAGgw2MDQ3NTgxMDI2NjUiDITVsh%2FbXDCTjerOWyrXA9skMbqdx8pG2RMMt4k5ms%2BFG30KSO3pLnufYGx%2B7nvR%2FVMNwS4lJZoAEhMED%2B6y22UVHgV1N3lUcEhAuc6X8i2ZiFHReKQDLODjfuL9ZoU3ubq%2FhOXmxaqZTw%2BSZ852w%2FeZK77wD0TsrUcabCDNLManoYWOMxkwqiPnyHChuwCON8529O6KlCMp%2BoXYl%2BMk7AXVdS3Fd1onYEy10MpPnhVN%2Bp0cdixCuaxfcmN3EPxfbY4q667z5fyg1fcDBfKgW8tk%2B5kzXMxpk0V%2Bq3Q3pII%2B9i3872DDEZFicjSLaLTt1w%2BHGg22mn9wBH7tFDTzmDF6UJWLpuRXnw3Ek8MWmDqdB8BWgxAbFmJbmQBGEcB9Bb%2FxAKIReMXFtrM2nThAMcOfx%2BwuiPlqVuHAdbjN6gtUrd3tX8Kd7WI3Q2mRzok5nf8XZ96EmYr3dA3ck7fBiC2ywkCzhn6U%2B4919EFAy3gJkOEbSNileqOKg1L%2Fv%2FOn4MRXwsQbAXJsb6Fq4XO%2BzYfUUvHHviqYMDdyIZU1Za9mPX9UER9XdrBJ9D6%2BDiJmOnbEz6MKAdjzAjDwmHqR0VHWj%2B8zVk98xa0x%2FLxol1CHliq5YjWT%2FfPBl3GoVRXFtorDyjkJzDCil4aQBjqlATd4Im%2B%2F5Hnm3WhZD3xNPje0qXVNEpEczPEUg77lN747bEEo47cI25tpDgVFxS%2BTkWv4Bqv2ylplFaPThOnwvUW1ms4f7Br3Z%2BI64bISJNaBXGv%2F7hb68bFvHsCc8yqBo1TcZcw6CaOyLyLfA5dAz1BKQNpdoI0K5m9DLHlBtZaAIn2uXUp75ttP0UqMXjhTDsHJWwkuM%2FQNDPGMGtbHJMynIEjZ7w%3D%3D&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20220207T221949Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIAYZTTEKKE2RWCPQVC%2F20220207%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Signature=3d12d4d450d382bd441b4d23e229780f45b867eae9a23cec52b1794716f8d1cc">Read the press release outlining this research project.</a>

Research - 2021

Deep Umbra: A Global-Scale Generative Adversarial Approach for Sunlight Access and Shadow Accumulation in Urban Spaces

COVID-19 EnsembleVis: Visual Analysis of County-level Ensemble Forecast Models

DASS

Longitudinal Spatial-Nonspatial Decision Support for Competing Outcomes in Head and Neck Cancer Therapy

Midwest Comprehensive Visualization Dashboards: A New Environmental Justice Tool for Chicago Communities

Contact