NCSA Welcomes 2021-22 Faculty Fellows

07.21.21 -

by Sophie Bui

NCSA has named 15 new Faculty Fellows for the 2021-22 academic school year. The NCSA Faculty Fellowship is a competitive program for faculty and researchers at the University of Illinois at Urbana-Champaign, which provides seed funding for new collaborations that include NCSA staff as integral contributors to the project. The new fellows and their projects are:

Air Pollution Prediction Using Traffic Surveillance Camera Footage and Deep Learning

Mei Tessum Christopher Tessum Faculty Fellows: Mei Tessum (Agricultural and Biological Engineering), Christopher Tessum (Civil and Environmental Engineering)
NCSA Collaborators: Volodymyr Kindratenko and Dawei Mu

About the Project: Traffic-related air pollution is a major health burden in the United States; however, measuring pollution with traditional direct-measurement techniques at the required levels of granularity doesn’t scale well and has high demand in cost and labor. This project proposes creating a system using traffic camera footage and deep learning to predict traffic-related pollution concentrations. Initial results will help establish a larger project in cooperation with multiple metropolitan areas as a step toward a scalable system for hyperlocal pollution estimates.

Analyzing and Visualizing Linguistic Variation in Monolingual and Bilingual Speakers

David Dubin Silvina Montrul Faculty Fellows: Silvina Montrul (Spanish and Portuguese, Linguistics, College of Liberal Arts and Sciences), David Dubin (School of Information Sciences)
NCSA Collaborator: Todd Nicholson

About the Project: The majority of human language theories, models, and assessments are based on monolingual and English language data. Acknowledging linguistic diversity as a normal condition of human language is necessary to regard multi-and bilinguals as native speakers. This project proposes using computational resources to build an annotated corpus (with speech samples of Spanish and English speakers in the U.S. and Mexico) to investigate the underutilization of automation tools and encoding standards and improve productivity to lead to new discoveries.

BiteMap: Tracking Invasive Mosquitoes, Ticks, and Emerging Pathogens Through A Community-Engaged Web Registry

Holly Tuten Chris Stone Faculty Fellows: Holly Tuten
, Chris Stone (Illinois Natural History Survey)
NCSA Collaborators: Jong Lee, Colleen Bushell

About the Project: Tracking invasive vectors and emerging pathogens poses an enormous challenge to public health surveillance. Implementing a community-engaged approach to gather vectors and vector-borne disease data could enhance surveillance, field collections and testing. This project proposes developing a web-based portal where people in Illinois can share encounters with mosquitoes and ticks. It will provide insights into changes in where and when people encounter vectors, leading to a better understanding of Illinois’ vector-borne disease landscape to detect overlooked and emerging pathogens.

Building Data Infrastructure to Assess the Health Impacts of Wildfire Smoke

David Molitor Faculty Fellow: David Molitor (Finance)
NCSA Collaborators: Ben Galewsky, Matias Carrasco Kind, Colleen Bushell

About the Project: Many countries regulate air pollution to reduce harm to human health. Marginal pollution reductions and health effects inform and determine optimal environmental policy, especially for developed countries with relatively low pollution levels and high reduction costs. This project proposes assessing the health impacts of wildfire smoke, a common air pollution event expected to increase under climate change. Combining the dose and health impacts of smoke shocks will allow the team to understand the relationship between pollution and health better.

Design of Multiphysics Structures Using Generative Adversarial Networks

Kai James Faculty Fellow: Kai James (Aerospace Engineering)
NCSA Collaborators: Diab Abueidda, Erman Guleryuz

About the Project: With the advent of generative adversarial networks, machine learning tools can now autonomously perform open-ended creative tasks. Researchers are applying this technology to structural design optimization but are relatively limited to simple single-discipline problems. This project proposes creating a generative design algorithm that can rapidly produce multifunctional structures that resemble real-world engineering structures. By applying this approach in aircraft and vehicle design, the team will contribute to better-performing lightweight systems and reduce emissions, thereby contributing to more sustainable vehicles.

Identifying Conflicting Claims in Clinical Literature Using Natural Language Processing and Knowledge Graphs

Halil Kilicoglu Faculty Fellow: Halil Kilicoglu (School of Information Sciences)
NCSA Collaborators: Michael Bobak, Colleen Bushell

About the Project: A vast amount of biomedical knowledge is published daily, making new scientific claims and confirming or refuting earlier ones. Natural language processing (NLP) techniques are used to extract information from scientific publications; however, little attention has been paid to interpreting knowledge claims and situating them within the larger scientific evidence base. This project proposes redesigning an existing biomedical NLP tool, SemRep, to extract knowledge claims and contextual characteristics from scientific publications more effectively and generate publication-specific knowledge graphs.

NCSA AI and Compute for Super-Resolution Ultrasound Advancement

Pengfei Song Faculty Fellow: Pengfei Song (Electrical and Computer Engineering, Beckman Institute)
NCSA Collaborators: Peter Groves, Colleen Bushell

About the Project: Super-resolution ultrasound localization microscopy (ULM) has great potential as a medical imaging technology due to its unique combination of imaging penetration and spatial resolution. Capturing a ULM image takes an excessive amount of time and has consequently prevented deploying the technology in a clinical setting. This project proposes using deep learning to develop a shareable microvascular database that shortens data acquisition and post-processing time to facilitate faster technique development and ultimately achieve the clinical translation of ULM.

Toward Ultrasound Brain Imaging via Full-Wave Acoustic Simulations and Deep Learning

Aiguo Han Faculty Fellow: Aiguo Han (Electrical and Computer Engineering)
NCSA Collaborator: Volodymyr Kindratenko

About the Project: Transcranial ultrasound could enable a broad variety of applications in brain imaging, including hemorrhage detection and stroke diagnosis, among others. Despite the great potential, it has not been widely used in adult brain imaging because their skulls cause severe phase aberration, leading to degraded ultrasound images. This project proposes using deep learning and a real-time pulse-echo ultrasound approach to estimate skull profile and speed of sound, allowing accurate skull aberration correction and establishing the feasibility of the proposed methods.

Using Machine Learning to Predict Eye Movements in Skilled and Unskilled Readers

Jon Willits Jessica Montag Anastasia Stoops Faculty Fellows: Jon Willits
, Jessica Montag
, Anastasia Stoops (Psychology)
NCSA Collaborators: Volodymyr Kindratenko, Eliu Huerta, Dawei Mu

About the Project: Attaining reading proficiency is an issue for both adults and children. To understand the underlying processes of skilled reading, researchers track eye movements that gather visual information efficiently. Typically, people move their eyes while reading and can only clearly see 7-10 letters at a time. This project proposes developing a better deep learning model that integrates visual and linguistic data to predict eye movements. It will help inform reading interventions and education by giving more detailed profiles on skilled and unskilled readers.

Voice Vitals: Novel Infrastructure for Disease Screening and Treatment Tracking

Mary Pietrowicz Faculty Fellow: Mary Pietrowicz (Applied Research Institute)
NCSA Collaborators: Chen Wang, Volodymyr Kindratenko, Colleen Bushell

About the Project: Many disease states, particularly in psychiatry, neurology, and cardiology, are often overlooked in our healthcare systems due to treatment barriers and untimely diagnoses. New disease screening methods are necessary to address these problems. This project proposes developing automated disease screening techniques that can infer clinical states, such as anxiety and manic depressive disorders, using machine learning, modeling, and human speech and language data. The team will integrate models with Clowder to demonstrate the automated annotation of speech/language/health data.

About NCSA

The National Center for Supercomputing Applications at the University of Illinois at Urbana-Champaign provides supercomputing and advanced digital resources for the nation's science enterprise. At NCSA, University of Illinois faculty, staff, students and collaborators from around the globe use these resources to address research challenges for the benefit of science and society. NCSA has been advancing many of the world's industry giants for over 35 years by bringing industry, researchers and students together to solve grand challenges at rapid speed and scale.