ECE faculty among recipients of 2021-2022 NCSA Fellowships

8/9/2021 Illinois ECE

Research Assistant Professor Aiguo Han and Assistant Professor Pengfei Song were both named NCSA Fellows, which will provide funding for research projects.

Written by Illinois ECE

Two Illinois ECE faculty members were among the University of Illinois Urbana-Champaign researchers selected as National Center for Supercomputing Applications (NCSA) Fellows for 2021-2022. Research Assistant Professor Aiguo Han and Assistant Professor Pengfei Song submitted winning research proposals that the fellowship program will fund.

According to the website, "NCSA's Fellowship program provides an opportunity for faculty and researchers at the University of Illinois at Urbana-Champaign to catalyze and develop long-term research collaborations between Illinois departments, research units, and NCSA. This competitive program provides seed funding for new projects that include NCSA staff as integral contributors to the project." Descriptions of their projects, taken from the NCSA website, are provided below.

Aiguo Han's Proposal

Aiguo Han, Research Assistant Professor, ECE ILLINOIS
Aiguo Han, Research Assistant Professor, ECE ILLINOIS


College: Grainger College of Engineering
Award year: 2021-2022
NCSA collaborators: Volodymyr Kindratenko

Transcranial ultrasound could enable a broad variety of applications in imaging, including functional brain imaging, intracerebral hemorrhage detection, brain perfusion evaluation, and cerebrovascular disease (e.g., stroke) diagnosis, among others. Despite the great potentials, transcranial ultrasound brain imaging has not been widely used in adults. This is in large part because adult human skulls cause severe phase aberration, leading to highly degraded ultrasound images. The research team proposes a real-time pulse-echo ultrasound approach to estimate the skull profile and speed of sound using deep learning methods with ultrasound radiofrequency signals backscattered from the skull. The hypothesis is that with sufficient training, deep learning is capable of extracting the skull profile and speed of sound from radiofrequency signals, and deep learning-extracted skull profile and speed of sound allow accurate skull aberration correction for transcranial ultrasound imaging. The objective of this project is to establish the feasibility of the proposed methods.

Pengfei Song's Proposal

Pengfei Song
Pengfei Song, Assistant Professor


College: Grainger College of Engineering
Award year: 2021-2022
NCSA collaborators: Peter Groves, Colleen Bushell

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. Currently, capturing an image using ULM takes a prohibitive amount of time, and consequently, the technology has not been deployed in a clinical setting. To address this problem, the team is exploring the use of deep learning to more efficiently utilize the microbubble signal to shorten the data acquisition time and post-processing. However, deploying deep learning for microbubble signal processing in ULM requires a large amount of labeled data for neural network training, which is difficult to accomplish in an experimental setting. Relying on synthesized data is a viable alternative solution for DL, but it requires in vivo microvascular graph models to generate realistic ultrasound simulation data for contrast-enhanced blood flow. Such a database of in vivo microvascular graph models does not currently exist, which the team is proposing to develop in this project. Successful completion of this project will produce a large set of labeled micro vessel data that can be shared with the ULM research community to facilitate the development of fast ULM techniques to ultimately achieve the clinical translation of ULM.

Share this story

This story was published August 9, 2021.