NCSA Student Spotlight: Arjun Gupta

12.10.19 -

By Shaylee Bent (NCSA)

Noise can often disrupt the effectiveness and signal of an electrocardiogram (ECG), a test used to measure the electrical activity of the heart. With the added noise, the process of interpreting the signal and consequently diagnosing patients becomes that much harder. Arjun Gupta, an undergraduate research assistant at the National Center for Supercomputing Applications (NCSA), has been working on a solution.

Using his research to eliminate noise in the search for gravitational waves billions of miles away, Gupta was able to apply the same techniques to improve ECG results, transcending from the galactic to human scale.

Gupta, currently a junior in computer engineering at the University of Illinois at Urbana-Champaign, has developed deep learning algorithms that have proven effective at removing non-Gaussian and non-stationary noise from gravitational wave signals that outperform existing methods to denoise and classify time-series data of ECG's.

Along with his deep learning algorithms, Arjun and his collaborators in the Gravity Group at NCSA worked to develop a model capable of detecting myocardial infarctions based on the ECG signal of a patient.

“Arjun has quantified what leads are the most relevant to correctly identify heart conditions and has also developed a neural network model that can take in multiple leads to boost its prediction accuracy, which currently represents the state-of-the-art in this field. His innovative work is a testament to the interdisciplinary AI research program conducted at the Center for Artificial Intelligence Innovation” said Eliu Huerta, director of the Center for Artificial Intelligence Innovation and lead of the Gravity Group.

Myocardial infarctions, also known as heart attacks, occur when there’s blockage in at least one of the coronary arteries. According to the Center for Disease Control and Prevention, the leading cause of death worldwide is heart disease, making this research especially relevant.

“Amongst patients with cardiovascular diseases, myocardial infarction is the main cause of death. Thus, detection of myocardial infarction in a timely manner is a serious challenge with a significant potential for impact,” said Gupta.

“If we are able to further increase the accuracy of our model, the model could potentially be deployed to aid cardiologists in their work,” continued Gupta.

Arjun’s interest in deep learning started when he attended a symposium at NCSA in which the main discussion was centered around the uses for deep learning across various disciplines. He then reached out to Daniel George, a then-member of the Gravity group and current Research Scientist at Google X. George spoke about the Gravity Group’s use of deep learning to de-noise gravitational waves, which intrigued Gupta, and he’s been a part of the group ever since.

“Arjun started his journey in the Gravity Group developing deep learning algorithms to remove noise anomalies from gravitational wave signals produced by the collision of black holes. He then started exploring the applicability of such algorithms to other disciplines, including cardiology,” said Huerta.

This ever-important crossover is yet another example of what the Center for Artificial Intelligence Innovation enables - researchers using a variety of AI methods to address challenges regardless of disciplinary boundaries.

“My experience with the group has been incredible,” said Gupta. Arjun is especially thankful for how understanding, caring, and patient mentors like Dr. Eliu Huerta have been to him. He’s also grateful for how his co-members have been willing to sacrifice their time, even on the weekends, to help him out.

In July of this year, Arjun attended the Complex Interventional Cardiovascular Therapy Symposium in Chicago, which features open forum discussions surrounding cardiovascular topics, such as methods to utilize in minimizing and handling complications during surgery.

In the future, Gupta plans on continuing to work with the Gravity Group on a multitude of problems. “Members of the Gravity Group work on very interesting problems with real significance,” Gupta said, “be it directly through potentially improving patient care, or more indirectly through advancing the field of physics.”

“Such a collaborative environment and such supportive people are reasons why I've been with the same group since freshman year and I plan on being with the group until the end of my time at Illinois,” said Gupta.

Gupta's research is supported by NCSA's Student Pushing Innovation (SPIN) Internship Program. He is also a recipient of Fiddler Innovation Fellowship at NCSA.

About NCSA

The National Center for Supercomputing Applications (NCSA) 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 advanced digital resources to address research grand challenges for the benefit of science and society. NCSA has been advancing one third of the Fortune 50® for more than 30 years by bringing industry, researchers, and students together to solve grand challenges at rapid speed and scale.