Project aims to bring cancer data to the bedside


cancer data


Jacob Luber is on a mission to improve cancer treatment by collecting and analyzing vast amounts of cancer data. The computer science assistant professor earned a five-year, $2 million grant from the Cancer Prevention and Research Institute of Texas to create a database that contains every publicly available cancer dataset from the National Cancer Institute, with the goal of enabling researchers and physicians to map where cancer patients have similar traits and improve and expand treatments based on that information. Such datasets are currently too large for physicians to access easily.

To do so, Dr. Luber is using deep-learning techniques to explore how to reduce the size of extremely large datasets from up to three petabytes to a more-manageable 12 terabytes. (One petabyte contains 1,000 terabytes.) A recent paper accepted at the 2023 IEEE International Symposium on Biomedical Imaging led by two members in Luber’s lab, Amir Hajighasemi and Mohammad Nasr, implemented this impressive compression ratio.

“With deep learning, we can take a high-dimensional image with proteomic data overlaid on top and compress it, then use it to find similarities between patients,” says Luber, who is also a member of the University’s Multi-Interprofessional Center for Health Informatics. “Once those similarities are identified, physicians could mine electronic health records to see what treatments were most effective and adjust treatment regimens accordingly.”

This research is already making an impact. The algorithm developed by Amir and Mohammad in their IEEE paper allows doctors to input patient data and access the larger index through a computer at the patient's bedside or in an exam room. Visiting Assistant Professor Helen Shang, a medical doctor, is working to start clinical trials using these algorithms for real-world applications.

Another lab member, Jillur Rahman Saurav, is making strides in the field of generative AI. His work uses generative models to synthesize medical images. Many have tried chatGPT, which is actually based on this type of algorithm, which uses AI to create realistic images or text based on the data it has been trained on. In Saurav’s work, AI is used to create medical images that help doctors diagnose and treat cancer more effectively, particularly in rural contexts where resources may be limited. Based on this work, Saurav was recently selected for the highly competitive 2023 Google CS Research Mentorship Program, which matches students from historically marginalized groups with peers and a Google mentor to support their pursuit of computing research pathways.

The groundbreaking work being done in Luber’s lab has the potential to revolutionize cancer treatment for patients worldwide. As the lab continues to innovate and develop new algorithms, they will help bring personalized and effective cancer care to more people in need.


Rick Wilder

Investigating Earth’s ionosphere-thermosphere

A space physicist at UTA is playing a key role in NASA’s Geospace Dynamics Constellation mission aimed at improving our understanding of the planet’s ionosphere-thermosphere (I-T) system.

Kyeong Rok Ryu

Quantifying the benefits of roadside vegetation

Construction management Assistant Professor Kyeong Rok Ryu is helping cultivate a better-looking Texas by creating best practices for roadside vegetation.

Illustration by Jing Jing Tsong

Evaluating smart traffic signal design

Civil engineer Pengfei “Taylor” Li hopes to make traffic signals smarter with new simulation techniques and big data.

Inquiry Magazine 2023

See More