Whittling down the wait list

Researcher’s AI kidney allocation software gains traction

Nearly 90,000 people in the U.S. are waiting for kidney transplants, yet a large number of kidneys never reach a recipient. Researchers from S&T are reducing that number using artificial intelligence.

Nearly 90,000 people in the U.S. are waiting for kidney transplants, yet a large number of kidneys never reach a recipient. Researchers from S&T are reducing that number using artificial intelligence.

Dr. Cihan Dagli, professor of engineering management and systems engineering; Dr. Richard Threlkeld, who earned a Ph.D. in systems engineering from S&T; and Dr. Lirim Ashiku, research associate at S&T, created AI software to improve kidney allocation. The software assists organ procurement organizations (OPOs) in determining which kidneys will be difficult to place.

The AI software improves the efficiency of matching donor kidneys to potential recipients. The longer a kidney spends on ice waiting to be matched, the greater chance it won’t be placed in time, so it gets discarded. By using OPO data and data from the Organ Procurement and Transplantation Network to model kidney disposition, organizations can more quickly identify those kidneys with a high likelihood for non-use and begin outreach to transplant centers.

Four organizations: Mid-America Transplant, Louisiana Organ Procurement Agency, ARORA (Arkansas) and Lifeshare of Oklahoma are currently integrating the model to improve their kidney use rates.

This lifesaving story began in 2020. Threlkeld, of Orlando, wanted to get an online Ph.D. after earning a master’s degree at Georgia Institute of Technology, so he reached out to Dagli in systems engineering at S&T.

“We were looking for topics of research. That’s when I got thyroid cancer,” Threlkeld says. “I decided that I wanted to do something to help the community. This health care opportunity showed up, and I said I’m going to dive straight in.”

Threlkeld’s cancer was removed, and he said people at S&T were very supportive. Now, he gets bloodwork and imaging done every three months to monitor his health and catch cancer early if it returns.

“It pivoted me toward health care,” Threlkeld says. “I am fortunate to have a second chance.”

Turns out, Dagli has a similar story. Years ago, he was diagnosed with cancer and needed a cornea transplant. Luckily, the cancer is now gone.

“I am on the receiving end of donation, so it is very fulfilling to work on this project,” Dagli says. “That experience led me to this project.”

While pursuing his Ph.D., Threlkeld, Ashiku and Dagli created the first model of the AI software. Threlkeld always wanted to start a business, so from this project he created Valiant AI. He is currently the chief executive officer and Ashiku is the chief technology officer.

The company has an exclusive license to the software, thanks to S&T’s technology transfer and economic development team. Valiant AI can sublicense the software to interested organ procurement organizations across America.

The software is expected to be adopted in all 50 states.

“It’s very unique for S&T to have this successful copyright without having a medical school,” Dagli says. “Soon, we will have to go international with this.”

Threlkeld hopes his work helps create a world where no one waits for a lifesaving organ transplant.

“If this was adopted nationally, I can conservatively say thousands of kidneys would be saved each year.” 

Dr. Richard Threlkeld

“That’s thousands more lives saved. The impact is truly incredible.”People’s perception and preference is informed by a variety of factors, but what’s the messaging we want to give to achieve a certain goal?  We have to understand what people think. We have to understand their pain points and their misconceptions,” Fikru says. “If there are misconceptions, we have to understand their concerns that policymakers would have to address.”