VIDEO: International researchers use AI to evaluate health of allograft
Click Here to Manage Email Alerts
Key takeaways:
- Researchers showed in a 5-year study that artificial intelligence can be a valuable tool in detecting signs of allograft failure.
- Protecting allografts can help reduce the impact on a limited organ supply.
In this interview with Healio, Alexandre Loupy, MD, PhD, discusses the results of a 5-year international study looking at the value of artificial intelligence in detecting signs of kidney allograft failure.
“There is an immediate need to diagnose allograft rejection,” Loupy, a professor in nephrology and epidemiology biostatistics at the Necker Hospital in Paris and director of the Paris Institute for Multiorgan Transplant and Organ Regeneration, told Healio. “Rejection represents the major cause of allograft loss. It is a huge burden for our patients and our health care systems, given the organ shortage.”
The research, conducted at 20 transplant centers in Europe and the United States, was funded by the French National Institutes of Health and published in Nature Medicine.
In the study, researchers evaluated the use of algorithms to reclassify diagnoses indicating allograft rejection in adult and pediatric kidney transplant recipients. The review of 4,409 biopsies from 3,054 patients showed the Banff Automation System used by the research team reclassified 83 of 279 (29.75%) antibody-mediated rejection cases and 57 of 105 (54.29%) T cell-mediated rejection cases to a diagnosis of no rejection.
A review of 3,239 (7.32%) biopsies diagnosed as non-rejection by pathologists were identified by the algorithm as rejection in 237 cases.
“This study demonstrates the potential of an automated histological classification to improve transplant patient care by correcting diagnostic errors and standardizing allograft rejection diagnoses,” the authors wrote.
“It is somewhat like a ChatGPT specialized for rejections," Daniel Yoo, MPH, a data scientist at the Paris Institute for Transplantation and Organ Regeneration and co-first author of the study, said in a press release. “We have developed an intelligent and user-friendly system.”
Collapse
Read more about
Click Here to Manage Email Alerts