The image analysis can rapidly decide which ones have the best transplant success rate!
Researchers at the University of Pennsylvania have developed a new AI system that can select donor organs for transplant with greater accuracy than humans. The system, called “Organ-Specific Matched Evaluation for Transplantation” (OSMET), uses machine learning algorithms to analyze data from donor organs and potential recipients.
Dr. Peter Reese, an associate professor of medicine at the University of Pennsylvania and one of the authors of the study, said, “OSMET is designed to help clinicians select the best organs for each patient based on a range of factors, including blood type, organ size, and the medical history of both the donor and the recipient.”
The system was trained on a dataset of over 15,000 real-life transplant cases, and it generates a score that indicates the likelihood of a successful transplant. In tests, OSMET was able to accurately predict the success rate of each transplant with a high degree of accuracy and outperformed human transplant experts in matching donor organs with suitable recipients.
“We believe that this technology could help reduce the number of organs that are discarded or transplanted suboptimally, which could lead to more successful outcomes for patients in need of a transplant,” said Dr. Roy Bloom, one of the lead authors of the study.
The development of this AI system has significant implications for the field of organ transplantation and has the potential to improve the success rate of transplants, potentially saving more lives. The system can also help clinicians make more informed decisions when selecting donor organs and could reduce the number of organs that are discarded due to mismatching.
As Dr. Reese explained, “Our hope is that this technology will help more patients receive the lifesaving organ transplants they need and improve outcomes for both donors and recipients.”
Via The Impactlab