In Lithuania, as in the rest of Europe, there is a shortage of organs suitable for transplantation, and therefore the use of machine learning provides hope that the forecasts for those waiting for a transplant will become more reliable and the overall procedure will be smoother. Transplantation of solid organs is a unique and complex branch of surgery. The success of the operation depends on the complex and multifaceted relationship of the donor, the preparation of the recipient waiting for the operation, and the variables of the surgical procedure. Machine learning models have been proposed for more accurate preoperative diagnosis and postoperative prognosis. Three teams joined forces to create the model planned in Lithuania: the VU Faculty of Mathematics and Informatics, Institute of Data Science and Digital Technologies, and VU Faculty of Medicine in partnership with the VULSK Organ Transplantation Coordination Center[ALH3] .
Prof. Gintaras Dzemyda.Personal archive.
"Our goal is to create an artificial intelligence model that will help the transplant team make an optimal decision regarding organ suitability for transplantation", Professor Gintautas Dzemyda said.
"Collaboration between faculties allows a broader view of the problem and creates the conditions for new research and inventions. I always encourage interdisciplinary work of specialists in introducing the latest innovations to improve the quality of patients' health," says Prof. D. Jatužis, Dean of VU Faculty of Medicine.
Prof. Dalius Jatužis. MF archive.
The solution that is created will be implemented in clinical practice and freely accessible to transplant teams not only in Lithuania, but also all around the world.
"Machine learning is a unique opportunity for a computer to learn to make the best decisions by applying the experience accumulated by doctors. The ability to single out the factors with the greatest influence on the donor and the recipient for a successful transplant is important as well. This is a big step forward in evaluating preoperative preparation and postoperative care, improving quality indicators, and preventing complications", Associate Professor Gulla , who is doing a postdoctoral internship at the VU Faculty of Mathematics and Informatics, said. "I am proud to have the opportunity to work with a top-level machine learning team and artificial intelligence specialists at VU's Faculty of Mathematics and Informatics and to gain new knowledge in a world-class centre: Memorial Sloan Kettering Cancer Center located in New York," she said.
Aistė Gulla, PhD. Personal archive.
The models developed during the project will undoubtedly improve the accuracy of predictions concerning the success of the transplantation process, allow the selection [ALH7] of the key factors affecting morbidity and mortality after transplantation, and optimise the process of organ allocation to patients waiting for transplantation.
The project "Machine Learning Application to Determine Preoperative Feasibility and Postoperative Outcomes of Liver Transplantation" is funded by LMT, no. P-PD-22-099