The broader impact of this Small Business Technology Transfer (STTR) Phase I project will be to improve solid organ transplantation outcomes. Few significant clinical or technological advancements have been made within the last two decades to improve organ matching success, and the accuracies of current models predicting survival outcomes are diminishing. There is a great need for clinicians to have better decision-making support tools. Every 12 minutes, a new person is added to the organ transplant waiting list, a number growing by about five percent each year. Within a single day, 21 people die waiting for a kidney, liver, or other organ match. Although 36,500 kidney and liver transplants are performed each year, the patient demand for donor organs far outweighs supply by four to one, so the need to improve donor-recipient matching is urgent. Optimizing the donor organ-patient match is a key determining factor for improving transplant success and patient survival. This project's artificial intelligence (AI) model will guide transplant surgeons, physicians, and other healthcare professionals will deliver precise, accurate, quantitative information for real-time predictions.
This Small Business Technology Transfer (STTR) Phase I project proposes artificial intelligence to predict outcomes after solid organ transplantation procedures. Clinicians currently consider several factors when determining organ allocation and candidate patient ranking on the recipient waitlist, including extent of disease pathology, functional status of the recipient, and intrinsic donor and recipient compatibility factors. Measures, indices and functional status scores have been designed to predict specific outcomes but are not easily combined into one optimized decision to guide organ allocation decisions. To date, no organ-matching predictive outcome model has comprehensively synthesized all available patient- and donor-specific variables at the time of transplantation. This project will train an artificial intelligence (AI) algorithm to comprehensively integrate all information available at the time of transplantation procedures (hundreds of variables) into a predictive model. An AI model of this nature would be a substantial improvement from linear models able to synthesize only a modest number of parameters (approximately 15) to date. It is expected that the proposed technology will predict both pre- and post-transplant survival more accurately than currently accepted models.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
