New metric evaluates kidney transplant survival prediction models

Researchers have developed a new evaluation framework for machine learning models predicting kidney transplant survival. This paired recipient-based approach compares graft outcomes for two recipients sharing the same deceased donor, aiming to assess the benefit of alternative recipient matching. Five different models, including deep learning, achieved approximately 60% accuracy using this new metric, which is argued to be more clinically relevant than the traditional concordance index (C-index). The study highlights limitations of the C-index and proposes the new metric to better reflect real-world organ allocation scenarios. AI

IMPACT Introduces a more clinically relevant evaluation metric for survival prediction models in organ transplantation.

RANK_REASON Academic paper published on arXiv detailing a new evaluation metric for ML models in a specific domain. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• arXiv
• Hugging Face
• Paired Recipient-based Evaluation of Survival Prediction for Deceased Donor Kidney Transplants

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →