Integrated AI-Driven Systems For Kidney Transplant Precision Medicine (AI-Care)

Introduction

Despite considerable improvements in short-term allograft and patient survival in kidney transplantation in the past two decades, long-term graft survival remains low, with 20% of transplants failing within five years. Rejection remains a leading cause of graft loss, with immediate consequences for patient morbidity and mortality, posing a major burden on healthcare systems.

Challenges in Graft Longevity

Increased allograft longevity requires improved patient monitoring and rejection risk stratification using mechanistically informed, non-invasive biomarker strategies and precision diagnostics.

Limitations

1. Lack of Deeply Phenotyped Cohorts
   The main limitation for such an endeavor is the lack of deeply phenotyped cohorts. Given the complexity and heterogeneity of allograft rejection, patient cohorts with sufficient volume, velocity, variety, and validity of data are critical in order to capture disease activity, disease stage, improve risk stratification, and optimize treatment.

2. Paucity of Multidimensional Research Strategies
   Another limitation is the paucity of multidimensional and integrative research strategies.

Project Goals

The goal of the ERC AI-Care project is therefore to go beyond the current state of the art by establishing a transplant precision medicine system to:

1. Facilitate personalized, non-invasive patient monitoring using cell-free DNA technology (Liquid biopsy).
2. Improve rejection diagnosis beyond conventional histology using molecular technologies and digital pathology.
3. Gain further mechanistic insights and improve diagnostic accuracy of transplant organ diseases and apply this knowledge to new research areas.