AI Model Combines EKG and Blood Tests to Assess Heart Transplant Rejection

Researchers at NYU Langone Health said in a press release that an AI model combining electrocardiogram readings with two blood tests could help identify heart transplant rejection without a biopsy in some cases. The study was published online September 25 in the Journal of Heart and Lung Transplantation.

The researchers trained models using 5,300 EKG readings from 2,357 adult heart transplant recipients. One model used EKG data alone, while another combined EKG readings with blood tests that measure gene activity linked to cellular rejection and fragments of donor DNA in a recipient's bloodstream. The team compared model results with biopsy records.

In a separate test group of 38 transplant recipients, the combined model correctly identified 94 percent of patients who were not experiencing rejection. Blood tests alone incorrectly identified 19 patients as potentially needing biopsies. The researchers plan to test the combined model with more patients across several transplant centers.