Added value of T1 mapping derived phenotypes and polygenic risk score (PRS) for the prediction of common cardiac diseases

Motivation: Limited attention has been given to the evaluation of the predictive capacity of T1 mapping in the context of cardiac diseases in the present. Goal(s): To improve clinical diagnostic precision by integrating T1 value of different AHA segments into the prediction model and find the important AHA segments for cardiac diseases prediction. Approach: Our study employed nnU-Net to segment T1 mapping images, then to quantify T1 values of AHA segments to establish a hybrid prediction model for some common cardiac diseases. Results: The results demonstrate that the incorporation of the hybrid prediction model with T1 mapping leads to enhanced performance. Impact: The added value of T1 mapping enhanced the performance of the common cardiac disease prediction model. It empowered clinicians to identify potential cardiac issues earlier and making clinicians pay more attention to certain AHA segments of the cardiac diseases.