Yiming Cheng

This paper introduces a Cauchy likelihood residual model for CAR-T cellular kinetics, showing it improves robustness and physiological plausibility compared to Gaussian and Student's t models.

This paper introduces an AI-assisted method to implement spatial PDE models in NONMEM, enabling efficient and transparent spatial pharmacometric analysis of drug distribution in tumors.

This paper develops a delay differential equation model for CAR-T cell kinetics, showing it better fits longitudinal data and supports delayed conversion from effector to memory cells.

This paper evaluates heavy-tailed residual distributions for outlier handling in population PK modeling, finding that the Student's t distribution provides more stable inference than exponential-tail alternatives under extreme contamination.

This study evaluates Bayesian modeling strategies for population pharmacokinetics, finding that Students t-based M3 censoring improves accuracy and precision in handling outliers and censored data.