Heyrim Cho

INFORME uses information-theoretic design with mixed-effects modeling to adaptively schedule measurements, reducing the number of scans needed for patient-specific treatment response prediction.

This paper develops a generalized resource-allocation growth law to model prostate cancer responses to intermittent androgen deprivation therapy, showing it predicts treatment outcomes and adapts to different schedules.

The paper develops an agent-based model to predict neural stem cell migration in mouse brains using multi-fiber tractography, showing injection location affects distribution and suggesting factors like chemotaxis may be important.

This study uses a mathematical model to explore combination CAR T-cell treatments for glioblastoma, showing that early intervention and spatially targeted delivery are effective strategies.

This paper develops a hybrid neural network/agent-based model to predict Pramlintide resistance in cutaneous squamous cell carcinoma and enable personalized treatment decisions.