Factorial Data-Driven Inverse Design of Granular Hydrogels for Targeted Therapeutic Release

Yasha Saxena, Po-An Lin, Jay Shah, Tracy Asamoah, Arthi Jayaraman, Gaurav Arya, Tatiana Segura

Keywords: inverse design; drug delivery; biomaterials; factorial experimental design

TL;DR: We demonstrate a data-efficient simulation approach to train an inverse design framework for biomaterials with targeted therapeutic release rates.

Abstract: Grandular hydrogels enmeshed with therapeutic particles offer an exciting modular platform for the delivery of targeted therapeutics, but this modularity also complicates the optimization of the design. Here, we present a programmable therapeutic release simulation for this material platform. Using factorial experimental design, we efficiently validate simulation parameters and identify a practical design space that supports precision medicine through the inverse design of unique and customizable drug release profiles, including tunable cumulative release profiles through random packing and tunable instantaneous release profiles through layered packing.

Submission Track: Paper Track (Short Paper)

Submission Category: AI-Guided Design

Institution Location: Durham, North Carolina

AI4Mat Journal Track: Yes

AI4Mat RLSF: Yes

Submission Number: 74