2026-07-06 ブラウン大学
＜関連情報＞
- https://www.brown.edu/news/2026-07-06/ai-drug-delivery
- https://www.sciencedirect.com/science/article/pii/S1773224726006787
物理法則に基づいたニューラルネットワークを用いた薬剤放出モデリング Drug release modeling using Physics-Informed Neural Networks
Daanish Aleem Qureshi, Khemraj Shukla, Vikas Srivastava
Journal of Drug Delivery Science and Technology Available online: 1 July 2026
DOI:https://doi.org/10.1016/j.jddst.2026.108654
Abstract
Accurate modeling of drug release is essential for designing and developing controlled-release systems. Classical models (Fick, Higuchi, Peppas) rely on simplifying assumptions that limit their accuracy in complex geometries and release mechanisms. Here, we propose a novel approach using Physics-Informed Neural Networks (PINNs) and Bayesian PINNs (BPINNs) for predicting drug release from complex geometries using planar, 1D-wrinkled, and 2D-crumpled films as examples. This approach uniquely integrates Fick’s diffusion law with limited and sparse experimental data to enable accurate long-term predictions from short-term measurements, and is systematically benchmarked against classical drug release models. We incorporated Fick’s second law into the PINN framework as a soft constraint being satisfied at randomly sampled collocation points. Previously published sparse experimental datasets are then used to evaluate drug release performance using mean absolute error (MAE) and root mean square error (RMSE), accounting for noisy conditions and limited data availability. Our approach reduced mean error by up to 40% relative to classical baselines across all film types. The PINN formulation achieved RMSE ＜0.05 utilizing only the first 6% of the release time data (reducing 94% of release time required for the experiments) for the planar film. For wrinkled and crumpled films, the PINN reached RMSE＜0.05 in 33% of the release time data. BPINNs provide tighter and more reliable uncertainty quantification under noise. By combining physical laws with experimental data, the proposed framework yields highly accurate long-term release predictions from short-term measurements, offering a practical route for accelerated characterization and more efficient early-stage drug release system formulation. Unlike conventional machine learning approaches, which often lack clear theoretical convergence guarantees, PINNs provide convergence assurances for broad classes of elliptic and parabolic equations. With Fick’s diffusion equation as parabolic, our proposed PINNs based method offers a theoretically consistent and stable framework for modeling drug diffusion processes. The codes are available at https://github.com/SrivastavaResearchLab/drug_release_modeling_PINN-2025
