Transformer-Based Deep Learning Framework for Predicting Stimuli-Responsive Release Profiles in Nano-Drug Delivery Systems
International Journal of Drug Delivery Technology · 30.05.2026 · 16 (38s)
DOI: 10.25258/ijddt.16.38s.44 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Drug Delivery Technology, online erschienen am 30.05.2026. Verfasst von Venkateswaran Govindarajulu und 1 weiteren Autor:innen.
Abstract (Originalsprache)
The design and optimization of stimuli-responsive nano-drug delivery systems (NDDS) present significant challenges, primarily because the release behaviour of encapsulated drugs varies dynamically based on environmental triggers such as local pH gradients, temperature fluctuations, and redox potential. Conventional computational methods— including empirical pharmacokinetic modelling and traditional machine learning regression—lack the capacity to capture the complex, non-linear temporal dependencies inherent in such release profiles. This study introduces a Transformer-based deep learning framework specifically designed to predict cumulative drug release kinetics from stimuli-responsive nanocarriers, including polymeric nanoparticles, liposomes, chitosan-based systems, and dendrimers. The framework encodes drug physicochemical descriptors alongside stimuli parameters through a multihead self-attention mechanism, enabling the model to learn cross-dimensional interaction patterns not previously captured by sequential models. Using a curated dataset of eight drugs including Doxorubicin, Paclitaxel, Insulin, and Curcumin across pH-responsive, thermo-responsive, and redox-responsive systems, the proposed model achieved an R² score of 0.941, mean absolute error (MAE) of 0.082, and an overall prediction accuracy of 94.2%, outperforming six comparative baseline models including Long Short-Term Memory (LSTM) networks and Graph Neural Networks (GNNs). These findings indicate that Transformer architectures are well-suited for accelerating the computational screening of intelligent drug delivery formulations, with potential for integration into clinical decision support platforms.
Zitieren
APA
Govindarajulu, V., & K, L. (2026). Transformer-Based Deep Learning Framework for Predicting Stimuli-Responsive Release Profiles in Nano-Drug Delivery Systems. International Journal of Drug Delivery Technology, 16(38s). https://doi.org/10.25258/ijddt.16.38s.44
Vancouver
Govindarajulu V, K L. Transformer-Based Deep Learning Framework for Predicting Stimuli-Responsive Release Profiles in Nano-Drug Delivery Systems. International Journal of Drug Delivery Technology. 2026;16(38s). doi:10.25258/ijddt.16.38s.44
BibTeX
@article{Govindarajulu2026,
author = {Govindarajulu, Venkateswaran and K, Lokesh},
title = {Transformer-Based Deep Learning Framework for Predicting Stimuli-Responsive Release Profiles in Nano-Drug Delivery Systems},
journal = {International Journal of Drug Delivery Technology},
year = {2026},
volume = {16},
number = {38s},
doi = {10.25258/ijddt.16.38s.44}
}
EUVIMED