Non-invasive drug-release monitoring via AI-enhanced ultrasound in gelatin-magnesium gallate hydrogels
Year: 2026
Type: article
Abstract: • Non-invasive monitoring of drug release from Gelatin-Magnesium Gallate hydrogels • Machine learning enables high-accuracy analysis of hydrogel acoustic signals. • Validated our strategy in varied dispersion medium and cell culture conditions. • Drug release-induced hydrogel degradation resulted in acoustic impedance variation. Hydrogel-based drug delivery systems have been actively explored for precision and personalized medicine, leveraging their localized, sustained, and biocompatible approach for drug administration. However, conventional platforms generally lack real-time monitoring capabilities, as therapeutic evaluation typically relies on invasive or end-point analyses. While stimuli-responsive hydrogels offer controlled release, non-invasive strategies to track in vivo release kinetics is still challenging. We addressed these limitations using an acoustic system coupled with a deep learning framework to analyze ultrasound (US) signal variations from a gelatin-magnesium gallate (GMG) hydrogel, thereby demonstrating the feasibility of non-invasive, time-dependent drug release monitoring. A machine-learning framework quantitatively decodes subtle acoustic variations, enabling non-invasive monitoring with > 95% accuracy. GMG hydrogel was fabricated using highly porous magnesium gallate MOFs (metal–organic frameworks), and the drug (gallic acid) release-mediated structural changes in the hydrogel were observed through US signal variations. Drug release was validated under diverse experimental conditions to ensure reproducibility of the results. Integration of US with AI-based analysis demonstrates the potential for non-invasive, real-time monitoring of drug release from hydrogel-based systems by analyzing the complex pattern of US signals and maintaining hydrogel integrity that enables insight into structural changes during drug release.
Source: Materials & Design
Institution: Pukyong National University
Language: English
Subfield: Molecular Medicine
Domain: Life Sciences
Open Access status: gold
Estimated APC paid: $3,650
Awards: G4601381929, G7077519219
OpenAlex ID: W7160540749
