The Prediction of the In Vitro Release Curves for PLGA-Based Drug Delivery Systems with Neural Networks DOI Creative Commons
Zheng Zhang,

Bolun Zhang,

Chen Ren

и другие.

Pharmaceutics, Год журнала: 2025, Номер 17(4), С. 513 - 513

Опубликована: Апрель 14, 2025

Background/Objectives: The accurate prediction of drug release profiles from Poly (lactic-co-glycolic acid) (PLGA)-based delivery systems is a critical challenge in pharmaceutical research. Traditional methods, such as the Korsmeyer-Peppas and Weibull models, have been widely used to describe vitro kinetics. However, these models are limited by their reliance on fixed mathematical forms, which may not capture complex nonlinear nature behavior diverse PLGA-based systems. Method: In response limitations, we propose novel approach—DrugNet, data-driven model based multilayer perceptron (MLP) neural network, aiming predict data at unknown time points fitting curves using key physicochemical characteristics PLGA carriers molecules, well data. We establish dataset through literature review, trained validated determine its effectiveness predicting different curves. Results: Compared traditional Korsmeyer–Peppas semi-empirical MSE DrugNet decreases 20.994 1.561, respectively, (R2) increases 0.036 0.005. Conclusions: These results demonstrate that has stronger ability fit better relationships It can deal with change better, adaptability advantages than overcomes limitations expressions models.

Язык: Английский

Nanocarriers and their Integrated Microneedle Systems-Mediated Drug Delivery for the Treatment of Moderate-Severe Dermatological Diseases: Recent Progress, Applications and Future Perspectives DOI
Mridusmita Das, Rabinarayan Parhi

Journal of Drug Delivery Science and Technology, Год журнала: 2025, Номер 106, С. 106748 - 106748

Опубликована: Фев. 22, 2025

Язык: Английский

Процитировано

0
Novel strategies in topical delivery for psoriasis treatment: nanocarriers and energy-driven approaches DOI
Cheng-Yu Lin, Zih‐Chan Lin,

Yen-Tzu Chang

и другие.

Expert Opinion on Drug Delivery, Год журнала: 2025, Номер unknown

Опубликована: Фев. 25, 2025

Psoriasis is characterized by abnormal differentiation and hyperproliferation of epidermal keratinocytes. This condition presents significant challenges for effective drug delivery. In addition to overcoming the thickness skin, topical treatments must navigate complex hydrophobic hydrophilic properties skin barrier. Recent advancements in nanocarrier technologies, including energy-driven methods microneedles that penetrate stratum corneum, present promising strategies enhancing permeation through tailored physicochemical properties. A literature search was performed using databases Google Scholar, PubMed, ScienceDirect. review highlights recent studies on novel delivery psoriasis treatment, addressing current therapeutic options their limitations. We provide a comprehensive overview chemical nanoformulations explore physical improve rates. Furthermore, we discuss advantages various formulations can carry different types payloads, offering patients diverse symptom management. The covers conventional treatments, emphasizing nanoparticle design macromolecular drugs. includes Ribonucleic acid (RNA)-based therapies protect drugs from rapid clearance body. argue intelligent approaches enhance efficacy across applications while allowing precision treatment strategies, ultimately improving patient outcomes.

Язык: Английский

Процитировано

0
The Prediction of the In Vitro Release Curves for PLGA-Based Drug Delivery Systems with Neural Networks DOI Creative Commons
Zheng Zhang,

Bolun Zhang,

Chen Ren

и другие.

Pharmaceutics, Год журнала: 2025, Номер 17(4), С. 513 - 513

Опубликована: Апрель 14, 2025

Background/Objectives: The accurate prediction of drug release profiles from Poly (lactic-co-glycolic acid) (PLGA)-based delivery systems is a critical challenge in pharmaceutical research. Traditional methods, such as the Korsmeyer-Peppas and Weibull models, have been widely used to describe vitro kinetics. However, these models are limited by their reliance on fixed mathematical forms, which may not capture complex nonlinear nature behavior diverse PLGA-based systems. Method: In response limitations, we propose novel approach—DrugNet, data-driven model based multilayer perceptron (MLP) neural network, aiming predict data at unknown time points fitting curves using key physicochemical characteristics PLGA carriers molecules, well data. We establish dataset through literature review, trained validated determine its effectiveness predicting different curves. Results: Compared traditional Korsmeyer–Peppas semi-empirical MSE DrugNet decreases 20.994 1.561, respectively, (R2) increases 0.036 0.005. Conclusions: These results demonstrate that has stronger ability fit better relationships It can deal with change better, adaptability advantages than overcomes limitations expressions models.

Язык: Английский

Процитировано

0