Proceedings of the
9th International Conference of Asian Society for Precision Engineering and Nanotechnology (ASPEN2022)
15 – 18 November 2022, Singapore
doi:10.3850/978-981-18-6021-8_OR-08-0147
Prediction of Drug Permeation Through Human Skin Treated with Microneedles Using Machine Learning Methods
1School of Pharmacy, Faculty of Medicine and Health, University of Sydney, NSW, 2006, Australia
2Department of Clinical Pharmacy and Pharmacy Administration, School of Pharmacy, Fudan University, Shanghai, 201203, China
3Harvard T.H. Chan School of Public Health, 677 Huntington Avenue Boston, MA 02115, USA
4National Healthcare Group, 1 Fusionopolis Link, Singapore 138542, Republic of Singapore
5Procter & Gamble, 70 Biopolis St, Singapore 138547, Republic of Singapore
6MGI Biotechnology, 21 Biopolis Rd, Nucleos, Singapore 138670, Republic of Singapore
ABSTRACT
Microneedles are sharp protrusions, which can penetrate the stratum corneum, the utmost layer of skin preventing external substances from getting into human body. Then drugs can passively diffuse through skin and are absorbed into blood circulation. The process of measuring the amounts of drugs permeated through skin is challenging yet crucial for transdermal drug delivery. In vitro drug permeation testing is a commonly used method, but it is costly and time-consuming. To this end, the simulation method may be a better alternative to predict the drug permeation than experimental testing. For in vitro drug permeation testing, Fick's law is used find the diffusion coefficient, which is hard to determine. This paper proposes to apply machine learning methods to predict the rate of drug permeation through skin, circumventing the process of measuring diffusion coefficient with experiments. XGBoost and Random Forest methods are used and compared with conventional methods. The results showed that XGBoost was the best method to predict of drug permeation through skin. It was found that drug loading, permeation time and microneedle surface area are critical parameters in the prediction models. With the prediction models enabled by machine learning, microneedles could be tailored to deliver prescribed drug dose through skin for precision medication.
Keywords: Microneedle, Transdermal, Machine learning, XGBoost, Random Forest, Drug delivery
1School of Pharmacy, Faculty of Medicine and Health, University of Sydney, NSW, 2006, Australia
2Department of Clinical Pharmacy and Pharmacy Administration, School of Pharmacy, Fudan University, Shanghai, 201203, China
3Harvard T.H. Chan School of Public Health, 677 Huntington Avenue Boston, MA 02115, USA
4National Healthcare Group, 1 Fusionopolis Link, Singapore 138542, Republic of Singapore
5Procter & Gamble, 70 Biopolis St, Singapore 138547, Republic of Singapore
6MGI Biotechnology, 21 Biopolis Rd, Nucleos, Singapore 138670, Republic of Singapore
ABSTRACT
Microneedles are sharp protrusions, which can penetrate the stratum corneum, the utmost layer of skin preventing external substances from getting into human body. Then drugs can passively diffuse through skin and are absorbed into blood circulation. The process of measuring the amounts of drugs permeated through skin is challenging yet crucial for transdermal drug delivery. In vitro drug permeation testing is a commonly used method, but it is costly and time-consuming. To this end, the simulation method may be a better alternative to predict the drug permeation than experimental testing. For in vitro drug permeation testing, Fick's law is used find the diffusion coefficient, which is hard to determine. This paper proposes to apply machine learning methods to predict the rate of drug permeation through skin, circumventing the process of measuring diffusion coefficient with experiments. XGBoost and Random Forest methods are used and compared with conventional methods. The results showed that XGBoost was the best method to predict of drug permeation through skin. It was found that drug loading, permeation time and microneedle surface area are critical parameters in the prediction models. With the prediction models enabled by machine learning, microneedles could be tailored to deliver prescribed drug dose through skin for precision medication.
Keywords: Microneedle, Transdermal, Machine learning, XGBoost, Random Forest, Drug delivery