|
KMID : 1146920230530060803
|
|
Journal of Pharmaceutical Investigation
2023 Volume.53 No. 6 p.803 ~ p.826
|
|
|
Recent trends and perspectives of artificial intelligence-based machine learning from discovery to manufacturing in biopharmaceutical industry
|
|
Ravi Maharjan
Lee Jae-Chul
Lee Kyeong
Han Hyo-Kyung
Kim Ki-Hyun
Jeong Seong-Hoon
|
|
|
Abstract
|
|
|
Background : Machine learning (ML) tools have become invaluable in potential drug candidate screening, formulation development, manufacturing, and characterization of advanced drug delivery systems. These tools are part of the Industry 4.0 revolution, which plays a vital role in microparticle and microfluidics, alongside mRNA-LNP vaccines, and stability in advanced protein therapeutics.
Area covered : This study summarizes the application of ML tools in drug discovery, formulation development, and optimization, in addition to continuous manufacturing and characterization of advanced drug delivery systems such as biopharmaceutical formulations including mRNA-LNP vaccines, microfluidics, and microparticle dosage forms. Furthermore, it includes stability concerns, and regulatory, technical, and ethical issues along with future perspectives.
Expert opinion : ML tools are essential for revolutionizing the drug development cycle, where it has been implemented to screen vast databases for drug discovery, optimize formulations, adopt Industry 4.0, and continuous manufacturing concepts, including characterizing and predicting the stability of biopharmaceuticals. However, a gap between regulatory authorities and industries is felt due to current ethical and technical issues in the drug approval process. The vast available databases can be used to train the ML models and such pre-trained ML models can address these concerns. Additionally, these pre-trained tools can predict stability, meaning that the optimization of the formulation is possible, which can save lots of time, efforts, and costs. Moreover, a multidisciplinary approach between ML tools and the drug delivery system promotes digital twin, which can lead to improved patient compliance and efficacy.
|
|
|
KEYWORD
|
|
|
Machine learning, Artificial intelligence, Drug development, Continuous manufacturing, Pharmaceutical application
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|
|