Machine learning approach for carrier surface design in carrier-based dry powder inhalation

•Combinations of 3 carrier and 3 drug were incorporated in machine learning models.•13 surface parameters were used to describe the carrier surface characteristics.•Effects of key surface roughness variables on DPI performance were elaborated.•Surfaces of highly irregular crevices and orientations a...

Full description

Saved in:
Bibliographic Details
Published in:Computers & chemical engineering 2021-08, Vol.151, p.107367, Article 107367
Main Authors: Farizhandi, Amir Abbas Kazemzadeh, Alishiri, Mahsa, Lau, Raymond
Format: Article
Language:English
Subjects:
Artificial neural network
Dry powder inhalation
Emitted dose
Fine particle fraction
Machine learning
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•Combinations of 3 carrier and 3 drug were incorporated in machine learning models.•13 surface parameters were used to describe the carrier surface characteristics.•Effects of key surface roughness variables on DPI performance were elaborated.•Surfaces of highly irregular crevices and orientations are favorable for high ED.•A balance between pointed peaks and wedged peaks is required to achieve high FPF. In this study, a machine learning approach was applied to evaluate the impact of operating and design variables on dry powder inhalation efficiency. Emitted dose and fine particle fraction data were extracted from the literature for a variety of drug and carrier combinations. Carrier surface properties are obtained by image analysis of SEM images reported. Models combining artificial neural network and genetic algorithm were developed to determine the emitted dose and fine particle fraction. Design strategies for the carrier surface were also proposed to enhance the fine particle fractions. [Display omitted]
ISSN:0098-1354
1873-4375
DOI:10.1016/j.compchemeng.2021.107367