Multiple linear regression applied to predicting droplet size of complex perfluorocarbon nanoemulsions for biomedical applications

Abstract

Multiple linear regression (MLR) modeling as a novel methodological advancement for design, development, and optimization of perfluorocarbon nanoemulsions (PFC NEs) is presented. The goal of the presented work is to develop MLR methods applicable to design, development, and optimization of PFC NEs in broad range of biomedical uses. Depending on the intended use of PFC NEs as either therapeutics or diagnostics, NE composition differs in respect to specific applications (e.g. magnetic resonance imaging, drug delivery, etc). PFC NE composition can significantly impact on PFC NE droplet size which impacts the NE performance and quality. We demonstrated earlier that microfluidization combined with sonication produces stable emulsions with high level of reproducibility. The goal of the presented work was to establish correlation between droplet size and composition in complex PFC-in-oil-in-water NEs while manufacturing process parameters are kept constant. Under these conditions, we demonstrate that MLR model can predict droplet size based on formulation variables such as amount and type of PFC oil and hydrocarbon oil. To the best of our knowledge, this is the first report where PFC NE composition was directly related to its colloidal properties and MLR used to predict colloidal properties from composition variables.

Graphical Abstract

Keywords:

• Multilinear regression
• modeling
• perfluorocarbons
• colloids
• nanoemulsions

Acknowledgements

The authors acknowledge Shikhar Mohan and Yuxiang Zhao for technical insight in statistical analysis.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was in part supported by National Institute of Biomedical Engineering and Imaging award R21EB023104-02 and AFMSA Department of Defense Award FA8650-17–2-6836.