Supervised Machine Learning Algorithms for Evaluation of Solid Lipid Nanoparticles and Particle Size.

Area of Science:

• Pharmaceutical Nanotechnology
• Materials Science
• Computational Chemistry

Background:

• Solid Lipid Nanoparticles (SLNs) are advanced drug delivery systems offering controlled release and stability.
• Particle Size (PS) is a critical parameter influencing SLN efficacy, affecting drug release and bio-distribution.
• Optimizing SLN formulation requires understanding the impact of processing parameters and ingredients on particle characteristics.

Purpose of the Study:

• To evaluate the formulation of SLNs using high-speed homogenization.
• To model the influence of mixing time and formulation ingredients on SLN particle size (PS).
• To assess the predictability of PS using various machine learning algorithms.

Main Methods:

• SLNs were prepared via high-speed homogenization.
• Particle size, size distribution, and zeta potential were measured.
• Machine learning algorithms were applied to a dataset of SLN formulations to predict PS.

Main Results:

• SLN particle sizes ranged from 263-498nm.
• Decision tree-based methods, specifically Random Forest, provided the most accurate PS estimations (MAE=0.028).
• Machine learning algorithms demonstrated the capability to estimate PS using both rule-based and function-fitting approaches.

Conclusions:

• Machine learning methods are highly effective for predicting SLN particle size.
• These findings highlight the utility of ML in optimizing SLN formulation parameters for future research.
• The study validates ML's role in pharmaceutical nanotechnology for efficient material design.