Optimal design of micromixer for preparation of nanoliposomes

Leveraging machine learning to streamline the development of liposomal drug delivery systems

“…These variables have relevant influences on the process and uniquely contribute to the formation and characteristics of liposomes. 22,24,52,53 Lipid molecular descriptors provide insights into the physicochemical properties of the lipids, influencing their behaviour in the microfluidic process. While formulation parameters have a direct impact on circulation and drug retention, chip geometry affects the mixing and assembly of lipids, influencing liposome size and uniformity.…”

“…Chips with different channel geometry were 3D printed and utilized in the production offering flexibility and independence at low cost (Supporting information Fig S1 ). 24 Immediately following liposome production, the sample underwent analysis via dynamic light scattering (DLS) 52 and the hydrodynamic diameter and the polydispersity index (PDI) of the liposomes were chosen as the primary response for optimization in this study. 6 The data set obtained from the lipid screening was analysed using the Tree map (TMAP) dimensionality reduction method (Fig.…”

“…22,24 Previous experiments have shown that total flow rate (TFR) becomes less important above a certain high turbulent flow, 27 where its influence on liposome formation is minimal which is supported by simulations elsewhere. 27,52 Therefore, the TFR was excluded from the initial feature vector after ensuring to be above the threshold of the used chips (>8 mL/min). A well-constructed feature vector (Fig.…”

“…15 Furthermore, variations in microfluidic chip design complicate the standardization of production processes. 52 Robustly 3D-printing these chips might offer a pathway to standardization, facilitating broader application in pharmaceutical manufacturing. 24 Additionally, microfluidic production depends on many controllable parameters (Formulation, Chip, FRR, etc.…”