Feature importance in multi-dimensional tissue-engineering datasets: Random forest assisted optimization of experimental variables for collagen scaffolds

Ice-templated collagen-based tissue-engineering scaffolds are ideal for controlled tissue regeneration since they mimic the micro-environment experienced in vivo. The structure and properties of fine-tuned during fabrication by controlling a number experimental parameters. However, this parameter space is large complex, rendering interpretation results selection optimal parameters to be challenging practice. This paper investigates impact cross section (drying conditions solute environment) on scaffold microstructure. Qualitative assessment revealed previously unreported drying temperature pressure pore wall roughness, confirmed influence collagen concentration, solvent type, addition morphology. For quantitative comparison, we demonstrate novel application random forest regression analyze multi-dimensional biomaterials datasets, predict microstructural attributes scaffold. Using these models, assessed relative importance input measurements. Collagen concentration pH were found largest factors determining size connectivity. Furthermore, circular dichroism peak intensities also good predictor structural variations, which that has not been investigated its effect Thus, demonstrates potential predictive models such as regressors discover relationships datasets. These between (such spectra connectivity) can therefore used identify design further avenues investigation within biomaterials.