Simulation of tissues for biomedical applications

Nanoporous particles with hydrophobically-modified surface have found applications in many areas of chemical and biomedical research. The understanding of their interactions with surfactants in aqueous solutions opens the opportunities for using such hydrophobic nanoporous particles in drug delivery, tissue phantom preparation and other biomedical fields.

We investigated surfactant-induced wetting of C18-modified silica gel with aqueous solution by means of fluorescence spectroscopy, confocal fluorescence imaging and multi-photon excitation imaging. Ratiometric imaging with solvatochromic probe Prodan revealed the nanotransport of intact surfactant micelles inside the hydrophobic nanopores. The influence of hydrophobic and electrostatic effects on the distribution of fluorescent probe molecules in the C18-particle-surfactant system was addressed by employing positively and negatively charged water-soluble probes, rhodamine 6G and fluorescein, respectively. The methodology for the preparation of surfactant-decorated particles with retained internal hydrophobicity was evaluated by means of confocal fluorescence imaging.

C18-derivatized nanoporous silica particles were incorporated in the design of semi-solid three-dimensional tissue phantoms. Patient-to-patient differences and complexity of the biochemical environments cause significant variation in the spectral characteristics of cancerous tissues. The use of tissue phantoms with controlled boundaries of optical features can assist in the development of optical biopsy methods. The phantoms were composed of hydrophobic nanoporous silica beads loaded with an endogenous tissue fluorophore, flavin adenine dinucleotide (FAD), and imbedded in 1% agar gel. Uniform 10-micron silica particles serve as perfect imaging objects for testing optical biopsy methods, providing features of known dimensions and shapes.

Efficacies of methods for the analysis of autofluorescence tissue spectra were compared. Evaluated methods included: principle peak ratio (PPR), differential normalized fluorescence (DNF), bivariate DNF (2D-DNF), principal component analysis (PCA) and correlation coefficient mapping (CCM). Comparison was achieved by performing diagnosis in a large simulated tissue-phantom data set. The use of simulation in this study ensured the comparison of methods in nearly the entire sample space, while avoiding environmental and patient-to-patient variation biases. The diagnostic boundaries were established by applying Bayes decision theory, which achieved statistically significant classification of the analyzed samples.