| Abstract Scope |
Riboflavin alleviates migraine symptoms by reducing neuroinflammation. Many migraine-relief nutraceuticals can be purchased over the counter. However, while most pharmaceutical products are highly regulated, supplements have less stringent regulatory assurances.
Multi-scale characterization of these supplements was used to understand the spatial distribution and morphology of components and demonstrate quality assurance in different riboflavin supplements.
Novel 3D techniques were used to characterize two capsules presenting very different efficacy. Nanoscale X-ray microscopy was used to image the powders non-destructively, applying advanced machine learning reconstruction and phase classification approaches to enhance visualization, phase contrast, and throughput. Correlative electron microscopy and EDS provided high resolution 2D quantification.
Machine learning methods significantly improve data quality. Complementary 3D automated phase classification identified binders, anti-caking agents and fillers.
Thus, multiscale nanoimaging and AI-based image processing revealed previously unobservable phases, quantifying previously unobservable undeclared components, demonstrating capability for quality assurance and regulation of supplements. |