Semisolid Pharmaceutical Product Characterization Using Non-invasive X-ray Microscopy and AI-Based Image Analytics

This work reports the use of X-ray microscopy (XRM) imaging to characterize the microstructure of semisolid formulations containing multiple immiscible phases. For emulsion-based semisolid formulations, the disperse phase globule size and its distribution can be critical quality attributes of the product. Optical microscopy and light diffraction techniques are traditionally used to characterize globule size distribution. These techniques are subjected to sample preparation bias and present challenges from matrix interference and data processing. XRM imaging is an emergent technique that when combined with intelligent data processing has been used to characterize microstructures of pharmaceutical dosage forms including oral solid formulations, controlled release microspheres, and lyophilized products. This work described our first attempt to use XRM imaging to characterize two complex emulsion-based semisolid formulations, a petrolatum-based ointment with a dispersed phase comprising a hydrophilic liquid, and an oil-in-water cream. This initial assessment of technology showed that microstructure details such as globule size distribution, volume fraction, spatial distribution uniformity, inter-globule spacing, and globule sphericity could be obtained and parameterized. It was concluded that XRM imaging, combined with artificial intelligence-based image processing is feasible to generate advanced characterization of semisolid formulation microstructure through 3D visualization and parameterization of globule attributes. This technique holds promise to provide significantly richer microstructure details of semisolid formulations. When fully developed and validated, it is potentially useful for quantitative comparison of microstructure equivalence of semisolid formulations.

Automated summary

This study investigates the use of X-ray microscopy (XRM) imaging to characterize the microstructure of semisolid formulations containing multiple immiscible phases. The results show that XRM imaging combined with artificial intelligence-based image processing can provide advanced characterization of semisolid formulation microstructure and offer richer details.