Photon counting multi-spectral x-ray computed tomography of multi-component cryoprotectant diffusion into tissue surrogate.

Objective.To demonstrate that multi-spectral x-ray computed tomography (CT) and an empirical energy bin selection technique can quantitatively measure heterogeneous diffusive transport of cryoprotectants in a tissue surrogate.Approach.We let a ternary cryoprotectant cocktail composed of dimethyl sulfoxide (DMSO), glycerol, and water diffuse into a hydrogel designed to mimic the diffusion coefficient of human muscle. Photon counting multi-spectral x-ray CT (MSCT) is used to non-invasively quantify the spatiotemporal distribution of the cryoprotectants diffusing into the hydrogel tissue-surrogate phantom. An empirical photon-energy bin selection algorithm is developed to achieve sensitivity to low-contrast cryoprotectants without contrast agents or fluorescence edges. The technique is validated on uniformly mixed DMSO-glycerol-water solutions.Main results.We resolve cryoprotectant volume fractions to within 5% accuracy and observe heterogeneous diffusion of the cryoprotectants, a phenomenon not observable with energy-integrating x-ray CT techniques that are currently standard in cryoprotectant perfusion experiments. The developed photon-energy bin selection method outperforms energy bins selected by maximizing flux or maximizing attenuation contrast.Significance.Cryopreservation by vitrification holds promise for extending biological cryopreservation to tissues and potentially whole organs, but a key step-loading cryoprotectant chemicals-is not well understoodin situ, leading to inconsistent and poor recovery. Insufficient cryoprotectant loading may lead to freezing, precluding successful recovery while over-loading results in chemical toxicity. Despite the importance of this step, existingin situmeasurements of cryoprotectant permeation remain largely unvalidated and do not resolve individual cryoprotectant concentrations. We show how MSCT can provide experimental data for validating perfusion models, assess perfusion status during cryoprotectant loading, and can quantitatively assess cryoprotectant loading protocols.