Insights into tissue microstructure using a double diffusion encoding sequence on a clinical scanner: Validation and application to experimental tumor models

Abstract

To present a double diffusion encoding MRI sequence on a clinical scanner to analyze micro-structure and micro-vasculature of tumors. Methods: The sequence was tested on phantoms, asparaguses, and 2 tumors allografts in a rodent. Results were analyzed using an adapted VERDICT model to estimate microstructural parameters. The technical feasibility of the sequence on a 3T clinical system was assessed on a water phantom. The accuracy of cell size estimation was assessed on asparaguses by comparison with light microscopy. Cell size estimations were also validated when limiting relative angles of diffusion encodings to 0 and 180°. Sensitivities to restricted diffusion and incoherent flow from the vasculature were investigated in experimental tumor models. Values of microstructural parameters in viable and decaying tumor tissue were compared with those obtained from histological analysis. Results: Measurements on the water phantom revealed no significant sequence artifacts and accurate apparent diffusion coefficient values within a 4% relative error. In asparaguses, quartiles and medians of pore size distributions typically deviated less than 6% from light microscopy regardless of whether the full or reduced set of relative angles was used. Signal analyses in tumors showed mixed effects of both blood flow and diffusion restriction. Microstructural parameter estimations in tumors were consistent with histology and allowed clear and histology-proven distinctions between decaying and viable tumor tissue. Conclusions: Double diffusion encoding with clinical gradients and scan times allows characterization of restricted diffusion and micro-circulation flow in tumors. Our estimated microstructural parameters are promising for further investigations in assessing microstructural evolutions in tumors.

MRI‐cytometry: Mapping nonparametric cell size distributions using diffusion MRI

“…Estimating nonparametric compartment size distributions without a priori knowledge of the distribution function has been performed to estimate: (1) pore size distribution of glass capillaries 18 ; (2) nonparametric axon size distributions in fixed tissues using pulse gradient spin echo (PGSE), 19 nonuniform oscillating gradient spin (OGSE) 20 or double diffusion encoding 21 ; and (3) cell size distributions in asparagus and animal allografts. 22 All of these methods addressed the complex optimization problems without assuming any specific distribution forms, ensuring the generalizability of both normal and abnormal tissues. However, these approaches either used special preclinical hardware or focused on animal applications only.…”