TRACED: In vivo imaging of extracellular intrinsic diffusivity, tortuosity, cell size distribution and cell density in human glioma patients

2026-05-04

AI Summary

Current diffusion MRI models struggle to accurately quantify the extracellular intrinsic diffusivity and microstructural parameters of tumors at intermediate time scales. This work proposes TRACED, a biophysical model that integrates Monte Carlo diffusion simulations based on spherical cell distributions with neural networks, and introduces Sim2PINN—a physics-informed transfer learning framework—to enable, for the first time, simultaneous in vivo quantification of extracellular intrinsic diffusivity, tortuosity, cell size distribution, and cellular density in glioma patients. Evaluated on eight patients with mixed-grade gliomas, TRACED significantly outperformed conventional two-compartment models and showed agreement with histological measurements in two cases, demonstrating its accuracy and generalizability.

Abstract

The lack of analytical models describing diffusion time dependence at intermediate time scales in complex tissue microstructure limits the accurate quantification of extracellular diffusivity and tissue microstructure. We introduce TRACED, a biophysical model that incorporates diffusion time dependence in cell distributions to quantify pathologically-relevant properties in solid tumors. Neural networks were trained on Monte Carlo diffusion simulations using sphere distribution-based geometries to enable the rapid computation of time-dependent diffusion MRI signals in cell populations of variable cell size. Model sensitivity and fit performance were assessed via simulation. Diffusion data from eight mixed-grade glioma patients was fitted using the TRACED model. Data fitting was performed using a novel physics-informed transfer learning pipeline, Sim2PINN. In two patients, cell size measurements were compared directly with image-localized histology. Simulation results indicate improved parameter estimation compared to the simple two-compartment model. TRACED enabled the simultaneous in vivo quantification of intracellular volume fraction, cell size distribution, extracellular intrinsic diffusivity, and tortuosity in glioma patients. Neural network implementations of diffusion time-dependence and tortuosity showed behavior consistent with coarse-graining and effective medium theory, respectively. Future work will explore the clinical utility of TRACED parameters in additional patients.

Problem

• diffusion time dependence
• tissue microstructure
• extracellular diffusivity
• cell size distribution
• glioma

Innovation

• TRACED
• diffusion time dependence
• physics-informed neural networks
• cell size distribution
• tortuosity