Meeting Banner
Abstract #3684

Non-Gaussian Analysis of Diffusion-Weighted Imaging in Nasopharyngeal Carcinoma at 3T

Jing Yuan1, Steven Kwok Keung Chow1, David Ka Wai Yeung1, Yujia Li1, Anil T. Ahuja1, Ann D. King1

1Imaging and Interventional Radiology, The Chinese University of Hong Kong, Shatin, NT, Hong Kong

This pilot study is to investigate the feasibility of non-Gaussian diffusion models, including diffusion kurtosis imaging (DKI), stretched exponential model (SEM), intravoxel incoherent motion (IVIM) and statistical model, for 3T diffusion-weighted imaging (DWI) analysis in patients with nasopharyngeal carcinoma (NPC). The results showed that at an extended b-value range up to 1500s/mm2, all non-Gaussian models yielded better diffusion signal fitting than the normal mono-exponential model for both primary tumors and metastatic nodes. Additional to the nominal diffusion coefficient, other non-Gaussian parameters could be extracted as potential biomarkers for NPC lesion detection and characterization in clinical applications.

Abstract PDF E-Poster Video

Keywords

accuracy anatomical apparent assumption behavior behaviors better biological brain carcinoma channel characteristics characterization chow clinical coefficient coil correlation correlations detection deviation diffusion displays distribution drawn except exhibited explored exponential extended extracted feasibility feasible fitted fitting future goodness grant head hong incoherent influence investigate investigated king kurtosis leads lesion limited listed maps matrix metastatic model modeling models mono motion neck nodes overlaid patients pilot pixel potentials practice primary proposed radiology rarely received reconstructed response reveal senior several significantly slices statistical statistics stretched studied supported suppressed table tissues treatment tumor tumors usually water weak widely wise yields yuan