Premium
MO‐FG‐CAMPUS‐JeP2‐01: 4D‐MRI with 3D Radial Sampling and Self‐Gating‐Based K‐Space Sorting: Image Quality Improvement by Slab‐Selective Excitation
Author(s) -
Deng Z,
Yang W,
Pang J,
Tuli R,
Bi X,
Hakimian B,
Fraass B,
Li D,
Fan Z
Publication year - 2016
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4957354
Subject(s) - image quality , undersampling , artifact (error) , nuclear medicine , sampling (signal processing) , biomedical engineering , materials science , physics , medicine , optics , computer science , artificial intelligence , image (mathematics) , detector
Purpose: A recent 4D MRI technique based on 3D radial sampling and self‐gating‐based K‐space sorting has shown promising results in characterizing respiratory motion. However due to continuous acquisition and potentially drastic k‐space undersampling resultant images could suffer from low blood‐to‐tissue contrast and streaking artifacts. In this study 3D radial sampling with slab‐selective excitation (SS) was proposed in attempt to enhance blood‐to‐tissue contrast by exploiting the in‐flow effect and to suppress the excess signal from the peripheral structures particularly in the superior‐inferior direction. The feasibility of improving image quality by using this approach was investigated through a comparison with the previously developed non‐selective excitation (NS) approach. Methods: Two excitation approaches SS and NS were compared in 5 cancer patients (1 lung 1 liver 2 pancreas and 1 esophagus) at 3Tesla. Image artifact was assessed in all patients on a 4‐point scale (0: poor; 3: excellent). Signal‐tonoise ratio (SNR) of the blood vessel (aorta) at the center of field‐of‐view and its nearby tissue were measured in 3 of the 5 patients (1 liver 2 pancreas) and blood‐to‐tissue contrast‐to‐noise ratio (CNR) were then determined. Results: Compared with NS the image quality of SS was visually improved with overall higher signal in all patients (2.6±0.55 vs. 3.4±0.55). SS showed an approximately 2‐fold increase of SNR in the blood (aorta: 16.39±1.95 vs. 32.19±7.93) and slight increase in the surrounding tissue (liver/pancreas: 16.91±1.82 vs. 22.31±3.03). As a result the blood‐totissue CNR was dramatically higher in the SS method (1.20±1.20 vs. 9.87±6.67). Conclusion: The proposed 3D radial sampling with slabselective excitation allows for reduced image artifact and improved blood SNR and blood‐to‐tissue CNR. The success of this technique could potentially benefit patients with cancerous tumors that have invaded the surrounding blood vessels where radiation therapy is needed to remove tumor from those regions prior to surgical resection. This work is partially supported by NIH R03CA173273; and CTSI core voucher award.