Affordable Access

Publisher Website

Achieving high spatial and temporal resolution with perfusion MRI in the head and neck region using golden-angle radial sampling.

  • Tomppert, Andrea1
  • Wuest, Wolfgang1, 2
  • Wiesmueller, Marco1
  • Heiss, Rafael1
  • Kopp, Markus1
  • Nagel, Armin M1, 3
  • Tomita, Hayato4
  • Meixner, Christian1, 3
  • Uder, Michael1, 2
  • May, Matthias Stefan5, 6
  • 1 Department of Radiology, University Hospital Erlangen, Maximiliansplatz 3, 91054, Erlangen, Germany. , (Germany)
  • 2 Imaging Science Institute, University Hospital Erlangen, Erlangen, Germany. , (Germany)
  • 3 Institute of Medical Physics, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, Germany. , (Germany)
  • 4 Department of Radiology, St. Marianna University School of Medicine, Miyamae-ku, Kawasaki, Japan. , (Japan)
  • 5 Department of Radiology, University Hospital Erlangen, Maximiliansplatz 3, 91054, Erlangen, Germany. m[email protected] , (Germany)
  • 6 Imaging Science Institute, University Hospital Erlangen, Erlangen, Germany. [email protected] , (Germany)
Published Article
European Radiology
Publication Date
Apr 01, 2021
DOI: 10.1007/s00330-020-07263-0
PMID: 32970184


Conventional perfusion-weighted MRI sequences often provide poor spatial or temporal resolution. We aimed to overcome this problem in head and neck protocols using a golden-angle radial sparse parallel (GRASP) sequence. We prospectively included 58 patients for examination on a 3.0-T MRI using a study protocol. GRASP (A) was applied to a volumetric interpolated breath-hold examination (VIBE) with 135 reconstructed pictures and high temporal (2.5 s) and spatial resolution (0.94 × 0.94 × 3.00 mm). Additional sequences of matching temporal resolution (B: 2.5 s, 1.88 × 1.88 × 3.00 mm), with a compromise between temporal and spatial resolution (C: 7.0 s, 1.30 × 1.30 × 3.00 mm) and with matching spatial resolution (D: 145 s, 0.94 × 0.94 × 3.00 mm), were subsequently without GRASP. Instant inline-image reconstructions (E) provided one additional series of averaged contrast information throughout the entire acquisition duration of A. Overall diagnostic image quality, edge sharpness and contrast of soft tissues, vessels and lesions were subjectively rated using 5-point Likert scales. Objective image quality was measured as contrast-to-noise ratio in D and E. Overall, the anatomic and pathologic image quality was substantially better with the GRASP sequence for the temporally (A/B/C, all p < 0.001) and spatially resolved comparisons (D/E, all p < 0.002 except lesion edge sharpness with p = 0.291). Image artefacts were also less likely to occur with GRASP. Differences in motion, aliasing and truncation were mainly significant, but pulsation and fat suppression were comparable. In addition, the contrast-to-noise ratio of E was significantly better than that of D (pD-E < 0.001). High temporal and spatial resolution can be obtained synchronously using a GRASP-VIBE technique for perfusion evaluation in head and neck MRI. • Golden-angle radial sparse parallel (GRASP) sampling allows for temporally resolved dynamic acquisitions with a very high image quality. • Very low-contrast structures in the head and neck region can benefit from using the GRASP sequence. • Inline-image reconstruction of dynamic and static series from one single acquisition can replace the conventional combination of two acquisitions, thereby saving examination time.

Report this publication


Seen <100 times