Comparison of cine cardiac MR imaging performance: 0.6 T versus 1.5 T, is 0.6 T fit for purpose?

Presenting Author(s)

• 

  Alessio Perazzolo, MD

  Radiologist
  Università Cattolica del Sacro Cuore, Rome Italy
  Rome, Lazio, Italy

Primary Author(s)

• JB

  Jacinta Browne, PhD

  Associate Professor
  Mayo Clinic, Department of Radiology, Minnesota, United States

Co-Author(s)

• 

  Jouke Smink, MSc

  MR Clinical Scientist
  Philips Healthcare
  Best, Noord-Brabant, Netherlands
• TC

  Tzu Cheng Chao, PhD

  Postdoctoral researcher
  Mayo Clinic, Department of Radiology, Minnesota, United States
• EE

  Ece Ercan, PhD

  Physicist/Biomedical engineer
  Leiden University Medical Center, Leiden, The Netherlands, Netherlands
• DW

  Dinghui Wang, PhD

  Postdoctoral researcher
  Mayo Clinic, Department of Radiology, Minnesota, United States
• 

  Camilla Vittoria Vita, MD

  MD
  Department of Radiological and Hematological Sciences, Università Cattolica del Sacro Cuore, Rome, Italy; Department of Radiology, Mayo Clinic, Rochester MN USA
  Rome, Lazio, Italy
• AD

  Ajit Deveraj

  Researcher Scientist
  Mayo Clinic, Rochester MN, United States
• 

  Spencer Waddle, PhD

  Clinical Scientist
  Philips Healthcare
  Rocheseter, Minnesota, United States
• MP

  Michal Povazan, PhD

  PhDPostDoc Position at Johns Hopkins Medicine
  Johns Hopkins Medicine, United States
• VS

  Vincenzo Scialo', MD

  Doctor, PhD student
  Humanitas Research Hospital; Department of Radiology, Mayo Clinic, Rochester MN USA
  Milan, Lombardia, Italy
• TL

  Tim Leiner, MD, PhD

  Professor of Radiology
  Mayo Clinic
  Rochester, Minnesota, United States

Background: In recent years, mid-field MRI has shown promise to meet the technical demands of cardiac imaging [1,2]. Mid-field strengths offer advantages to CMR due to shorter T1-relaxation times, lower specific absorption rate and improved magnetic field homogeneity [3]. The purpose of this study was to evaluate the diagnostic capabilities of high-performance 0.6 T CMR imaging for quantification of ventricular function in healthy subjects.

Methods: In this prospective study, 10 healthy subjects underwent paired cardiac MRI exams on a 1.5 T scanner (Ingenia Evolution, Philips, The Netherlands) scanner and a prototype 0.6 T Philips MRI system (Philips Healthcare, Best, The Netherlands). All subjects underwent cine imaging of the heart, using breath-held balanced steady state free precession (bSSFP) and volumetric coverage with standard cardiac views.


The acquisition parameters used at 1.5 T and 0.6 T were matched apart from the flip angle which was 60° and 90° at 1.5 T and 0.6 T, respectively. For all cine acquisitions, left ventricular volumes (LVEF) and global longitudinal strain (GLS) were assessed by three radiologists using contouring software (AVW, Philips, The Netherlands). Bland-Altman were used to compare paired quantitative measurements of ventricular volumes and mass parameters between the two CMR exams.


Two radiologists provided gradings of image quality on a 4-point scale where a score of 1 was the best image quality (IQ).

Results: Representative images (Figure 1a) demonstrated comparable diagnostic quality across scanners. Image quality scores were similar at both field strengths (Figure 1b). Short-axis LV and RV volumes, as well as LVEF, demonstrated excellent agreement between 0.6 T and 1.5 T (all p > 0.05, r > 0.9). GLS was also comparable (−17.1% vs −18.2%, p = 0.11; r = 0.60). Bland-Altman plots of ventricular volumetric and mass metrics (Figure 2), LVEF and GLS (Figure 3) showed no significant difference between scanners.

Conclusion: This study demonstrates that a high-performance mid-field 0.6 T CMR scanner provides diagnostic image quality and reliable assessment of ventricular volumes, ejection fraction, and GLS, showing equivalence to 1.5 T for comprehensive cardiac functional analysis.

Figure 1. (a) Representative images demonstrating the Image quality of 0.6 T and 1.5 T breath-held cine bSSFP in short axis and long axis slices from one healthy subject (1- excellent, diagnostic without significant artefacts; 2- good, minor degradation in image quality not impacting diagnosis; 3- fair, major degradation in image quality not impacting diagnosis; and 4 – poor, major degradation in image quality rendering images non-diagnostic) and (b) comparison of five image quality metrics used to evaluate diagnostic confidence at both 0.6 T and 1.5 T field strengths provided by two radiologists provided on a 4-point scale where a score of 1 was the best image quality (IQ). Abbreviation: bSSFP = breath-held balanced steady state free precession.
Figure 2. Bland-Altman comparisons of short-axis LV measurements at 0.6 T and 1.5 T: (a) LVEDV, (b) LVESV, (c) stroke volume, (d) end-systolic mass, (e) end-diastolic mass, and (f) LVEF. No significant differences were observed between field strengths. Abbreviations: LV = left ventricle; LVEDV = left ventricular end-diastolic volume; LVEF = left ventricle ejection fraction.
Figure 3. Bland-Altman comparisons of monoplanar LV ejection fraction from (a) 2-chamber, (b) 3-chamber, (c) 4-chamber views, and (d) global longitudinal strain (GLS). All parameters showed good agreement between 0.6 T and 1.5 T.