Comparison of a 0.55 T dental-dedicated magnetic resonance imaging system with a 1.5 T system in evaluation of the temporomandibular joint regarding subjective image quality assessment and rater agreement.
Donald R Nixdorf, Andreas Greiser, Carmel Hayes, Laurence Gaalaas, Beth R Groenke, João Marcus de Carvalho E Silva Fuglsig, Katrine Mølgaard Johannsen, Cory R Herman, Shanti Kaimal, Estephan J Moana-Filho, Mariona Mulet, Can Özütemiz, Rubens Spin-Neto
Author Information
Donald R Nixdorf: Department of Diagnostic & Biological Sciences, School of Dentistry, Minneapolis, Minnesota; Department of Radiology, Medical School, University of Minnesota, Minneapolis, Minnesota. Electronic address: nixdorf@umn.edu.
Andreas Greiser: Siemens Healthineers AG, Erlangen, Germany.
Laurence Gaalaas: Department of Diagnostic & Biological Sciences, School of Dentistry, Minneapolis, Minnesota.
Beth R Groenke: Department of Diagnostic & Biological Sciences, School of Dentistry, Minneapolis, Minnesota.
João Marcus de Carvalho E Silva Fuglsig: Department of Dentistry and Oral Health, Aarhus University, Aarhus, Denmark.
Katrine Mølgaard Johannsen: Department of Dentistry and Oral Health, Aarhus University, Aarhus, Denmark.
Cory R Herman: Department of Diagnostic & Biological Sciences, School of Dentistry, Minneapolis, Minnesota.
Shanti Kaimal: Department of Diagnostic & Biological Sciences, School of Dentistry, Minneapolis, Minnesota.
Estephan J Moana-Filho: Department of Diagnostic & Biological Sciences, School of Dentistry, Minneapolis, Minnesota.
Mariona Mulet: Department of Diagnostic & Biological Sciences, School of Dentistry, Minneapolis, Minnesota; HealthPartners Park Nicollet, Minneapolis, Minnesota.
Can Özütemiz: Department of Radiology, Medical School, University of Minnesota, Minneapolis, Minnesota.
Rubens Spin-Neto: Department of Dentistry and Oral Health, Aarhus University, Aarhus, Denmark.
OBJECTIVE: We aimed to compare subjective assessment of image quality and rater agreement of magnetic resonance (MR) images of the temporomandibular joint (TMJ) acquired with dental-dedicated 0.55 T and clinical 1.5 T magnetic field strengths. STUDY DESIGN: Six blinded raters evaluated randomized TMJ MR images from 5 healthy participants imaged with 3 pulse sequences on 2 scanners: a 0.55 T system using a custom-built surface coil for dental-dedicated applications and a clinical 1.5 T system using a standard head/neck coil. TMJ anatomy and image quality parameters were assessed using a three-point scale (0 = unacceptable, 1 = acceptable, and 2 = excellent). Data was pooled across participants, pulse sequences, and sides. Kappa coefficients were calculated for intra- and inter-rater agreement. Chi-square tests were applied to explore differences between scanners. Significance was established at P<0.05. RESULTS: Four different raters preferred 0.55 T images for visualizing hard and/or soft tissue anatomy (P≤.02). One rater preferred the 0.55 T system regarding contrast. No other differences were detected regarding qualitative ratings. Kappa values for intra-rater agreement ranged from 0.49 to 0.70 for 0.55 T images and 0.42 to 0.71 for 1.5 T images. Inter-rater kappa values ranged from 0.02 to 0.29 for 0.55 T images and 0.02 to 0.23 for 1.5 T images. CONCLUSIONS: Images from the two systems were rated similarly. Further research is needed to assess diagnostic performance in patients.
