Physical characteristics of images obtained with '0.5T' MR equipment

A. Crespi, R. G. Pellegrini, N. Paruccini, E. De Ponti, G. Montanari, A. Alfieri, G. Pedroli

Research output: Contribution to journalArticlepeer-review


We studied the physical parameters characterising the MR device and the images produced using a phantom. We evaluated the device performance, using the Modulation Transfer Function (MTF). For this purpose, two methods for MTF calculating were compared: the 'classic' one, operating on modulus images, and the 'previsualisation MTF' obtained by using real and imaginary images separately. With the latter method all the information concerning phase is not lost and difficulties of 'classic MTF', due to non-linearity of the MRI system, are avoided. Image noise was evaluated by means of the Noise Spectrum (or Wiener Spectrum), after having verified that the signal distribution was gaussian in a homogeneous region. Lastly, we evaluated the Signal-to-Noise Ratio (SNR) versus spatial frequencies by means of the ratio of MTF and Wiener Spectrum. The results show that the MTF is constant up to the characteristic frequency of the system, before visualisation. The image visualisation, obtained by means of the modulus operator applied to Bidimensional Fourier Transform, causes the loss of phase information, but, on the other hand, gives a better image uniformity. The Noise appears to be instrumentation dependent and 'white', that is non-correlated and structured like ground noise. Thus the SNR results constant up to the maximum resolution of the system, then the signal falls naturally.

Original languageEnglish
Pages (from-to)218-221
Number of pages4
JournalPhysica Medica
Issue numberSUPPL. 1
Publication statusPublished - 1997


  • Modulation Transfer Function
  • MR Imaging
  • Signal to Noise Ratio
  • Wiener Spectrum

ASJC Scopus subject areas

  • Biophysics
  • Physics and Astronomy(all)


Dive into the research topics of 'Physical characteristics of images obtained with '0.5T' MR equipment'. Together they form a unique fingerprint.

Cite this