N.V. Anisimov - Dr. Sc. (Phys.-Math.), Senior Reserch Scientist, Faculty of Fundamental Medicine, Lomonosov Moscow State University E-mail: anisimovnv@mail.ru
E.I. Shalamova - Student, Faculty of Physics, Lomonosov Moscow State University. E-mail: katya2006-91@mail.ru
K.L. Volkova - Student, Faculty of Physics, Lomonosov Moscow State University. E-mail: trinvi@rambler.ru
M.V. Gulyaev - Dr. Sc. (Phys.-Math.), Reserch Scientist, Lomonosov Moscow State University, Faculty of Fundamental Medicine. E-mail: mihon-epsilon@yandex.ru
A.A. Samoylenko - Ph.D. (Chem.), Head of Laboratory, Semenov Institute of Chemical Physics RAS (Moscow). E-mail: sam_nmr@chph.ras.ru

Methods for evaluating the content of fatty tissue in the human body are described. They are based on the analysis of MR images and recording the NMR spectra of the whole body. Particular attention is paid to the spectroscopic method, where the evaluation is made by analysis of the intensities ratio of water and fat peaks. This method was introduced by Mystkowski et al. in 2000. He studed laboratory animals and measurements were performed by a high field NMR spectrometer. Interest in such measurements is due to the fact that they are easy to implement and take little time. But the main thing is that there is correlation between the intensity ratio of peak and the content of fat in the body of an animal. The aim of our work was to adapt the method used on small laboratory animals to human studies. In our case, the measurements were performed on a standard (with a horizontal bore magnet) 0.5 Tesla MR scanner. NMR spectra were recorded from all parts of the body, and then summed. In the total spectrum, peaks of water and fat were defined. The analysis of peak intensities gave information about the content of fat in the human body. Registration of the NMR spectra from all parts of the human body was carried out in a homogeneous magnetic field. To do this, the patient's body was moved stepwise along the horizontal axis of the magnet. Scanning area was limited by slice thickness of 20 cm. Slice plane was oriented perpendicularly to the mentioned axis. Local NMR spectroscopy scanning methods were used to fix the slice thickness. These methods use inhomogeneous (gradient) fields which are applied synchronous with exciting RF pulses. Spectral NMR data were compared with the values of the average density of the body for each object of research, as well as the volume of fat determined by MR images. There were T1 and T2-FSE weighted images. Abdominal and subcutaneous fat areas were determined visually by anatomical landmarks. Segmentation of these areas was carried out. It gave possibility to count the total amount of whole body fat. The problem connected with low magnetic field was revealed. Width of the lines appeared to be comparable with the distance between them. It lead to the difficulties of measuring integrals of spectral lines. Result of measurements for 8 subjects indicates a correlation between the average density of body and fat in it. It is consistent with the results of work by Mystkowski et al. The conclusion is that it is preferable to perform spectroscopic measurements at a higher magnetic field (1.5 T and more).