• Magnetic resonance

Nuclear Magnetic Resonance has acquired more and more popularity in medical diagnostics over the years because of its safety and flexibility. The research in this field is moving towards Magnetic Resonance (MR) scanners with Ultra High Field (UHF), i.e. with a value of static field of 7 T and above. They allow to investigate human districts with higher spatial resolution and higher signal to noise ratio with respect to the scanners used in clinical procedures.
The main current open issues are to highlight on which pathological conditions they are useful to provide more accurate diagnoses and to evaluate their safety.
An aspect to be considered, for the safety of the patient, is the exposure to the electromagnetic waves which, in case of 7 T scanner, are RF of 300 MHz.
The radiation deposits energy inside the tissue under exam, causing an increase in temperature. The European laws imposed limits on this increase in temperature, which has to be contained within 1C and on global SAR (Specific Absorbtion Rate), which indicates the electromagnetic energy absorbed for unit of time by an element of tissue of one unit of mass. The biggest limitation of SAR is that it is not measurable directly, but it is an indirect measure carried out by the scanner. The estimate of the global SAR is sufficient in case it has an uniform distribution, which is when the wavelength of the RF is longer with respect to the object to investigate. It is no more valid when the wavelength is shorter with respect to the object, or when the field transmitted is not uniform in tissue because of a variation of dielectric constant or in presence of prostheses.
In this case hotspots can be produced with consequent heating of the tissues over the allowed limit.
The main objective of this thesis is to setup and validate a technique able to monitor directly the increase of temperature during a magnetic resonance imaging session.
These techniques are referred as Magnetic Resonance Thermometry (MRT).
MRT are based on the relationship between the water resonance frequency shift and the temperature. They have been implemented in this thesis according to two methods both based on the different behavior of water and fat spins.
Static magnetic field drifts over the time, and it affects both water and fat phase. Moreover the water magnetization acquires an additional phase component, related to the temperature change, while fat magnetization phase only depends on the time-dependent inhomogeneities of the static field, thus it can be used as a reference for field drift. It represents the key point of the two techniques called "referenced" and "self-referenced" MRT.
In the "referenced" technique, proposed by Oh et al., external vials with oil are inserted inside the RF coil, around the sample at fixed spatial positions. The phase signal extrapolated inside the vials will provide information on the timedependent inhomogeneities of the static field and the corresponding baseline image is applied on the water phase map. This technique is useful when you want to investigate fat poor human districts, such as the brain.
In the "self-referenced" technique, implemented by Hofstetter et al., the fat signal that is naturally present in the investigation area is exploited; by means of the fat-water separation Dixon technique, the fat signal is distinguished from that of the water. The corresponding water phase map is corrected using the fat phase map as a reference.
Both these MRT methods have been implemented in MR scanners and compared through cooling down experiments carried out in a temperature ranges useful for the in vivo imaging .