Physica Medica: European Journal of Medical Physics - Volume 32 - Supplement 1 - February 2016

129 / 146

Developments in MRI
E.420
EXPOSURE TO STATIC MAGNETIC FIELDS IN NMR RESEARCH
LABORATORIES
G. Acri *, C. Sansotta, B. Testagrossa, A. Salvo, G. Vermiglio. Biomorf
Department, University of Messina, Messina, Italy
Introduction: To obtain cumulative exposure of workers to static mag-
netic fields in research laboratories, isotropic magnetic flux density values
generated by MR spectrometers were integrated over time. One need only
to consider that in a non medical environment, exposure limits are regu-
lated only by international guidelines and by good practice, while in MR
diagnostic environment there is a proper legislation. In this paper, the AA
also introduced a weighting function, incorporating the international guide-
lines limits, and tested it with the Italian legislation existing for medical
workers. The results were stored on a personal dose card, in order to eval-
uate the daily exposure of the worker and to better study the long term
effects to static magnetic fields.
Materials and Methods: A Hall-sensor probe was used for the measure-
ments on MR scanners in research laboratories (300–900 Mhz; 7–21 T) and
the obtained values were integrated over time. A weighting function cali-
brated on the field of strength was applied.
Results: The obtained values were considered on the basis “as is” and
corrected by the weighting function, and then compared with the Italian
exposure limits for medical operators working in MRI environment.
All values did not exceed the recommended threshold international
limits; in addition, the so calculated values permit to determine the
number of operations that a single operator can perform in a working day,
since the simple measure of absorbed dose can lead to a false sense of
security.
Conclusions: The calculus of the total dose absorbed can be considered
as an index of the global exposition in MR research laboratories and re-
ported in a special recording card, to evaluate long term effects. It seems
to be important to introduce a weight factor guidelines-oriented in the cal-
culus, to better evaluate the real exposure. The weight factor is easily
customizable, taking into account the specific guidelines and/or the ap-
plicable legislation.
E.421
DIFFUSE AND LOCAL MYOCARDIAL FIBROSIS EVALUATED BY MEANS OF
T1-3D VOLUMETRIC MAPPING. INITIAL CLINICAL EXPERIENCE
P. Di Renzi A. Abella E. Belligotti L. Begnozzi P. Pasqualetti
I. Simonelli A. Conigli a Department of Radiology, Fatebenefratelli Hospital,
S. Giovanni Calibita, Roma, Italy; b Specialization school of Medical Physics, Tor
Vergata University, Roma, Italy; c Medical Physics Unit, Fatebenefratelli Hospital,
S. Giovanni Calibita, Roma, Italy; d Service of Medical Statistics and Information,
AFaR Division, Fatebenefratelli Foundation for Health Research and Education,
Roma, Italy
Introduction: Evaluation of myocardial T1 relaxation time is an emerg-
ing cardiovascular magnetic resonance (CMR) technique used to evaluate
diffuse and regional myocardial fibrosis. We report our initial clinical ex-
perience using volumetric T1 mapping of the heart.
Materials and Methods: We performed CMR on 34 patients, 21 with isch-
emic and non ischemic cardiomyopathy diseases (CM) and 13 normative
subjects (NS). Left ventricular morphology, function and distribution of re-
gional myocardial fibrosis were evaluated with cine imaging and late
gadolinium enhancement (LGE). A three-dimensional inversion recovery
sequence was used to generate volume T1 maps of the heart and a custom
Matlab® code was developed to divide myocardium into 48 different sectors
(6 sectors for 8 slices). Acquisitions were performed after 20 minutes from
contrast medium administration. For each patients mean myocardium T1
value, T1 histograms and T1 maps were obtained and evaluated. Quality
assessment of T1 maps was performed on the basis of the goodness of T1
fit. Differences between NS and CM were tested with parametric t-test. The
receiver operating characteristic curve was adopted to evaluate the ability
of myocardium T1 in discriminating CM and NS groups. “Optimal” cut-
offs were calculated by the Youden’s index.
Results: We found that CM group has significantly lower mean myocar-
dium T1 value compared to the NS group (593 + 72 ms vs 630 + 36 ms;
p = 0.03). Liver, skeletal muscle and blood T1 values were similar in both
groups (451 + 40 ms vs 430 + 33 ms, p = 0.33, 777 + 57 ms vs 773 + 71 ms,
p = 0.95, 500 + 85 ms vs 486 + 19 ms, p = 0.53).
Conclusions: Our results demonstrate that 3D-T1 mapping provides
good qualitative and quantitative evaluation of myocardium. 3D-T1
maps allow direct evaluation of differences in myocardium T1 pattern
distribution of normal and pathological myocardium. The T1 cut-off
value helps distinguish normal and pathological subjects in the clinical
scenario.
E.422
NOVEL MRI QUALITY ASSURANCE PROCEDURE BASED ON SIGNAL TO
NOISE RATIO LINEARITY
M. Bettiol L. Lorenzon R. Rauco B. Cassano S. Murri D. Aragno
a Dipartimento di Medicina Molecolare, Sapienza, Roma, Italy; b Azienda
Ospedaliera S. Camillo Forlanini, Roma, Italy
Introduction: Both the signal to noise ratio (SNR) and the slice thickness
(ST) are sensitive and non-specific quality control (QC) parameters since
they are affected by several and different system features. The SNR depends
on changes of the magnetic field strength over time, system calibration,
gain, coil tuning and coupling, radiofrequency (RF) shielding, pre-amplifier
and receiver. The ST is affected by gradient and RF fields’ non-uniformity,
poor RF pulse shape due to RF transmitter amplitude nonlinearity and in-
stability of the RF phase and/or amplitude.
Since SNR of MRI image is proportional to the sample volume per image
voxel, V, we propose as a novel CQ parameter the SNR trend as a function
of V, by varying the ST.
By construction, the SNR linearity, SNRL, ensures the distinct monitoring
of all factors that degrade ST and SNR.
Materials and Methods: Four superconductive scanners belonging to dif-
ferent manufacturers, with a field of 1.5 T located at A.O. S.Camillo-
Forlanini, were used: Philips (NT15), Siemens (Avanto), Toshiba (Vantage),
General Electric (Signa Excite).
Measures were performed on the head and body coils. Two cylindrical phan-
toms of 16 cm and 20 cm diameter were used, both filled with a 8 mM
CuSO4 solution.
The voxel size, V = δxδyδz, was varied by changing ST (δz), in the range from
2 mm to 10 mm.
SNR is obtained according to AAPM directive using the spin echo
sequence.
Results: The trend of SNR as a function of ST has been measured thanks
to the linear fit. The obtained R2 values range from 0.950 to 0.999.
Conclusion: The evaluation of SNRL allows controlling the MRI factors af-
fecting SNR and ST, separately.
Contents lists available at
Physica Medica
journal homepage: