Measurement of liver iron overload: noninvasive calibration of MRI-R2* by magnetic iron detector susceptometer.

An accurate assessment of body iron accumulation is essential for the diagnosis and therapy of iron overload in diseases such as thalassemia or hemochromatosis. Magnetic iron detector susceptometry and MRI are noninvasive techniques capable of detecting iron overload in the liver. Although the transverse relaxation rate measured by MRI can be correlated with the presence of iron, a calibration step is needed to obtain the liver iron concentration. Magnetic iron detector provides an evaluation of the iron overload in the whole liver. In this article, we describe a retrospective observational study comparing magnetic iron detector and MRI examinations performed on the same group of 97 patients with transfusional or congenital iron overload. A biopsy-free linear calibration to convert the average transverse relaxation rate in iron overload (R(2) = 0.72), or in liver iron concentration evaluated in wet tissue (R(2) = 0.68), is presented. This article also compares liver iron concentrations calculated in dry tissue using MRI and the existing biopsy calibration with liver iron concentrations evaluated in wet tissue by magnetic iron detector to obtain an estimate of the wet-to-dry conversion factor of 6.7 ± 0.8 (95% confidence level).

Total skin electron therapy (TSET): a reimplementation using radiochromic films and IAEA TRS-398 code of practice.

PURPOSE: The aim of this work is to present an updated implementation of total skin electron therapy (TSET) using IAEA TRS-398 code of practice for absolute dosimetry and taking advantage of the use of radiochromic films. The optimization of quality control tests is also included. METHODS: A Varian 2100 C/D linear accelerator equipped with the special procedure HDTSe- (high dose rate total skin electron mode, E=6 MeV) was employed to perform TSET irradiations using the modified Stanford technique. The commissioning was performed following the AAPM report 23 recommendations. In particular, for dual-field beams irradiation, the optimal tilt angle was investigated and the dose distribution in the treatment plane was measured. For a complete six dual-field beams irradiation, the treatment skin dose on the surface of a cylindrical phantom was evaluated by radiochromic films and the B factor which relates the single dual-field skin dose to the six dual-field skin dose was assessed. Since the TRS-398 reference conditions do not meet the requirements of TSET absolute dosimetry, GafChromic EBT films were also employed to check and validate the application of the protocol. Simplified procedures were studied to verify beam constancy in PMMA phantoms without the more difficult setup of total skin irradiation. RESULTS: The optimized geometrical setup for dual-field beams was: Tilt angle = +/- 19 degrees, SSD=353 cm, and the beam degrader (200 x 100 X 1 cm3) placed at 320 cm from the source. As regards to dose homogeneity in the treatment plane, for dual-field beams irradiation, the mean relative dose value was 97% +/- 5% (normalizing to 100% at the calibration point level). For six dual-field beams irradiation, the multiplication factor B was 2.63. In addition, beam quality, dose rate, and bremsstrahlung contribution were also suitable for TSET treatments. The TRS-398 code of practice was used for TSET dosimetry, as dose measurements performed by ionization chamber and radiochromic film agreed within 2.5%. Simplified quality control tests and baseline values were presented in order to check flatness, symmetry, and field size with radiochromic films and output and beam quality constancy with ionization chamber. Short-term reproducibility and MU linearity tests were also included. CONCLUSIONS: Commissioning parameters met the requirements of TSET treatments and the matching of AAPM guidelines with the IAEA code of practice was successful. Frequent beam performance controls can be easily performed through the presented quality assurance tests. Radiochromic dosimetry facilitated the TSET commissioning and played a major role to validate the application of TRS-398.

Dosimetric verification of a commercial Monte Carlo treatment planning system (VMC++) for a 9 MeV electron beam.

The aim of this work was to evaluate the performance of the voxel-based Monte Carlo algorithm implemented in the commercial treatment planning system ONCENTRA MASTERPLAN for a 9 MeV electron beam produced by a linear accelerator Varian Clinac 2100 C/D. In order to realize an experimental verification of the computed data, three different groups of tests were planned. The first set was performed in a water phantom to investigate standard fields, custom inserts, and extended treatment distances. The second one concerned standard field, irregular entrance surface, and oblique incidence in a homogeneous PMMA phantom. The last group involved the introduction of inhomogeneities in a PMMA phantom to simulate high and low density materials such as bone and lung. Measurements in water were performed by means of cylindrical and plane-parallel ionization chambers, whereas measurements in PMMA were carried out by the use of radiochromic films. Point dose values were compared in terms of percentage difference, whereas the gamma index tool was used to perform the comparison between computed and measured dose profiles, considering different tolerances according to the test complexity. In the case of transverse scans, the agreement was searched in the plane formed by the intersection of beam axis and the profile (2D analysis), while for percentage depth dose curves, only the beam axis was explored (1D analysis). An excellent agreement was found for point dose evaluation in water (discrepancies smaller than 2%). Also the comparison between planned and measured dose profiles in homogeneous water and PMMA phantoms showed good results (agreement within 2%-2 mm). Profile evaluation in phantoms with internal inhomogeneities showed a good agreement in the case of "lung" insert, while in tests concerning a small "bone" inhomogeneity, a discrepancy was particularly evidenced in dose values on the beam axis. This is due to the inaccurate geometrical description of the phantom that is linked to the calculation voxel size, a feature over which the user has no control.