Igo Primo Dem Files Download
Katerine Aldrige
Primo VE enables you to expose your library's unique materials to search engines (such as Google) so that users are directed back to your Primo VE interface. Sitemap files enable Web crawling tools (such as Google) to find Primo records that are dynamic (as a result of search queries) and to index them. The following publishing profiles enable you to define up to three sitemap files that enable you to create back links to different views or domain names within your Primo VE environment:
The PRIMO system is a computer software that allows the Monte Carlo simulation of linear accelerators and the estimation of the subsequent absorbed dose distributions in phantoms and computed tomographies. The aim of this work is to validate the methods incorporated in PRIMO to evaluate the deviations introduced in the dose distributions by errors in the positioning of the leaves of the multileaf collimator recorded in the dynalog files during patient treatment.
Modern radiation therapy techniques are based on the combination of multiple variables, such as the modulation of the beam intensity and the variation of the gantry rotation speed and the fluence output rate to maximize conformity of the dose to the planned target volumes (PTVs) and to spare organs-at-risk (OARs). The increased complexity of the treatment planning and delivery attained by those techniques reinforces the necessity of implementing refined patient-specific quality assurance (QA) procedures.
The PRIMO system is a software that allows the Monte Carlo simulation of linear accelerators for the generation of phase-space files (PSFs) and the estimation of dose distributions in phantoms and computed tomographies (CT) [4]. The interaction with the system is managed by a friendly graphical-user interface designed to spare the user of having to deal with the intricacies of the Monte Carlo method applied to radiation transport simulation. Furthermore, PRIMO has integrated functions for the analysis and visualization of simulated results including an environment for the comparison of dose distributions. PRIMO (version 0.3.1.1681) uses PENELOPE (version 2011) [5] as its main radiation transport engine. The Dose Planning Method (DPM v1.1) [6], a fast Monte Carlo radiation transport algorithm, has been recently implemented in PRIMO as an alternative Monte Carlo dose computation engine used to simulate dynamic plans [7, 8].
A function to create a treatment plan employing data extracted from the dynalog files was coded in PRIMO. Hereafter, we shall call this plan the reconstructed plan to differentiate it from the original plan created in the TPS and exported as a DICOM RTPLAN file. Consequently, we shall refer to the original dose and to the reconstructed dose as the dose distributions estimated by the Monte Carlo simulation of the original and reconstructed plans, respectively. The control points of the reconstructed plan can be generated either from the expected or the actual MLC positions, both recorded in the dynalog files. For both cases the following options have been coded:
Uniform reconstruction (UR): Reconstructing by uniformly sampling the records in the dynalog files, that is, by taking records at a given time interval. This interval can be freely chosen, with a minimum value of 50 ms (or 20 ms for TrueBeam linacs), in which case all records are considered.
Per-segment-reconstruction (PSR): The segment number stored in the dynalog files is used to sample only those records in which a change of segment occurs. This reconstruction method renders the same number of control points as the original plan.
Per-segment-reconstruction with error detection (PSR-ED): The reconstruction is made by including the records in which a change of segment occurs, in addition to all other records where at least one leaf is found having a position error above a given tolerance. The tolerance can be freely chosen starting from zero, in which case all records are considered. When the selected tolerance equals to or exceeds the maximum leaf error in the dynalog file, this reconstruction becomes equivalent to the PSR.
The PSR option reduces the number of control points to those in the original plan. This approach has the advantage of a faster Monte Carlo simulation because less time is employed in re-arranging the simulation geometry from one segment to the next one. However, this method has the limitation that segments with large errors in the position of the leaves can be missed in the reconstruction. In order to overcome this limitation, the PSR-ED reconstruction option was coded, which allows to include segments with significant position errors.
Two volumetric-modulated arc therapy (VMAT) clinical cases of prostate and head&neck were considered in this work. They were selected because of their differences in the region of the body treated, in the complexity of the MLC dynamics and in the range of leaves involved. In both cases the region inside the contour of the body of the patient is hereafter identified as body.
For the prostate case five PTVs were included in the analysis. Four were drawn as irregular rings involving the region of the prostate. Hereafter, they will be identified as PTV1 to PTV4 where PTV1 is the inner one. The fifth PTV, identified as PTV total is an envelope of all other PTVs. The selected OARs were bladder and rectum.
For the head&neck case, two PTVs were considered, PTV1 a large region encompassing the lymph nodes of the left side of the neck, while PTV2 included the gross tumor plus margins. The spinal canal and the left and right parotid glands were selected as OARs.
Both clinical cases included in this work were real cases of treated patients. The treatment plans produced clinically acceptable dose distributions and successfully passed a TPS independent plan verification process.
Monte Carlo simulations were run using the PRIMO system. The simulation of the patient-independent part of the linac was done using PENELOPE as the Monte Carlo engine. That part was simulated once to tally a PSF with nominal energy 6 MV and initial beam parameters E=6.2 MeV, FWHM E=0.186 MeV, FWHM focal spot size=0.15 cm and beam divergence 2.5 degrees. Splitting roulette [10, 11] was employed as variance-reduction technique. The rest of simulation parameters, including absorption energies, were those provided as default in PRIMO. The tallied PSF produces a dose distribution in water that reproduces well the measured dose profiles for the particular linac used, with a gamma pass rate GPR, i.e., the percentage of voxels that pass gamma analysis [12] with criteria 1%,1 mm, better than 95%. The size of the PSF is 23 Gigabytes. For the patient-dependent part of the linac and the voxelized geometries, DPM was selected as the Monte Carlo radiation transport engine. Simulations were run for 1108 histories in a dual Xeon E5-2670V3 CPU with 12 cores each, and hyper-threading. The simple splitting variance-reduction technique was applied in the patient geometry with a splitting factor of 300. The obtained dose distributions had an average standard statistical uncertainty less than 1% in all cases.
The accuracy of the implemented reconstruction algorithm was assessed by comparing the original dose (reference) with the expected dose i.e., the dose obtained from the simulation of the plan reconstructed from the expected positions (evaluated). The comparison of dose distributions was made by calculating the gamma pass rate with criteria 2%, 1 mm (GPR 2,1) and by evaluating the DVHs percentage of agreement. All the analysis was done with the functions available in the PRIMO system.
Sensitivity of the dose to the magnitude of errors in the position of the MLC leaves was evaluated by using the gamma pass rate (GPR) and the PA. For this purpose, the position errors captured in the dynalog files of the two clinical cases were magnified. Magnification was made by rescaling the errors up to a maximum error Σ. Only errors larger than 0.01 mm were magnified. For scaling, the altered actual" position of a leaf, \(P^\prime _a,\) was calculated as,
Results of the comparison of the original and expected doses are shown in Table 1. The expected plans were reconstructed considering all the records in the dynalog files, i.e., 1536 and 1584 for the prostate and head&neck cases, respectively. Therefore, they describe the treatment dynamics with a higher time resolution than the original plans that included 177 and 194 control points (taken from the DICOM files) for the prostate and head&neck cases, respectively. However, the good agreement of the original dose of these low-resolution plans with the expected dose shown in Table 1, indicates that the impact of time resolution on the dose distribution is negligible. Table 1 also shows the comparison of the expected doses with original doses estimated from original plans in which the number of control points were increased to 1594 and 1561 for the prostate and head&neck cases, respectively. The additional control points were generated by linear interpolation of the MLC leaf positions and of the fractional dose. The agreement in these high-resolution cases is not significantly better than for the low-resolution plans.
The impact on the dose of magnifying leaf position errors ε by conserving its sign in Eq. 3 was small. This can be observed in Table 3 which shows the results of comparing the expected dose with the actual doses estimated for plans in which errors were scaled up to large values of 10 and 30 mm. For Σ=10 mm with RMS of 0.68 and 0.47 mm for the prostate and head&neck cases, respectively, the values obtained for PA and GPR 2,2 are similar to those obtained for the comparison of the original doses with the expected doses. The impact on the dose is however noticeable for Σ=30 mm with RMS of 2.03 and 1.41 mm for the prostate and head&neck cases, respectively.
It was verified that the different time resolution of the original plan with respect to a reconstructed plan that includes all the records of the dynalog files does not have a significant impact in the dose distribution for the clinical cases analyzed in this work. That justifies to make the comparison of the dose obtained from the original low-resolution plan with the dose obtained from a plan reconstructed from (all) the actual positions in the dynalog files and still be valid to attribute dose deviations to errors in leaf positioning during treatment. The advantage of selecting this approach is a faster simulation of the low-resolution plan.
b1e95dc632