• 2022-09
  • 2022-08
  • 2022-07
  • 2022-05
  • 2022-04
  • 2021-03
  • 2020-08
  • 2020-07
  • 2018-07
  • br Patient iframe width height src https www youtube


    2.7. Patient-specific QA
    Prior to plan approval at the treatment console, patient-specific quality assurance (QA) of the adapted plan was performed using an independent Monte-Carlo dose calculation algorithm and gamma ana-lysis (3%/3 mm) [27–30]. The Monte-Carlo engine for QA purposes uses phase space data recorded in a plane just above the MLC and the transport in the patient is loosely based on the DPM Monte-Carlo code [31]. It used the same beam parameters, segments shapes and electron density map as the treatment plan made with the MRIdian, resulting in a second 3D dose distribution. At each fraction a pdf-report was gen-erated including gamma pass-rates and gamma mean values for the comparison of both dose distributions. In addition, other plan para-meters related to the IMRT modulation in the plan were also reported [27].
    2.8. Patient reported questionnaires
    From the start of clinical MRgRT, patient experiences were assessed using an in-house developed patient-reported outcome questionnaire (PRO-Q) [32]. From July 2016 till December 2017 we collected 89 questionnaires in prostate cancer patients. This PRO-Q included
    Fig. 2. Workflow for MRgRT with online plan TRIzol for prostate cancer. HR = high resolution, MR = magnetic resonance, CTV = clinical target volume, OAR = organs at risk, QA = quality assurance.
    questions on potential MR-related complaints and experiences, such as anxiety, temperature, and noise. These items could be scored on a 4-point scale as: “not at all”, “a little”, “moderate”, and “very much”. PRO-Qs were collected once, immediately following the last MRgRT fraction, taking the completion of the PRO-Q on average 5 min.
    3. Results
    3.1. Target coverage and patient-specific QA
    Due to common manual adjustment of the CTV and the 3 mm PTV margin used, the predicted plan is generally suboptimal particularly for target coverage. In 97% (N = 677/700 fractions) of all fractions for prostate MRgRT the plan has been re-optimized. All adapted treatment plans have passed the patient-specific QA and the obtained average γ-pass rate for all 700 adapted fractions is 99.8 ± 0.1%, with γ mean = 0.38 ± 0.01.
    3.2. Treatment delivery
    Beam-on delivery treatment time was on average 10 min and con-stituted approximately one quarter of the total treatment duration. At the onset of treatment delivery a brief cine movie of 10 s duration was performed at a single sagittal plane (4 frames-per-second, slice thick-ness 5 mm) previously selected by the physician in order to check the tracking accuracy (Fig. 4). At the same time, it was verified that the position of the CTV had not changed from the first 3D MR-scan at the beginning of the fraction. Gated IMRT delivery was performed using a 3 mm gating boundary around the CTV. The system automatically shut off radiation delivery when the system detected that more than 7% (institute specific setting) of the CTV area is outside of the gating boundary (PTV) during MR-planar acquisition for intra-fraction mon-itoring. Prostate drifts and intra-fraction prostate rotation/deformation led to application of 2D shifts during treatment delivery in more than 20% of all delivered fractions (149/700 fractions). Larger prostate shifts requiring repeat 3D imaging were observed in approximately 6% of fractions (39/700 fractions).
    On average, the duration of an uneventful MRgRT fraction is ap-proximately 45 min (range for all patients, 40–70 min). An overview of 
    the relative duration of all the steps in our MRgRT TRIzol workflow is shown in Table 1, being recontouring the step which took the longest.
    3.3. Patient experiences