Automated segmentation in pelvic radiotherapy: A comprehensive evaluation of ATLAS-, machine learning-, and deep learning-based models.

Artificial intelligence can standardize and automatize highly demanding procedures, such as manual segmentation, especially in an anatomical site as common as the pelvis. This study investigated four automated segmentation tools on computed tomography (CT) images in female and male pelvic radiotherapy (RT) starting from simpler and well-known atlas-based methods to the most recent neural networks-based algorithms. The evaluation included quantitative, qualitative and time efficiency assessments. A mono-institutional consecutive series of 40 cervical cancer and 40 prostate cancer structure sets were retrospectively selected. After a preparatory phase, the remaining 20 testing sets per each site were auto-segmented by the atlas-based model STAPLE, a Random Forest-based model, and two Deep Learning-based tools (DL), MVision and LimbusAI. Setting manual segmentation as the Ground Truth, 200 structure sets were compared in terms of Dice Similarity Coefficient (DSC), Hausdorff Distance (HD), and Distance-to-Agreement Portion (DAP). Automated segmentation and manual correction durations were recorded. Expert clinicians performed a qualitative evaluation. In cervical cancer CTs, DL outperformed the other tools with higher quantitative metrics, qualitative scores, and shorter correction times. On the other hand, in prostate cancer CTs, the performance across all the analyzed tools was comparable in terms of both quantitative and qualitative metrics. Such discrepancy in performance outcome could be explained by the wide range of anatomical variability in cervical cancer with respect to the strict bladder and rectum filling preparation in prostate Stereotactic Body Radiation Therapy (SBRT). Decreasing segmentation times can reduce the burden of pelvic radiation therapy routine in an automated workflow.

Dosimetric commissioning and quality assurance of scanned ion beams at the Italian National Center for Oncological Hadrontherapy.

PURPOSE: To describe the dosimetric commissioning and quality assurance (QA) of the actively scanned proton and carbon ion beams at the Italian National Center for Oncological Hadrontherapy. METHODS: The laterally integrated depth-dose-distributions (IDDs) were acquired with the PTW Peakfinder, a variable depth water column, equipped with two Bragg peak ionization chambers. fluka Monte Carlo code was used to generate the energy libraries, the IDDs in water, and the fragment spectra for carbon beams. EBT3 films were used for spot size measurements, beam position over the scan field, and homogeneity in 2D-fields. Beam monitor calibration was performed in terms of number of particles per monitor unit using both a Farmer-type and an Advanced Markus ionization chamber. The beam position at the isocenter, beam monitor calibration curve, dose constancy in the center of the spread-out-Bragg-peak, dose homogeneity in 2D-fields, beam energy, spot size, and spot position over the scan field are all checked on a daily basis for both protons and carbon ions and on all beam lines. RESULTS: The simulated IDDs showed an excellent agreement with the measured experimental curves. The measured full width at half maximum (FWHM) of the pencil beam in air at the isocenter was energy-dependent for both particle species: in particular, for protons, the spot size ranged from 0.7 to 2.2 cm. For carbon ions, two sets of spot size are available: FWHM ranged from 0.4 to 0.8 cm (for the smaller spot size) and from 0.8 to 1.1 cm (for the larger one). The spot position was accurate to within ± 1 mm over the whole 20 × 20 cm(2) scan field; homogeneity in a uniform squared field was within ± 5% for both particle types at any energy. QA results exceeding tolerance levels were rarely found. In the reporting period, the machine downtime was around 6%, of which 4.5% was due to planned maintenance shutdowns. CONCLUSIONS: After successful dosimetric beam commissioning, quality assurance measurements performed during a 24-month period show very stable beam characteristics, which are therefore suitable for performing safe and accurate patient treatments.

Dosimetric accuracy of a treatment planning system for actively scanned proton beams and small target volumes: Monte Carlo and experimental validation.

This study was performed to evaluate the accuracy of a commercial treatment planning system (TPS), in optimising proton pencil beam dose distributions for small targets of different sizes (5-30 mm side) located at increasing depths in water. The TPS analytical algorithm was benchmarked against experimental data and the FLUKA Monte Carlo (MC) code, previously validated for the selected beam-line. We tested the Siemens syngo(®) TPS plan optimisation module for water cubes fixing the configurable parameters at clinical standards, with homogeneous target coverage to a 2 Gy (RBE) dose prescription as unique goal. Plans were delivered and the dose at each volume centre was measured in water with a calibrated PTW Advanced Markus(®) chamber. An EBT3(®) film was also positioned at the phantom entrance window for the acquisition of 2D dose maps. Discrepancies between TPS calculated and MC simulated values were mainly due to the different lateral spread modeling and resulted in being related to the field-to-spot size ratio. The accuracy of the TPS was proved to be clinically acceptable in all cases but very small and shallow volumes. In this contest, the use of MC to validate TPS results proved to be a reliable procedure for pre-treatment plan verification.

Prospective intervention trial of a screening protocol to identify fetal trisomy 18 using maternal serum alpha-fetoprotein, unconjugated oestriol, and human chorionic gonadotropin.

Two prenatal centres in New England, routinely using a screening protocol for fetal Down syndrome that included maternal serum alpha-fetoprotein (AFP), unconjugated oestriol (uE3), and human chorionic gonadotropin (hCG) measurements in combination with maternal age, adopted a separate screening protocol for trisomy 18. That protocol identified a pregnancy as being at high risk when AFP, uE3, and hCG measurements all fell at or below specified cut-offs (0.75, 0.60, and 0.55 multiples of the median, respectively), regardless of maternal age. Among the first 19,491 women screened, 98 (0.5 per cent) were found to have values which placed them in the high-risk category. Four of these women were subsequently found not to be pregnant. In two others, samples from non-pregnant individuals were found to have been incorrectly submitted for analysis in place of the samples from the pregnant women. All of the remaining 92 women were counselled and offered amniocentesis and fetal karyotyping. Eighty-eight (96 per cent) accepted. Karyotypes or birth outcomes were available on all 92 pregnancies. Six cases of trisomy 18 and one case of Turner syndrome were identified by karyotype. One case of trisomy 18 was identified for every 14 unaffected pregnancies offered amniocentesis. In the present prospective study, an estimated 85 per cent of the cases of trisomy 18 were identified. However, given the small number of cases (six), the 95 per cent confidence interval for the detection rate is broad (40-95 per cent).

Effect of atenolol on car drivers in a prolonged stress situation.

The effects of atenolol were investigated in car drivers participating in an amateur car rally. Atenolol or placebo were administered to two groups of 20 crews (driver and assistant) in this double-blind randomized trial. When compared with placebo, there was no deterioration in subjective appraisal of alertness and reaction ability in subjects given atenolol, and stress symptoms were markedly reduced. When compared with previous experience, the incidence of improved subjective appraisal of general driving ability was higher with atenolol (30 reports) than with placebo (5 reports). Furthermore the incidence of unchanged or worsened driving ability was significantly higher (p less than 0.001) with placebo (34 reports) than with atenolol (9 reports). At the end of the rally (16 hours after administration) subjects receiving atenolol 50 mg had significant reductions in heart rate (p less than 0.001), diastolic pressure (p less than 0.01), and systolic pressure (p less than 0.01) when compared with placebo.