An artificial neural network based approach for predicting the proton beam spot dosimetric characteristics of a pencil beam scanning technique.

Utilising Machine Learning (ML) models to predict dosimetric parameters in pencil beam scanning proton therapy presents a promising and practical approach. The study developed Artificial Neural Network (ANN) models to predict proton beam spot size and relative positional errors using 9000 proton spot data. The irradiation log files as input variables and corresponding scintillation detector measurements as the label values. The ANN models were developed to predict six variables: spot size in thex-axis,y-axis, major axis, minor axis, and relative positional errors in thex-axis andy-axis. All ANN models used a Multi-layer perception (MLP) network using one input layer, three hidden layers, and one output layer. Model performance was validated using various statistical tools. The log file recorded spot size and relative positional errors, which were compared with scintillator-measured data. The Root Mean Squared Error (RMSE) values for the x-spot and y-spot sizes were 0.356 mm and 0.362 mm, respectively. Additionally, the maximum variation for the x-spot relative positional error was 0.910 mm, while for the y-spot, it was 1.610 mm. The ANN models exhibit lower prediction errors. Specifically, the RMSE values for spot size prediction in the x, y, major, and minor axes are 0.053 mm, 0.049 mm, 0.053 mm, and 0.052 mm, respectively. Additionally, the relative spot positional error prediction model for the x and y axes yielded maximum errors of 0.160 mm and 0.170 mm, respectively. The normality of models was validated using the residual histogram and Q-Q plot. The data over fit, and bias were tested using K (k = 5) fold cross-validation, and the maximum RMSE value of the K fold cross-validation among all the six ML models was less than 0.150 mm (R-Square 0.960). All the models showed excellent prediction accuracy. Accurately predicting beam spot size and positional errors enhances efficiency in routine dosimetric checks.

Late urinary toxicity and QOL with pelvic radiotherapy for high-risk prostate cancer: Dose-effect relations in the POP-RT randomized phase III trial.

PURPOSE: The POP-RT phase III randomized trial showed improved biochemical failure-free survival and metastasis-free survival with whole pelvic radiotherapy (WPRT) versus prostate-only radiotherapy (PORT) for high/very high-risk prostate cancer, albeit with worse RTOG late urinary toxicity. We report updated late urinary adverse effects and bladder dose-effect relations within this trial. MATERIALS/METHODS: Late urinary toxicity, and cumulative severity of each symptom over the follow-up period was graded using CTCAE v5.0. Bladder dosimetry in 5-Gy increments (V5, V10, V15...V65Gy, V68Gy) in the approved radiotherapy plans was compared for urinary symptoms and overall grade 2+ toxicity. Potential factors influencing urinary toxicity were tested using multivariable logistic regression analysis. Updated urinary quality of life (QOL) scores were compared between the trial arms. RESULTS: Complete combined data for late urinary symptoms and dosimetry was available for 193/224 patients. At a median follow-up of 75 months, cumulative late urinary CTCAE grade 3 toxicity was low and similar for WPRT and PORT (5.2% vs 4.1%, p=0.49), while grade 2 toxicity was 31.3% vs 22.7% respectively (p=0.12). Cumulative rates of each urinary symptom were similar between both arms. Multivariable analysis with age at diagnosis, known diabetes, tumor stage, trial arm, prior TURP, grade 2+ acute urinary toxicity, low bladder dose (V10Gy) and moderate bladder dose (V40Gy) did not identify any significant association with late urinary toxicity. Urinary QOL scores was similar between both the arms for all the symptoms. CONCLUSION: Over long term follow up, whole pelvic radiotherapy resulted in low (∼5%) and similar grade 3 cumulative urinary toxicity as prostate-only radiotherapy. The long term patient-reported QOL scores were similar. No causative factors affecting the late urinary toxicity were identified.

Moving towards process-based radiotherapy quality assurance using statistical process control.

Monitoring Radiotherapy Quality Assurance (QA) using Statistical Process Control (SPC) methods has gained wide acceptance. The significance of understanding the SPC methodologies has increased among the medical physics community with the release of Task Group (TG) reports from the American Association of Physicists in Medicine (AAPM) on patient-specific QA (PSQA) (TG-218) and Proton therapy QA (TG-224). Even though these reports recommend using SPC for QA analysis, physicists have ambiguities and doubts in choosing proper SPC tools and methodologies. This review article summarises the utilisation of SPC methods for different Radiotherapy QAs published in the literature, such as PSQA, routine Linac QA and patient positional verification. QA analysis using SPC could assist the user in distinguishing between 'special' and 'routine' sources of variations in the QA, which can aid in reducing actions on false positive QA results. For improved PSQA monitoring, machine-specific, site-specific, and technique-specific Tolerance Limits and Action Limits derived from a two-stage SPC-based approach can be used. Adopting a combination of Shewhart's control charts and time-weighted control charts for routine Linac QA monitoring could add more insights to the QA process. Incorporating SPC tools into existing image review modules or introducing new SPC software packages specifically designed for clinical use can significantly enhance the image review process. Proper selection and having adequate knowledge of SPC tools are essential for efficient QA monitoring, which is a function of the type of QA data available, and the magnitude of process drift to be monitored.

Quantitative Comparison of Dosimetric Data Between the Indigenous Baseplate and Commercially Available Carbon Fiber Baseplate for 6 and 15 MV Photon Energy.

Background: This study aims to design an indigenous baseplate (ID baseplate) that is economically viable and dosimetrically comparable for radiotherapy patient treatment. An ID baseplate was designed and manufactured using wood plastic composition materials that are readily available in the market and were compared dosimetrically with the commercially available carbon fiber baseplate (CF baseplate). Materials and Methods: Surface dose and beam attenuation properties of both the baseplates (ID and CF) were measured using a parallel plate chamber and compared with the dose calculated from the treatment planning system (TPS). Separate computer tomography images of both the baseplates were acquired by placing solid water phantoms. These images were used for surface dose calculation in the TPS and were validated with experimental measurements. Proper densities were assigned to the couch and baseplates to avoid uncertainties in dose calculations. All measurements were performed at field sizes 10 cm × 10 cm for 6 MV and 15 MV photon beams. Results: The percentage surface dose measured for the ID baseplate and CF baseplate was found to be matching for 6 MV beam (98.2% and 97%, respectively); however, for the 15 MV beam, the ID baseplate showed a higher surface dose of 98.6% compared to CF baseplate (87.4%). For the ID baseplate, the percentage difference in the surface dose between that TPS calculated value and the measured values were 1.6% and 1.4% for 6MV and 15MV, respectively. The ID baseplate showed higher beam attenuation than the CF baseplate by 2.2% for the 6MV beam and 3.4% for the 15MV beam when proper electron densities were not assigned. The difference between the TPS calculated dose and delivered dose was achieved within 3% after assigning proper electron density to the couch and baseplate. Conclusions: The ID baseplate has shown acceptable dosimetric results and can be an economically viable alternative to the commercially available CF baseplates. The manufacturing cost of the ID baseplate was ten times cheaper than the CF baseplate.

Acute and Late Adverse Effects of Prostate-Only or Pelvic Stereotactic Radiation Therapy in Prostate Cancer: A Comparative Study.

PURPOSE: To compare the urinary and gastrointestinal adverse effects with or without the inclusion of pelvic nodal regions in patients treated with extreme hypofractionated stereotactic radiation therapy (SBRT) for prostate cancer. METHODS AND MATERIALS: Patients treated with definitive SBRT for nonmetastatic adenocarcinoma prostate were identified from prospectively maintained institutional database, and details of radiation therapy volume, dose, acute, and late adverse effects were analyzed. Symptoms of acute (within 90 days of completing SBRT) and late gastrointestinal and urinary toxic effects were graded using Common Terminology Criteria for Adverse Effects version 5.0. Each symptom was scored according to the worst reported grading during treatment and the follow-up period. Cumulative rates of adverse effects between prostate-only SBRT (PO-SBRT) and whole pelvic SBRT (WP-SBRT) were compared using the χ2 test. Univariable and multivariable analysis was performed for possible factors affecting acute gastrointestinal and late urinary toxic effects. RESULTS: A total of 220 patients were analyzed (PO-SBRT = 118, WP-SBRT = 102), with a median follow-up of 28 months (interquartile range, 14-40). Most patients had locally advanced disease (PO-SBRT 60% high risk and 40% intermediate risk, WP-SBRT 79% node positive, and 21% high risk). The median SBRT dose was 36.25Gy (interquartile range, 35-36.25) to the prostate (2-Gy equivalent, EQD2 = 90.6Gy, a/b = 1.5Gy) and simultaneous integrated 25Gy to the pelvis (EQD2 = 46.3Gy) in 5 fractions on alternate days. No grade 3 to 4 acute adverse effects were observed except 1 grade 3 urinary obstruction (PO-SBRT). WP-SBRT was associated with a significantly higher rate of acute grade 2 gastrointestinal toxic effects (29.4% vs 14.7%, P = .008) and late grade 2 urinary adverse effects (45.6% vs 25.0%, P = .003). Both groups had low incidence of late grade 3 adverse effects (urinary 2.5%, gastrointestinal 1%). CONCLUSIONS: WP-SBRT was associated with significantly higher acute gastrointestinal and late urinary adverse effects compared with PO-SBRT, although overall incidence of severe adverse effects was low.

Validation of Dolphin dosimetry in three dimensional patient-specific quality assurance programme.

AIM: The aim of this study is to commission and validate Dolphin-Compass dosimetry as a patient-specific Quality Assurance (QA) device. BACKGROUND: The advancement of radiation therapy in terms of highly conformal delivery techniques demands a novel method of patient-specific QA. Dolphin-Compass system is a dosimetry solution capable of doing different QA in radiation therapy. MATERIALS AND METHODS: Dolphin, air-vented ionization detector array mounted on Versa-HD Linear Accelerator (LINAC) was used for measurements. The Compass is a dose computation algorithm which requires modelling of LINAC head similar to other Treatment Planning Systems (TPS). The dosimetry system was commissioned after measuring the required beam data. The validation was performed by comparison of treatment plans generated in Monaco TPS against the measurement data. Different types of simple, complex, static and dynamic radiation fields and highly conformal treatment plans of patients were used in this study. RESULTS: For all field sizes, point doses obtained from Dolphin-Compass dosimetry were in good agreement with the corresponding TPS calculated values in most of the regions, except the penumbra, outside field and at build-up depth. The results of gamma passing rates of measurements by using different Multi-leaf Collimator patterns and Intensity Modulated Radiation Therapy fluence were also found to be in good correlation with the corresponding TPS values. CONCLUSIONS: The commissioning and validation of dosimetry was performed with the help of various fields, MLC patterns and complex treatment plans. The present study also evaluated the efficiency of the 3D dosimetry system for the QA of complex treatment plans.