Purpose: To establish the first DRLs for contrast-enhanced diagnostic pediatric fluoroscopic procedures in Brazil, stratified by body weight, based on data collected in a single tertiary referral public hospital of Rio de Janeiro city.
Methods: This descriptive, cross-sectional study included 928 diagnostic fluoroscopy examinations performed in patients aged 0-18 years between December 2020 and December 2024 at a national pediatric referral center. Dosimetric parameters such as Air kerma-area product (PKA), Air kerma at patient entrance reference point (Ka,r), fluoroscopy time, and number of images were extracted from automatically generated reports. Results were compared with existing international DRLs.
Results: Local median doses showed overall agreement with European and UK reference levels in lower weight and age groups, but higher values were observed in patients with higher body mass. Compared with other single-center studies, local KAP values were higher, likely reflecting differences in case complexity and exposure parameters. Weight-based stratification proved more reliable than age-based grouping for defining pediatric DRLs, supporting its use as the preferred reference criterion.
Conclusion: This is the first Brazilian study to systematically report DRLs for diagnostic pediatric fluoroscopy using weight-based stratification, that support the need for structured dose monitoring, protocol optimization, and continuous professional training.
Background and purpose: Stereotactic radiosurgery/radiotherapy (SRS/SRT) has emerged as a less invasive alternative to enucleation in the management of ocular malignancies. SRS/SRT planning is time-consuming, complex, and the plan quality depends on the experience of the treatment planner. We demonstrate the use of RapidPlan modeling in conjunction with HyperArc for the treatment of ocular diseases to improve planning efficiency, consistency, and quality.
Materials and methods: A HyperArc-based RapidPlan (HARP) model was iteratively trained on 80 patient datasets that simulated ocular malignancies. Twenty additional patient datasets were reserved for model testing and comparison with manual plans. Each testing dataset was both manually planned and replanned with the final RapidPlan model. Target volumes were defined in the HyperArc module as the GTV and PTV. The PTV was generated by a 2 mm static expansion of the GTV. 25 Gy in 1 fraction was prescribed, and all plans were normalized such that PTVD95% = 25 Gy. Treatment planning was done using Eclipse V16 with the Acuros XB dose engine on a TrueBeam LINAC with Millennium 120 MLCs (5 mm width). All plans underwent EPID-based patient-specific QA and an independent Monte Carlo (MC) second-check.
Results: Model-based plans demonstrated similar degrees of conformality, gradient, and target coverage compared to manual planning. OAR sparing showed statistically significant improvements with model-based plans. Optic nerve Dmax (D0.03 cc) decreased to 4.56 Gy (manual: 7.18 Gy), and lens Dmax decreased to 7.64 Gy (manual: 9.36 Gy). Lacrimal gland mean dose decreased from 11.61 to 6.47 Gy. Modulation factor, monitor units, and beam-on times also all decreased. Patient-specific QA results showed improvements compared to manually generated plans. MC second check results also improved, increasing from an average of 97.81%-98.74%. Plan optimization times decreased significantly from approximately 120 min for manual planning to 15 min on average for model-based plans.
Conclusion: RapidPlan-generated ocular SRS plans showed either comparable performance or improvements in all plan metrics measured, compared to manual planning. Planning times were significantly reduced while maintaining or improving plan quality. Clinical implementation of this HARP model is ongoing at our clinic.
Background: An ionization chamber and electrometer allow measurement of the absorbed dose to water. A sensitivity comparison between electrometers is essential for quality control, and an efficient method is available to accurately measure the electrometer sensitivity coefficient without using a linear accelerator (linac). Although dual- circuit electrometers are becoming increasingly common, no calculation method for the sensitivity coefficient of their second-circuit is available. Hence, we propose a method for calculating this sensitivity coefficient using the first-circuit as the reference and evaluate its accuracy.
Methods: Using the first-circuit of a RAMTEC pro electrometer as a reference, the RAMTEC duo and SuperMAX electrometers were connected as test units to the simple yet accurate Japanese-patented SCG002 current source powered by a dry cell battery. Sensitivity ratio relec was calculated from the average of three charge measurements using RAMTEC Pro. This ratio was multiplied by the calibration coefficient of the first- circuit to obtain the sensitivity coefficient of the second-circuit. The accuracy was obtained from the relative error of each electrometer based on the calibration coefficient (kelec) provided by a standards laboratory.
Results: The sensitivity coefficient of the second-circuit of RAMTEC pro was 1.0004 (relative error, +0.030%). For RAMTEC duo, the first- and second-circuit coefficients were 1.0014 and 1.0013, respectively (relative errors, +0.080% and +0.070%). For SuperMAX, the coefficients were 0.9986 and 0.9983 (relative errors, 0.0% and -0.050%) for the first and second circuits, respectively. Thus, the proposed method provided accurate measurements.
Conclusion: We accurately determine the sensitivity coefficient of the second-circuit in a dual-circuit electrometer using the first-circuit of the same or another electrometer as the reference. If the electrometer performance is verified, the coefficient kelec of the first-circuit is likely applicable to the second-circuit. This method may reduce the costs associated with electrometer calibration in clinical settings.
Background: Deep learning algorithms can synthesize pulmonary functional images from CT images. However, previous studies have only been able to predict either ventilation or perfusion from CT, limiting the holistic evaluation of lung function.
Purpose: This study aimed to develop a deep learning-based framework for simultaneously generating lung perfusion and ventilation images from three-dimensional CT.
Methods: A total of 98 cases who underwent single-photon emission CT perfusion images (SPECT PI) with 99mTc-labeled macroaggregated albumin, ventilation images (VI) with 99mTc-Technegas, and three-dimensional CT images were collected. The three-dimensional CT and SPECT images were registered and cropped to include only the lung parenchyma. A dual-decoder residual attention network (DDRAN) was constructed to generate both PI and VI simultaneously from three-dimensional CT images. For comparative assessment, we additionally employed a conventional single-decoder residual attention network (RAN) to individually generate PI and VI. The structural similarity index (SSIM) and Spearman's rank correlation coefficient (Rs) were utilized to assess voxel-wise agreement. Additionally, the Dice similarity coefficient (DSC) was applied to evaluate function-wise concordance. We used the Wilcoxon signed-rank test to statistically evaluate the differences between the images synthesized by DDRAN and RAN. Beyond image-similarity metrics, we evaluated overall model performance using threshold-based classification. Lastly, a two-part reader study was conducted: (I) qualitative image acceptability for clinical review, and (II) illustrative diagnostic interpretation based on synthesized image pairs alone.
Results: Overall, DDRAN and RAN achieved comparable performance. The average SSIM values of the DDRAN/RAN model were 0.871/0.866 (p < 0.05) for PI and 0.830/0.825 (p < 0.05) for VI, and the Rs values were 0.836/0.819 and 0.732/0.731, respectively. The DDRAN/RAN model achieved average DSC values of 0.795/0.796 for PI and 0.708/0.718 for VI in low-function regions, and 0.857/0.849 for PI and 0.793/0.793 for VI in high-function regions. In the two-part reader study, the synthesized perfusion and ventilation images received almost acceptable scores across all experience levels and demonstrated diagnostic potential.
Conclusions: We have developed a dual-decoder residual attention network that enables the simultaneous synthesis of lung perfusion and ventilation images from three-dimensional CT. Preliminary results indicate moderate-to-high structural-wise and functional-wise concordances, and our proposed model demonstrates comparable accuracy when benchmarked against single-decoder models. The synthesized perfusion and ventilation images can potentially be used for precise diagnosis and guiding functional lung avoidance radiotherapy.
Purpose: The purpose of this work was to commission and validate GPUMCD, a GPU-accelerated Monte Carlo dose calculation engine for c-arm Elekta linear accelerators (linac). This algorithm was recently released for clinical implementation in the Elekta One Treatment Planning System (v6.2.3, EOP).
Methods: A GPUMCD beam model was generated for all photon energies of a VersaHD linac (6X, 6FFF, 10X, 10FFF, 18X). A validated version of the Monaco Commissioning Utility was used to compare calculated percent depth dose (PDD) profiles as well as lateral profiles for open fields against measurements. An adapted MPPG 5.b methodology was used to verify point-doses and 3D dose distributions in homogeneous and heterogeneous media using the ArcCheck, solid water, the CIRS ZEUS phantom, and the IROC HN and spine phantoms.
Results: The average agreement between measured and calculated PDDs and beam profiles using a local 2% dose difference (DD) in the high dose region for fields greater than 5 × 5 cm2 was 98.4% ± 2.3% for all energies. Using a 2% DD and 2 mm distance-to-agreement (DTA) gamma criteria for all fields using a 5% dose threshold yielded an agreement of 99.9% ± 0.5%. For open fields, GPUMCD reduced the calculation time by 93% as compared to X-ray voxel Monte Carlo (XVMC) using the same hardware. All MPPG 5.b. recommended testing was within the suggested tolerance limits. All plan measurements passed at the recommended gamma criteria. GPUMCD heterogeneity agreement and point dose measurements were found to agree within 3%.
Conclusion: The GPUMCD algorithm in EOP was successfully tested and commissioned for clinical use for the VersaHD linac.
The purpose of this study is to assess American Association of Physicists in Medicine (AAPM) members' availability to volunteer, share their expertise, and promote global collaboration in medical physics by partnering with international colleagues who seek support in strengthening their medical physics practice.
The questionnaire was administered to AAPM members to assess their willingness to volunteer, preferred engagement modalities, and availability for different types of volunteer activities. The questionnaire was integrated into the AAPM member profile and covered therapy, diagnostic, and nuclear medicine disciplines, capturing participants' visiting preferences and expertise in teaching and training. Respondents were categorized based on demographic factors such as years of experience, expertise areas, and geographic location.
A total of 409 AAPM members completed and submitted 427 responses to the questionnaire (18 members submitted two discipline-specific sets). The 4% response rate confines this report to exploratory findings rather than predictive conclusions for all AAPM members. Most respondents expressed a preference for short-term engagements, with 49.2% favoring visits of less than one week. Volunteers reported significant expertise, with a substantial portion having more than 15 years of professional experience. Additionally, respondents demonstrated active participation in international activities, with 57 engaged in AAPM initiatives and over 100 involved in private nonprofit or nongovernmental organizations (NGOs).
This report highlights the breadth of expertise and willingness of AAPM members to collaborate and support education across all specialties, contributing to the advancement of global medical physics. While the volunteer data shows significant experience and expertise, targeted recruitment is needed in certain specialized areas. This initiative offers a scalable and sustainable model for global volunteerism in medical physics, complementing traditional in person training and addressing the needs of low- and middle-income countries (LMICs).
Radiochromic film dosimeters are commonly used in radiation oncology in both clinical and research settings to measure the radiation dose with high spatial resolution over a two-dimensional area. Radiation-induced solid-state polymerization occurring in films leads to a change in their opacity. It is critical to evaluate the temporal stability of the response of new film models to ensure an appropriate post-irradiation time is selected for film scanning.
To investigate the time-dependent growth of optical density (OD) in the recently released EBT4 radiochromic films.
EBT4 film samples were irradiated at 0.125, 0.25, 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, and 15 Gy dose levels using a 6 MV X-ray beam produced by a clinical linear accelerator. Two film samples were irradiated at each dose level. The films were repeatedly scanned using a flatbed scanner upon irradiation and at various time intervals for approximately 2 months (1325 h). The net OD was measured, and the films’ sensitivity and calibration curves were obtained at several time points.
The net OD grows with time in all color channels, the rate of which reduces with dose. Red channel OD increased by ∼8%–30% over the studied time period for films irradiated at 15-1 Gy. A similar trend was observed in the green channel with slightly higher (∼1%–2%) growth rates compared to the red channel. Films irradiated at > 1 Gy dose reached over ∼90% of their final net OD within the first 12 h post-irradiation, followed by a gradual growth over the following weeks. The OD-per-Gy decreased with dose and increased with elapsed time post-irradiation.
Since the response of the films continues to develop with time, to adhere to good practice of film dosimetry, the calibration and test films must be read at the same time interval post-irradiation, otherwise significant errors in dose measurement can occur. Since the temporal dynamics of EBT4 and EBT3 films were found to be similar, both models can be treated similarly in terms of the optimum post-irradiation wait-time prior to film readout.
Usability engineering is essential for ensuring the safety and effectiveness of medical software, as design-related issues are a leading cause of use errors in clinical settings. Heuristic evaluation provides a practical approach to identifying usability problems, but its outcomes depend heavily on expert interpretation. Large Language Models (LLMs), such as ChatGPT, offer a potential means to augment heuristic evaluation by generating structured, context-aware usability feedback. This study explored the use of ChatGPT to support heuristic assessment of the Radiation Planning Assistant (RPA), a web-based radiotherapy planning tool designed to support clinical teams in low- and middle-income countries.
ChatGPT was provided with the RPA user and technical guides, training videos for each functional dashboard, and Zhang et al.’s 14 usability heuristics. The model was instructed to score each dashboard according to these heuristics, using Zhang's 0–4 severity scale, and to propose concrete interface improvements. The resulting feedback was reviewed and scored independently by the RPA developer team and by 13 users during a dedicated User Meeting. Comparative analysis was performed between ChatGPT, developer, and user ratings.
ChatGPT identified 26 potential usability issues across six heuristic domains. The developer team considered nine of these actionable, though all were classified as minor (severity ≤ 2). User ratings showed wide variability, with nine suggestions achieving mean scores ≥ 1.5. Qualitative agreement between users and developers was limited, underscoring the importance of diverse perspectives in heuristic evaluation. Three suggestions—enhanced upload logs, reversible actions (“reopen request”), and stronger error prevention—were rated as potentially high priority by a minority of users. ChatGPT's ratings were consistent across dashboards.
While ChatGPT did not reveal any critical usability failures, its heuristic assessment proved valuable in prompting discussion, identifying minor refinements, and enriching both developer and user engagement with the RPA's interface design. This study demonstrates that LLMs can serve as an effective, low-cost complement to conventional heuristic evaluation, supporting early-stage usability review and stakeholder dialogue in the development of medical software.
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：