Purpose: The Geant4-DNA source code was locally modified by users to enable water radiolysis at various temperatures and pH by previous work, however, without linking those two parameters. This study aimed to integrate a quantitative correlation between temperature and pH in a user-modified local version of the Geant4-DNA. Methods: We quantified the temperature-pH relationship via polynomial regression and integrated it into the Geant4-DNA chemistry constructor. The integration was validated using a user-modified local version of Geant4-DNA (version 11.3.0), simulating isotropic irradiation of a 1.0 km3 water cube (pH 7) by a central point source across 25-150 degrees C. Results were compared against literature data. Results: The effects of temperature and pH became observable at around 100 ns (the end of the non-homogeneous stage), as pH mainly affects reactions between generated and background species. While pure water pH is 7.0 at 25 degrees C, it decreases at higher temperatures, shifting away from the value of 7.0. When accounting for this temperature-pH interdependence, higher temperatures amplify the deviation from pH 7, thereby enhancing ionic reactions after the intra-spur and non-homogeneous stages. In our simulations of water radiolysis, the resulting G-values agree with experimental measurements within 15.33% and with other simulated data within 11.99%, with uncertainties below +/- 0.60%. The trends we observe in these G-values also closely mirror those reported in the literature. Conclusions: The successful integration of the temperature-pH correlation in a custom version of Geant4-DNA chemistry module was demonstrated. This integration improves the physical realism of water radiolysis and contributes to various scientific fields.

Objective.RapidBrachyMCTPS is a Monte Carlo (MC)-based treatment planning system for advanced brachytherapy dosimetry applications. In this work, we present the latest updates in RapidBrachyMCTPS v3.0 and benchmark its dose calculations against published guidelines.Approach.RapidBrachyMCTPS v3.0 introduces a major redevelopment of RapidBrachyMC, its underlying MC dose calculation engine. Simulation times were reduced, greater user control over simulation geometry was enabled, and material definitions were greatly simplified and standardized. Test cases developed by the working group on model-based dose calculation algorithms in brachytherapy (WGDCAB) were used to benchmark dose calculation accuracy and quantify improvements in simulation efficiency relative to v2.0.Main results.For four water phantom-based WGDCAB test cases, mean and root-mean-square dose differences between reference data and RapidBrachyMCTPS v3.0-calculated dose distributions were within the estimated total uncertainty budget of 0.4%. Target structure dose-volume histogram metrics for site-specific WGDCAB test cases, one for gynecological tandem and ring brachytherapy and the other for interstitial breast brachytherapy, showed generally good agreement, within 0.01 Gy or 0.10 percentage points, with the same metrics calculated from reference dose distributions. Observed simulation run times for test cases were on average between1.63±0.02and1.66±0.03times faster than v2.0 without an applicator model and between1.45±0.01and1.91±0.02times faster with an applicator model, depending on the applicator modeling scheme used.Significance.RapidBrachyMCTPS v3.0 was well-validated to deliver fast, accurate MC dose calculations alongside robust treatment planning capabilities. It remains freely available to the research community.

Objective.In radioembolization, SPECT/CT planning scans are often acquired during free breathing, which can introduce motion-related blurring and misregistration between SPECT and CT, leading to dosimetric inaccuracies. This study quantifies the impact of respiratory motion on absorbed dose metrics-tumor-to-normal tissue (T/N) ratio, dose volume histograms, and mean dose-using several voxel-based dosimetry methods. This study supports standardization efforts through experimental measurements using a motion-enabled phantom.Approach.Motion effects in pre-therapy imaging were evaluated using a Jaszczak phantom filled with technetium-99 m (99mTc), simulating activity in lesion and background volumes. SPECT/CT scans were acquired with varying cranial-caudal motion amplitudes from the central position ± 0, ± 5, ± 6.5, ± 10, ± 12.5, and ± 15 mm. The impact of motion-related misregistration during scanning on dosimetry was also examined. Five dosimetry methods, including Monte Carlo simulation with uniform reference activity (MC REF), Monte Carlo simulation based on SPECT images (MC SPECT), SimplicityTM(Boston Scientific), local deposition method, and voxel-S-value convolution. Absorbed dose metrics of mean dose, dose volume histogram dosimetric indices (D50, D70, D90), and T/N ratio were obtained to quantify motion effects and evaluate clinical suitability.Main results.Mean absorbed dose values for the lesion and background were consistent across methods within uncertainties, though discrepancies were noted in non-lesion low-density regions. Respiratory motion reduced lesion dose by 16%-25% and increased background dose by 13%-32%, although the latter represented only a 1-2 Gy change. These shifts led to a 28%-43% decrease in the T/N ratio at ± 12.5 mm motion amplitude. Misregistration due to motion also significantly impacted dosimetric accuracy.Significance.The study demonstrated agreement between five dosimetry methods and revealed that respiratory motion can lead to substantial underestimation of the lesion dose and T/N ratio. Since T/N ratio is critical for patient selection and activity prescription, accounting for respiratory motion is essential for accurate radioembolization dosimetry.

Objective. Diffusing alpha-emitters radiation therapy (Alpha-DaRT) involves implanting stainless-steel seeds containing 70-200 kBq of224Ra into solid tumors. The decay of224Ra generates alpha particles and short-lived alpha-emitting atoms, which diffuse and deposit absorbed dose up to millimeters from the seed. As magnetic res- onance imaging (MRI)-guided brachytherapy is commonly used in colorectal cancer, it has also been adopted in pre-clinical orthotopic intra-rectal studies to investigate the therapeutic potential of Alpha-DaRT. However, the metallic seeds cause metal artifacts in MRI, distorting signals and degrading images. Overcoming these artifacts is essential for accurate tumor assessment and treatment planning. This study aimed to reduce metal artifacts in images acquired with 7 T pre-clinical MRI in the presence of Alpha-DaRT seeds using an orthotopic colorectal adenocarcinoma animal model.Approach. Inert seeds were imaged in gelatin phantoms and the dorsal cavity of a mouse carcass. For thein-vivoanimal model, NOD scid gamma mice with HT-29 colorectal adenocarcinoma tumors (∼5-7 mm) were injected with inert seeds. MRI sequence parameters such as echo and repetition times, slice thickness, and readout bandwidth were tested using the phantom and refined with carcass imaging, leading toin-vivoimaging.Main results. Multiple sequence protocols were tested on gelatin phantoms and compared to the original T2-weighted turbo spin echo (TSE) sequence protocol. An improved T2-weighted TSE sequence protocol reduced metal artifact volumes in the gelatin phantom from 147.8 ± 31.0 mm3to 87.0 ± 22.4 mm3in the axial slices, resulting in a 41% reduction. Validation in the mouse carcass confirmed high-quality soft-tissue imaging.In-vivoimages with live mice showed a statistically significant reduction (p<0.001) in metal artifact volume with the improved sequence.Significance. A T2-weighted turbo spin echo protocol effectively mitigated metal artifacts from Alpha-DaRT seeds in a colorectal adenocarcinoma animal model, allowing for clearer visualization of seed placement within the tumor and improving the accuracy of pre-clinical studies.

Quantitative PET studies require a measurement of the arterial input function (AIF), the time-dependent radiotracer concentration in arterial blood plasma. Several groups are developing non-invasive detectors to measure the AIF from the radial artery. This study quantifies the depth and cross-sectional area of the radial artery and accompanying veins at different wrist positions using ultrasound. These anatomical data will guide the design of a non-invasive, wrist-worn detector for AIF acquisition—a practical, patient-friendly alternative to invasive blood sampling. Ultrasound imaging of the wrist was performed on 154 healthy individuals at specified distances from the distal wrist crease (2 cm, 4 cm, and 6 cm). The depths of the radial artery at distances of 2 cm, 4 cm, and 6 cm from the distal wrist crease are 3.36 (1.25) mm, 4.08 (1.81) mm, and 4.66 (2.23) mm, respectively (mean (SD)). Similarly, the cross-sectional areas of the radial artery at these distances are 4.23 (1.75) mm$$^{2}$$, 3.92 (1.71) mm$$^{2}$$, and 3.90 (1.88) mm$$^{2}$$, respectively. The radial artery becomes larger and more superficial near the wrist, suggesting a radiation detector be placed 2 cm from the distal wrist crease on the left arm, where it is generally more superficial than on the right.

The HECKTOR 2025 challenge provides a platform to benchmark automatic segmentation methods for Head and Neck (H N) primary tumors and lymph nodes in FDG-PET and CT scans (Task 1). This study presents a challenge submission based on a Residual Encoder U-Net within the nnU-Net framework, enhanced with modality-specific preprocessing and data augmentations. The proposed solution, submitted under the user name sebquet, achieved mean Dice scores of 71.81 https://github.com/sebquetin/Hecktor2025.git .

Background and Purpose:Brachytherapy is a highly conformal and cost-effective radiotherapy modality, yet its clinical utilization has declined in multiple regions. This study investigated trends in brachytherapy utilization across different Canadian provinces from 2011 to 2020. Materials and Methods:A national survey was distributed to medical physicists in cancer centres across ten Canadian provinces, collecting data on types of brachytherapy, annual number of treatments, clinical indications, and logistical factors. In Québec, treated clinical indications were obtained through a complementary survey of all radiotherapy centres, extending previously published provincial brachytherapy data. Results:Out of 39 radiotherapy centres in Canada conducting brachytherapy treatments, 25 centres were included in this study. HDR accounted for the majority of treatments (60%-100%) and increased in most provinces, while LDR declined across reporting provinces; PDR comprised ≤ 10% of treatments and was limited to Alberta and Ontario. After adjusting for indication-specific cancer incidence and combining HDR and LDR brachytherapy, utilization trends showed decreases in Alberta and British Columbia, increases in Ontario, Nova Scotia, and Saskatchewan, and no significant change in Manitoba, Québec, or nationally. Conclusions:This study revealed significant regional variations in brachytherapy utilization and clinical practices across Canadian radiotherapy centres. Factors influencing these trends include reimbursement structures, personnel and infrastructure availability, and clinician preferences. The use of HDR modalities is increasing nationally, while LDR is declining. Despite significant regional differences, brachytherapy utilization in Canada was largely sustained over the decade from 2011 to 2020.

PURPOSE:To develop clinically relevant test cases for commissioning Model-Based Dose Calculation Algorithms (MBDCAs) for 192Ir High Dose Rate (HDR) gynecologic brachytherapy following the workflow proposed by the TG-186 report and the WGDCAB report 372. ACQUISITION AND VALIDATION METHODS:Two cervical cancer intracavitary HDR brachytherapy models were developed based on a real patient, using either uniformly structured regions or realistic segmentation. The patient's computed tomography (CT) images were processed, converted to a series of digital imaging and communications in medicine (DICOM) CT images, and imported into two treatment planning systems (TPSs), the Oncentra Brachy and BrachyVision. The original segmentation of the clinical case was augmented to enable a thorough dosimetric analysis. The actual clinical treatment plan was generally maintained, with the source replaced by a generic 192Ir HDR source. Dose to medium in medium calculations were performed using the MBDCA option of each TPS, and three different Monte Carlo (MC) simulation codes. MC results demonstrated agreement within statistical uncertainty, while comparisons between the commercial TPS MBDCAs and a general-purpose MC code highlighted both the advantages and limitations of the studied MBDCAs, suggesting potential approaches to overcome the challenges. DATA FORMAT AND USAGE NOTES:The datasets for the developed cases are available online at https://doi.org/10.5281/zenodo.15720996. The DICOM files include the treatment plan for each case, TPS, and the corresponding reference MC dose data. The package also contains a TPS- and case-specific user guide for commissioning the MBDCAs, as well as files necessary to replicate the MC simulations. POTENTIAL APPLICATIONS:The provided datasets and proposed methodology can serve as a commissioning framework for TPSs that employ MBDCAs, as well as a benchmark for brachytherapy researchers using MC methods and MBDCA developers. They also facilitate intercomparisons of MBDCA performance and provide a quality assurance resource for evaluating future TPS software updates.

This review shares the ongoing work of the global Worldwide Innovative Network (WIN) Consortium for Precision Medicine to synthesize emerging cancer treatment data and to define the requirements for a common global cancer database that can truly support precision oncology. We performed a narrative review of emerging cancer treatment data, molecular profiling technologies, and existing clinicogenomic databases, focusing on how tumors are characterized, how subgroups are defined, and how demographic, lifestyle, and environmental factors are captured. The growth in molecular profiling technologies and the development of new targeted therapies are transforming cancer care. Tumors, regardless of tissue origin, are increasingly defined as composites of multiple, often rare, subgroups, each with distinct biology and likely response to specific therapies, based on multidimensional profiling of the tumor and its microenvironment. The solution lies in building vast databases that capture racial and ethnic diversity, reflected in genomic data, as well as diet and lifestyle factors that may have epigenetic impact on gene expression and post-translational modifications. A truly inclusive and informative data set must reflect global diversity, and there are multiple examples of demography-dependent differences in genomic signals. With members caring for and studying patients with cancer across five continents, WIN is actively exploring pathways to create a global cancer database, rich in clinical and molecular detail, granular enough for precise analysis, and large enough to power artificial intelligence-driven insights, provided appropriate data quality, validation, and governance frameworks are in place. This review surveys the current landscape and outlines practical paths forward to achieve this goal.

Background and purpose: Although radiotherapy response involves biological processes shared across cancers, most biomarker studies rely on small, disease-specific sample sizes that limit the reliability of prognostic models. This study evaluated whether pan-cancer data can increase the information available for estimating genomic associations with progression after radiotherapy, based on Molecular Signatures Database hallmark gene sets applied to The Cancer Genome Atlas patients.Methods and materials: Cross-cancer predictive potential was evaluated in cancer types with at least 100 patients who received radiotherapy and had information on gene expression, progression, age, stage, and gender: breast invasive carcinoma (BRCA, n=553), cervical squamous cell carcinoma (CESC, n=183), head-neck squamous cell carcinoma (HNSC, n=320), low-grade glioma (LGG, n=316), lung adenocarcinoma (LUAD, n=104), skin cutaneous melanoma (SKCM, n=125), thyroid cancer (THCA, n=326), and uterine corpus endometrial carcinoma (UCEC, n=257). Gene set expression scores were calculated at the single-sample level. For each evaluation cancer type, scores from all other types were used to fit pan-cancer Cox proportional hazards models. Linear predictions from these models were evaluated in per-cancer prognostication via Kaplan Meier curves, Cox proportional hazards (CoxPH) estimates, including adjustment for age, stage, and gender, and optimism-corrected C-statistic and Nagelkerke pseudo-R2.Results: In Kaplan Meier analysis, pan-cancer scores predicted progression in all cancer types, with varying statistical significance. Univariable CoxPH models also demonstrated consistent association with progression, again with varying significance. Associations persisted when adjusted for clinical variables, and model performance metrics indicated the potential to improve per-cancer prognostication. Associations were particularly strong for BRCA, CESC, HNSC, LGG, and LUAD, with statistically insignificant results for SKCM, THCA, and UCEC.Conclusion: Pan-cancer gene expression data can predict per-cancer progression after radiotherapy, with variable strength across cancer types. These findings support the use of cross-cancer genomic information to address sample size limitations that affect radiotherapy biomarker development.

BACKGROUND:The Xoft electronic brachytherapy source is commonly used to treat superficial lesions and tumors located at shallow depths. However, uncertainties in material composition and geometry, mainly arising due to the manual assembly of the x-ray tube components, contribute to inter-source variability in the tube output spectrum. In addition, aging of the x-ray tube may lead to intra-source variability in the tube spectrum. PURPOSE:To investigate the inter- and intra-source variability of the spectrum for the Xoft S7500 model experimentally and through simulations, as well as to study the impact of this variability on dosimetry. METHODS:The Amptek X123 CdTe x-ray spectrometer was used to measure the spectrum of the Xoft S7500 source model. First, the spectrometer was calibrated using 137Cs, 152Eu, and 57Co. Second, the source output spectrum was measured on the side and at the tip of the source at a distance of 15.5 cm from the source tip. Throughout the measurement, the source was fixed to the optical table using a 3D-printed holder and was aligned using a laser beam. The intra-source variability was studied by investigating the spectrum emitted by one single source measured in five trials. Similarly, the inter-source variability was investigated by measuring five different sources of the S7500 model at the tip and the side. The symmetry of the output was investigated by comparing the side and tip spectra. The measurements were then compared with the simulated spectra using a previously developed E-Brachy software package, taking into account the range of variation in material composition reported by the manufacturer. Finally, to determine whether or not the uncertainty in the material composition and the output spectrum affect the dosimetry by a clinically significant amount, we compared the depth dose curves created by the upper and lower limits of the range of uncertainty in the material composition. RESULTS:The calibration line was obtained to correspond the detector channels to the energy in keV using linear regression. Escape peak and background corrections were performed on the spectrum, and the counts in the K-edges of cadmium (26.7 keV) and tellurium (31.8 keV) were readjusted so that the silver peaks were more clearly resolved. Intra-source variability demonstrated consistent peak resolution for tungsten, yttrium, and silver, with a coefficient of variation (CV) ranging from 0.5% to 9.8%. Inter-source variability highlighted significant differences between tip and side measurements, with up to 13.5% variation. Simulated spectra revealed the impact of material composition on the characteristic peak intensities, with the intensity of the peaks in the measured spectra lying between simulated spectra in the case of higher and lower percentage limits of yttrium and silver present in the material composition. Depth-dose simulations showed minimal differences (<1.2%) between material compositions. CONCLUSIONS:This study comprehensively compared the measured and simulated spectra of the Xoft S7500 source model. The differences in the depth-dose curves are within 2% as recommended by the AAPM TG-568 requirements, and the variations in the spectra for the source model S7500 lie within the recommended range.

Privacy regulations and limited expert-validation constrain the deployment of large language models (LLMs) for electronic health record structuring. We evaluated locally deployed LLMs to extract 30 prognostic variables from 1,360 head and neck cancer reports (882 patients) using zero-shot prompting. A stratified 50-case subset was reviewed by three radiation oncologists (50 cases, 30 fields, 3 reviewers; 4,500 decisions) to form a majority-vote reference for Llama3.3-70B, which achieved 98.6% F1 with high clinician agreement and processed reports in 53 s/report. Among seven additional models (2.6B-70B) benchmarked against this reference, GPT-OSS-20.9B (F1 89.4%) and MedGemma-27B (F1 88.5%) performed best. Integrating LLM-extracted HPV status, smoking history, and Charlson Comorbidity Score into a multivariate Cox Proportional Hazards model (age, sex, T/N stage) improved disease-free survival (likelihood ratio test p = 0.014; ΔC-index + 0.071) and locoregional failure-free survival (p = 0.026; ΔC-index + 0.108) with 1,000-bootstrap internal validation. This clinician-AI collaborative evaluation shows that on-premises LLMs enable privacy-preserving and efficient tumour board support, longitudinal data curation, and outcome prediction.

PURPOSE:Drug-eluting stents fail in up to 20% of patients. In failed cases, intravascular brachytherapy (IVBT) is administered with β-emitting 90Sr90Y through a guidewire. Current clinical dosimetry is water-based, neglecting attenuation from patient-specific materials such as plaques, stents, and the off-centered guidewire, leading to a discrepancy between prescribed and delivered dose. This study retrospectively performed patient-specific IVBT dose calculations using Optical Coherence Tomography (OCT) to quantify uncertainties in clinical dosimetry. METHODS AND MATERIALS:Dose calculations on OCT images from ten patients were performed using RapidBrachyIVBT, a Monte Carlo-based dose calculation software. Heterogeneities, including guidewire(s), stents, and fibrotic and calcified plaques, were contoured and assigned material properties; surrounding tissue was modeled as smooth muscle. Absorbed dose to water and medium were calculated. The prescribed dose to water was 18.4 or 23 Gy at 2 mm from the source, depending on lumen diameter. The dose homogeneity index was defined as the ratio of the maximum to the minimum dose in the target volume. RESULTS:When heterogeneities were included, median maximum dose attenuation was 76.7% (75.0-77.1) in the artery segment and 56.2% (52.2-65.1) in the target volume. The median dose homogeneity index increased from 1.29 in water to 2.93 (2.44-3.33) with patient-specific materials. The guidewire produced asymmetric dose distributions in all patients, with the greatest attenuation where it opposed thick calcified plaques. CONCLUSIONS:Standard water-based IVBT dosimetry is inaccurate due to dose-attenuating materials present during treatment. Personalized, image-guided IVBT planning that accounts for patient-specific heterogeneities may improve treatment accuracy and clinical outcomes.

Background: Dynamic positron emission tomography is an under utilized clinical technique. It requires knowledge of the time-course activity concentration in the patient’s arterial blood, called arterial input function (AIF), normally acquired via arterial blood sampling. Alternative methods to acquire the AIF do exist, but each has its own limitations. This study presents a non-invasive radiation detector, called Radian-PET, designed to measure the AIF non-invasively. Methods: 10 cm long scintillating fibers arranged in two bands of 32 fibers were read out by a 64-channel silicon photomultiplier array. Calibration measurements were performed using 68Ge rod sources placed over the sensitive volume of the detector. Inter-channel variability was measured using the scans with the source perpendicular to the fibers. Two decay measurements were performed using 18F samples with varying activity. Cross-validation measurements were performed by placing a microfluidic blood sampling detector before Radian-PET in a microfluidic circuit. 18F, 11C and 68Ga injections were used to simulate the AIF. Results: Calibration measurements show that the electronic discriminator threshold, with minimal detectable activity below 100 kBq per mL for a 1-second integration window. The measured 18F half-lives were 109.75 ± 0.04 min and 110.82 ± 0.73 min and agreed within uncertainties with the literature. Linear regressions from the cross-validation measurements showed good agreement (R2 > 0.96) for all scans. Conclusions: Radian-PET can accurately measure clinically relevant activity concentrations and detect quickly changing activity levels like those used to perform kinetic analysis. Further work is required to increase detector efficiency and reduce sources of noise.

We apply the precision tools of cavity-enhanced absorption sensing to clinical oncology, demonstrating a dosimeter paradigm in which a centimeter-scale volume of water serves as a tissue-equivalent sensing medium. Our proof-of-concept, all-optical scheme achieves real-time readout of clinical radiation pulses with a nominal single-pulse resolution of 90 μGy. This demonstration paves the way toward fiber-integrated, micron-scale devices for in situ universal absolute dosimetry during treatment.

BrachyUtils is a modular Python framework designed to streamline reproducible research in high dose rate (HDR) brachytherapy. It addresses limitations in complex clinical workflows that constrain plan quality under time pressure, where automation tools exist but lack standardized benchmarking and integration. The framework integrates open-source and in-house tools via Docker for reproducibility, connecting key components: registration (OpenTPS, Plastimatch, SimpleElastix), dose calculation/analysis (RapidBrachyTG43, RapidBrachyMC), and optimization (Gurobi, AMPL, ORTools). It also manages data anonymization and file conversions. Benchmarking on the μ-RegPro prostate dataset and 13 in-house HDR prostate plans demonstrated its effectiveness. Contour-based registration outperformed image-based methods; SimpleElastix-BSpline achieved the top prostate Dice score (0.94 ± 0.02) and the lowest Hausdorff distance (4.87 mm ± 1.58 mm) for biopsy contours, while OpenTPS-rigid was fastest (0.89 s ± 0.01 s). Dosimetry comparisons showed that RapidBrachyTG43’s dose-to-water overestimated dose volume histogram (DVH) metrics by 3.3%–6.8% versus RapidBrachyMC’s dose-to-medium. Among solvers, Gurobi excelled with the shortest times (0.59 s ± 0.28 s) for linear/ quadratic penalties. Overall, BrachyUtils unifies data handling, registration, dosimetry, and optimization in a reproducible environment, enabling automated evaluations and accelerating treatment planning research. The rapidly evolving framework can be accessed at https://github.com/engerlab/brachyutils.git.

Alpha-DaRT, an interstitial treatment, uses 224Ra-based diffusing alpha-emitters to treat solid tumors by creating a high-dose region up to 5 mm around the source. However, diffusion lengths (Ldiff) remain uncertain across cancer types, with current estimates from subcutaneous and limited orthotopic models. This study measured the Ldiff in an orthotopic colorectal adenocarcinoma mouse model. HT-29 colorectal adenocarcinoma cells were injected into the rectal submucosa of 38 Nod scid gamma mice. Tumor growth was monitored with 7 T Magnetic Resonance Imaging (MRI). At ~ 5-7 mm, groups included active (n = 20), inert (n = 9), and control (n = 9). Active mice had Alpha-DaRT sources implanted in tumors (n = 15) and rectal muscle (n = 5). Ex-vivo liver tissue (n = 3) was also analyzed. After four days, gamma spectroscopy measured 212Pb activity, and autoradiographs and histopathology assessed Ldiff, tissue damage, and vascularity (H&E, CD-31, CC-3). Ldiff was 0.36-0.84 mm in tumors, 0.30 mm in rectal muscle, and 1.0 mm in liver tissue. 212Pb showed a 50-93% escape probability, with kidneys having the highest activity. Active tumors exhibited more necrosis (p = 0.034) and reduced vascularity. This study provides the first in-vivo Ldiff measurements of Alpha-DaRT in an orthotopic colorectal adenocarcinoma model, highlighting Ldiff variability and the need for optimization based on cancer type.

PURPOSE:Endoscopy is critical in the identification of rectal tumors, but is prone to observer errors. The aim of this study was to assess the inter- and intra-observer variability in delineating rectal lesions in endoscopic images taken during high-dose-rate (HDR) brachytherapy and develop a deep learning-based automatic tumor segmentation model. MATERIALS AND METHODS:Three expert annotators identified tumors, scaring, ulcers and radiation proctitis in 801 endoscopic images from 24 patients. Inter-observer variability was evaluated at both whole-image and contour levels. Intra-observer variability was assessed by re-annotating 15 images from 14 patients after six months. Four DeepLabV3 models with a ResNet50 backbone were trained using a nested cross-validation approach: one per annotator and a fourth trained on majority-vote contours. Model performance was evaluated on 60 unseen images, which the annotators rated using a five-point Likert scale. RESULTS:Manual annotations showed significant variability for ulcers and radiation proctitis (average Dice: 0.36 and 0.57) versus tumors (0.83). Intra-observer Dice scores were 0.72, 0.68, and 0.87 across annotators. The majority-vote model outperformed individual annotator models (average Dice: 0.77) but generated many false positives, misclassifying ulcers and proctitis as tumors. Annotators generally rated the model trained on their own contours higher on the unseen test set. CONCLUSIONS:This work highlights the variability in expert annotations used as ground-truth for deep learning-based segmentation of rectal tumors in endoscopic images acquired during HDR brachytherapy. Automated contouring may provide a foundation for adaptive, AI-assisted brachytherapy workflows.

Background and purpose: Clinical brachytherapy treatment planning is performed assuming the patient is composed entirely of water and infinite in size. In this work, the effects of this assumption on calculated dose were investigated by comparing dose to water in water (Dw,w) in an unbound phantom mimicking TG-43 conditions, and dose to medium in medium (Dm,m) for breast cancer patients treated with high dose rate brachytherapy. Materials and methods: Treatment plans for 123 breast cancer patients were recalculated with a Monte Carlo-based treatment planning software. The dwell times and dwell positions were imported from the clinical treatment planning system. The dose was computed and reported as Dw,w and Dm,m. Dose-volume histogram (DVH) metrics were evaluated for target volumes and organs at risk. Results: Dw,w overestimated the dose for most studied DVH metrics. The largest median overestimations between Dm,m and Dw,w were seen for the planning target volume (PTV) V200% (5.8%), lung D0.1 cm3 (6.0%) and skin D0.1 cm3 (4.2%). The differences between Dm,m and Dw,w were statistically significant for all investigated DVH metrics. The PTV V90% had the smallest deviation (0.7%). Conclusion: There was a significant difference in the DVH metrics studied when tissue heterogeneities and patient-specific scattering are accounted for in high dose rate breast brachytherapy. However, for the studied patient cohort, the clinical coverage goal (PTV V90%), had the smallest deviation.