PURPOSE:To develop a machine learning-based workflow for patient-specific breast radiation dosimetry in CT. MATERIALS AND METHODS:Two hundred eighty-six chest CT examinations, with corresponding right and left breast contours, were retrospectively collected from the radiotherapy department at our institution to develop and validate breast segmentation U-Nets. Additionally, Monte Carlo simulations were performed for each CT scan to determine radiation doses to the breasts. The derived breast doses, along with predictors such as X-ray tube current and radiomic features, were then used to train deep neural networks (DNNs) for breast dose prediction. RESULTS:The breast segmentation models achieved a mean dice similarity coefficient of 0.92, with precision and sensitivity scores above 0.90 for both breasts, indicating high segmentation accuracy. The DNNs demonstrated close alignment with ground truth values, with mean predicted doses of 5.05 ± 0.50 mGy for the right breast and 5.06 ± 0.55 mGy for the left breast, compared to ground truth values of 5.03 ± 0.57 mGy and 5.02 ± 0.61 mGy, respectively. The mean absolute percentage errors were 4.01 % (range: 3.90 %-4.12 %) for the right breast and 4.82 % (range: 4.56 %-5.11 %) for the left breast. The mean inference time was 30.2 ± 4.3 s. Statistical analysis showed no significant differences between predicted and actual doses (p ≥ 0.07). CONCLUSION:This study presents an automated, machine learning-based workflow for breast radiation dosimetry in CT, integrating segmentation and dose prediction models. The models and code are available at: https://github.com/eltzanis/ML-based-Breast-Radiation-Dosimetry-in-CT.

This study calculated the radiation dose to young patients with high-risk abdominal neuroblastoma from therapeutic and imaging procedures. Computational XCAT phantoms representing typical patients aged 5–15 years were used. Intensity modulated radiotherapy (IMRT) and volumetric modulated arc therapy (VMAT) plans were generated with 6 MV photons for a planning target volume (PTV) on the left and right abdominal side. Dose-volume-histograms from the plans were used to find the average dose (Dav) to critical normal abdominal and thoracic organs. The imaging dose to these organs and PTV was calculated by simulating kV cone-beam computed tomography (CBCT) scanning for patient setup before radiotherapy. Different CBCT protocols were simulated with Monte Carlo methods. The IMRT and VMAT plans provided similar PTV coverage and organ sparing. For a 21.6 Gy target dose, the Dav of the abdominal organs from the treatment plans was 3.6–19.6 Gy and that of thoracic organs was 0.1–2.3 Gy. Daily CBCT scans on 15–year-old patients with the standard adult protocol gave total PTV and organ doses of 95.3–485.3 mGy. The doses from the modified standard protocol for 5- and 10-year-old patients were 74.2–159.6 mGy. The dose calculations with a specially designed CBCT protocol for patients up to 10 years were 6.0–27.8 mGy. The total imaging dose to the PTV was up to 2.2

BACKGROUND:Advanced radiotherapy techniques such as single-isocenter volumetric modulated arc therapy (VMAT) and simultaneous integrated boost (SIB-VMAT) require precise quality assurance (QA) due to their complexity and sensitivity to geometric and dosimetric uncertainties, especially for multi-target configurations. PURPOSE:To evaluate and compare two dosimetric systems-a diode array and a polymer gel dosimeter-for quality assurance in single-isocenter multi-target VMAT and SIB-VMAT plans through 3D gamma index and statistical agreement analysis. METHODS:A diode array system (Delta⁴-Scandidos, Uppsala, Sweden) and a 3D printed phantom (Prime-RTsafe, Athens, Greece) embedded with in-house polymer gel were utilized. The treatment plans for VMAT and SIB-VMAT were created using the Monaco treatment planning system (TPS) and irradiations were performed with the Elekta Infinity linear accelerator with a 6-MV photon beam on both Prime and Delta⁴. Analyses of the irradiated gels were performed using a 1.5T clinical MRI system. Additionally, 3D gamma indexes and Bland-Altman analyses were conducted to evaluate the agreement between relative doses from MRI-derived gel data and diode array's detector measurements. RESULTS:Diode array system achieved gamma passing rates (GPRs) >99%, while polymer gel showed >95% GPR for both irradiation plans. Bland-Altman analysis indicated minimal bias (mean difference: 0.1%) and narrow limits of agreement (-1.9% to 2.2%), confirming good consistency between the two dosimetric methods for both plans delivered. CONCLUSIONS:Acceptable agreement between the two systems was observed. Both demonstrated complementary capabilities, making both essential tools for ensuring precision in advanced radiotherapy techniques. The polymer gel system offers more detailed insights compared to the diode array method, showing increased sensitivity in challenging cases that involve small targets at greater distances from the isocenter.

Artificial intelligence (AI) is transforming medical radiation applications by handling complex data, learning patterns, and making accurate predictions, leading to improved patient outcomes. This article examines the use of AI in optimising radiation doses for x-ray imaging, improving radiotherapy outcomes, and briefly addresses the benefits, challenges, and limitations of AI integration into clinical workflows. In diagnostic radiology, AI plays a pivotal role in optimising radiation exposure, reducing noise, enhancing image contrast, and lowering radiation doses, especially in high-dose procedures like computed tomography (CT). Deep learning (DL)-powered CT reconstruction methods have already been incorporated into clinical routine. Moreover, AI-powered methodologies have been developed to provide real-time, patient-specific radiation dose estimates. These AI-driven tools have the potential to streamline workflows and potentially become integral parts of imaging practices. In radiotherapy, AI's ability to automate and enhance the precision of treatment planning is emphasised. Traditional methods, such as manual contouring, are time-consuming and prone to variability. AI-driven techniques, particularly DL models, are automating the segmentation of organs and tumours, improving the accuracy of radiation delivery, and minimising damage to healthy tissues. Moreover, AI supports adaptive radiotherapy, allowing continuous optimisation of treatment plans based on changes in a patient's anatomy over time, ensuring the highest accuracy in radiation delivery and better therapeutic outcomes. Some of these methods have been validated and integrated into radiation treatment systems, while others are not yet ready for routine clinical use mainly due to challenges in validation, particularly ensuring reliability across diverse patient populations and clinical settings. Despite the potential of AI, there are challenges in fully integrating these technologies into clinical practice. Issues such as data protection, privacy, data quality, model validation, and the need for large and diverse datasets are crucial to ensuring the reliability of AI systems.

Purpose: To develop a machine learning-based methodology for patient-specific radiation dosimetry in thoracic and abdomen CT. Methods: Three hundred and thirty-one thoracoabdominal radiotherapy-planning CT examinations with the respective organ/patient contours were collected retrospectively for the development and validation of segmentation 3D-UNets. Moreover, 97 diagnostic thoracic and 89 diagnostic abdomen CT examinations were collected retrospectively. For each of the diagnostic CT examinations, personalized MC dosimetry was performed. The data derived from MC simulations along with the respective CT data were used for the training and validation of a dose prediction deep neural network (DNN). An algorithm was developed to utilize the trained models and perform patient-specific organ dose estimates for thoracic and abdomen CT examinations. The doses estimated with the DNN were compared with the respective doses derived from MC simulations. A paired t-test was conducted between the DNN and MC results. Furthermore, the time efficiency of the proposed methodology was assessed. Results: The mean percentage differences (range) between DNN and MC dose estimates for the lungs, liver, spleen, stomach, and kidneys were 7.2 % (0.2-24.1 %), 5.5 % (0.4-23.0 %), 7.9 % (0.6-22.3 %), 6.9 % (0.0-23.0 %) and 6.7 % (0.3-22.6 %) respectively. The differences between DNN and MC dose estimates were not significant (p-value = 0.12). Moreover, the mean processing time of the proposed workflow was 99 % lower than the respective time needed for MC-based dosimetry. Conclusions: The proposed methodology can be used for rapid and accurate patient-specific dosimetry in chest and abdomen CT.

PurposeTo propose an artificial intelligence (AI)-based method for personalized and real-time dosimetry for chest CBCT acquisitions.MethodsCT images from 113 patients who underwent radiotherapy treatment were collected for simulating thorax examinations using cone-beam computed tomography (CBCT) with the Monte Carlo technique. These simulations yielded organ dose data, used to train and validate specific AI algorithms. The efficacy of these AI algorithms was evaluated by comparing dose predictions with the actual doses derived from Monte Carlo simulations, which are the ground truth, utilizing Bland-Altman plots for this comparative analysis.ResultsThe absolute mean discrepancies between the predicted doses and the ground truth are (0.9 ± 1.3)% for bones, (1.2 ± 1.2)% for the esophagus, (0.5 ± 1.3)% for the breast, (2.5 ± 1.4)% for the heart, (2.4 ± 2.1)% for lungs, (0.8 ± 0.6)% for the skin, and (1.7 ± 0.7)% for integral. Meanwhile, the maximum discrepancies between the predicted doses and the ground truth are (14.4 ± 1.3)% for bones, (12.9 ± 1.2)% for the esophagus, (9.4 ± 1.3)% for the breast, (14.6 ± 1.4)% for the heart, (21.2 ± 2.1)% for lungs, (10.0 ± 0.6)% for the skin, and (10.5 ± 0.7)% for integral.ConclusionsAI models that can make real-time predictions of the organ doses for patients undergoing CBCT thorax examinations as part of radiotherapy pre-treatment positioning were developed. The results of this study clearly show that the doses predicted by analyzed AI models are in close agreement with those calculated using Monte Carlo simulations.

PurposeTo propose a novel deep-learning based dosimetry method that allows quick and accurate estimation of organ doses for individual patients, using only their computed tomography (CT) images as input.MethodsDespite recent advances in medical dosimetry, personalized CT dosimetry remains a labour-intensive process. Current state-of-the-art methods utilize time-consuming Monte Carlo (MC) based simulations for individual organ dose estimation in CT. The proposed method uses conditional generative adversarial networks (cGANs) to substitute MC simulations with fast dose image generation, based on image-to-image translation. The pix2pix architecture in conjunction with a regression model was utilized for the generation of the synthetic dose images. The lungs, heart, breast, bone and skin were manually segmented to estimate and compare organ doses calculated using both the original and synthetic dose images, respectively.ResultsThe average organ dose estimation error for the proposed method was 8.3% and did not exceed 20% for any of the organs considered. The performance of the method in the clinical environment was also assessed. Using segmentation tools developed in-house, an automatic organ dose calculation pipeline was set up. Calculation of organ doses for heart and lung for each CT slice took about 2 s.ConclusionsThis work shows that deep learning-enabled personalized CT dosimetry is feasible in real-time, using only patient CT images as input.

PURPOSE:The aim of this study was to compare the organ doses assessed through a digital phantom-based and a patient specific-based dosimetric tool in adult routine thorax computed tomography (CT) examinations with reference to physical dose measurements performed in anthropomorphic phantoms. METHODS:Two Monte Carlo based dose calculation tools were used to assess organ doses in routine adult thorax CT examinations. These were a digital phantom-based dosimetry tool (NCICT, National Cancer Institute, USA) and a patient-specific individualized dosimetry tool (ImpactMC, CT Imaging GmbH, Germany). Digital phantoms and patients were classified in four groups according to their water equivalent diameter (Dw). Normalized to volume computed tomography dose index (CTDIvol), organ dose was assessed for lungs, esophagus, heart, breast, active bone marrow, and skin. Organ doses were compared to measurements performed using thermoluminescent detectors (TLDs) in two physical anthropomorphic phantoms that simulate the average adult individual as a male (Alderson Research Labs, USA) and as a female (ATOM Phantoms, USA). RESULTS:The average percent difference of NCICT to TLD and ImpactMC to TLD dose measurements across all organs in both sexes was 13% and 6%, respectively. The average ± 1 standard deviation in dose values across all organs with NCICT, ImpactMC, and TLDs was ± 0.06 (mGy/mGy), ± 0.19 (mGy/mGy), and ± 0.13 (mGy/mGy), respectively. Organ doses decreased with increasing Dw in both NCICT and ImpactMC. CONCLUSION:Organ doses estimated with ImpactMC were in closer agreement to TLDs compared to NCICT. This may be attributed to the inherent property of ImpactMC methodology to generate phantoms that resemble the realistic anatomy of the examined patient as opposed to NCICT methodology that incorporates an anatomical discrepancy between phantoms and patients.

BackgroundPatient-specific organ-dose estimation in diagnostic CT examinations can provide useful insights on individualized secondary cancer risks, protocol optimization, and patient management. Current dose estimation techniques mainly rely on time-consuming Monte Carlo methods or/and generalized anthropomorphic phantoms. PurposeWe proposed a proof-of-concept rapid workflow based on deep learning networks to estimate organ doses for individuals following thorax Computed Tomography (CT) examinations. MethodsCT scan data from 95 individuals undergoing thorax CT examinations were used. Monte Carlo simulations were performed and three-dimensional (3D) dose distributions for each patient were obtained. A fully connected sequential deep learning network model was constructed and trained for each organ considered in this study. Water-equivalent diameter (WED), scan length, and tube current were the independent variables. Organ doses for heart, lungs, esophagus, and bones were calculated from the Monte Carlo 3D distribution and used to train the deep learning networks. Organ dose predictions from each network were evaluated using an independent data set of 19 patients. ResultsThe trained networks provided organ dose predictions within a second. There was very good agreement between the deep learning network predictions and reference organ dose values calculated from Monte Carlo simulations. The average difference was -1.5% for heart, -1.6% for esophagus, -1.0% for lungs, and -0.4% for bones in the 95 patients dataset, and -5.1%, 4.3%, 0.9%, and 1.4% respectively in the 19 patients test dataset. ConclusionsThe proposed workflow demonstrated that patient-specific organ-doses can be estimated in nearly real-time using deep learning networks. The workflow can be readily implemented and requires a small set of representative data for training.

Medical ionizing radiation procedures and especially medical imaging are a non negligible source of exposure to patients. Whereas the biological effects of high absorbed doses are relatively well known, the effects of low absorbed doses are still debated. This work presents the development of a computer platform called Image and Radiation Dose BioBank (IRDBB) to manage research data produced in the context of the MEDIRAD project, a European project focusing on research on low doses in the context of medical procedures. More precisely, the paper describes a semantic database linking dosimetric data (such as absorbed doses to organs) to the images corresponding to X-rays exposure (such as CT images) or scintigraphic images (such as SPECT or PET images) that allow measuring the distribution of a radiopharmaceutical. The main contributions of this work are: 1) the implementation of the semantic database of the IRDBB system and 2) an ontology called OntoMEDIRAD covering the domain of discourse involved in MEDIRAD research data, especially many concepts from the DICOM standard modelled according to a realist approach.