Physics and Imaging in Radiation Oncology最新文献

{"title":"Accuracy-dependent dose-constraints and dose-based safety margins for organs-at-risk in radiotherapy","authors":"Joep C. Stroom ,&nbsp;Sandra C. Vieira ,&nbsp;Carlo Greco ,&nbsp;Sebastiaan M.J.J.G. Nijsten","doi":"10.1016/j.phro.2025.100713","DOIUrl":"10.1016/j.phro.2025.100713","url":null,"abstract":"<div><h3>Background and purpose</h3><div>Geometrical uncertainties in radiotherapy are generally accounted for by margins for tumors, but their effect on organs-at-risk (OARs) is often ignored. We developed a model that incorporates dose- and geometry-based uncertainties in OAR planning using dose constraints.</div></div><div><h3>Materials and methods</h3><div>Radiotherapy uncertainties cause real dose-volume histograms (DVHs) to spread around the planned DVH. With a <em>published</em> OAR dose constraint D(V_crit) &lt; D_crit such that complication probability &lt; Y%, real differences from planned D_crit can be described by mean- (MD_Dcrit) and standard deviations (SD_Dcrit). Assuming complications are associated with the worst DVHs, <em>New</em> dose constraints that maintain complication probability can be derived for new treatments:<span><span><span>D_crit,New = D_crit,publ + Φ⁻¹(1 - Y%) * (SD_Dcrit,publ - SD_Dcrit,New) + (MD_Dcrit,publ - MD_Dcrit,New),</span></span></span>with Φ⁻¹(x) the inverse cumulative normal distribution function. Setting SD_Dcrit,New = MD_Dcrit,New = 0 in the recipe yields the “True” critical dose, and D_crit,True - D_crit,publ can be considered a dose-based safety margin (DSM).</div><div>As hypothetical example, we estimated MD_Dcrit and SD_Dcrit values by simulating geometric errors in our clinical treatment plans and adding dose-based uncertainty. Over 1000 OARs with 108 different regular- and hypo-fractionation constraints were simulated. We assumed accuracy SDs to change from 2.5mm/3% to 1.5mm/2%.</div></div><div><h3>Results</h3><div>Results varied per OAR, fractionation, and constraint-type. If our 2.5mm/3% MD_Dcrit and SD_Dcrit values approximated dose-constraint studies, on average the DSM would be 4.5 Gy (18%) and our dose constraints would increase with 1.2 Gy (5%).</div></div><div><h3>Conclusions</h3><div>We introduced a first model relating dose constraints and complication probabilities with treatment uncertainties and safety margins for OARs. Among other things, it quantified how higher constraints can be applied with increasing radiotherapy accuracy.</div></div>","PeriodicalId":36850,"journal":{"name":"Physics and Imaging in Radiation Oncology","volume":"33 ","pages":"Article 100713"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143130342","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel multimodality anthropomorphic phantom enhances compliance with quality assurance guidelines for magnetic resonance imaging in radiotherapy","authors":"Meshal Alzahrani ,&nbsp;David A Broadbent ,&nbsp;Irvin Teh ,&nbsp;Bashar Al-Qaisieh ,&nbsp;Emily Johnstone ,&nbsp;Richard Speight","doi":"10.1016/j.phro.2025.100707","DOIUrl":"10.1016/j.phro.2025.100707","url":null,"abstract":"<div><h3>Background and purpose</h3><div>The use of magnetic resonance imaging (MRI) for radiotherapy (RT) simulation has grown, prompting quality assurance (QA) guidelines by the Institute of Physics and Engineering in Medicine (IPEM) and the American Association of Physicists in Medicine (AAPM). This study compares a novel multimodality anthropomorphic phantom to an American College of Radiology (ACR) phantom for a subset of these MRI-specific QA tests in RT.</div></div><div><h3>Materials and methods</h3><div>A novel 3D-printed multimodality head-and-neck anthropomorphic phantom was compared to an ACR large MRI phantom. IPEM and AAPM-recommended QA tests were conducted, including informatics/connectivity/data transfer, MRI-CT registration, end-to-end QA, and signal-to-noise ratio (SNR)/percentage integral uniformity (PIU) assessments using RT accessories.</div></div><div><h3>Results</h3><div>Both phantoms were suitable for informatics/connectivity/data transfer. In MRI-CT registration, no errors were found; the ACR phantom offered more quantitative landmarks, while the anthropomorphic phantom provided limited structures. Both phantoms achieved target registration errors (TREs) below 0.97 mm and dice similarity coefficient (DSC) values above 0.9, meeting guidelines. For end-to-end QA, the anthropomorphic phantom facilitated dose measurements of 1.994 Gy versus a calculated 2.01 Gy (−0.8 %). SNR and PIU assessments showed higher values in radiology setups compared to RT setups for both phantoms.</div></div><div><h3>Conclusions</h3><div>Multimodality anthropomorphic phantoms compatible with dosimetric equipment allow realistic end-to-end QA, unlike the ACR phantom. While the ACR phantom is suitable for informatics and MRI-CT registration, anthropomorphic phantoms better represent clinical scenarios. For comprehensive QA, both ACR and anthropomorphic phantoms are required. Additionally, large field-of-view (FOV) phantoms are crucial for evaluating large FOV MRI distortions.</div></div>","PeriodicalId":36850,"journal":{"name":"Physics and Imaging in Radiation Oncology","volume":"33 ","pages":"Article 100707"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143130585","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantitative magnetic resonance imaging responses in head and neck cancer patients treated with magnetic resonance-guided hypofractionated radiation therapy","authors":"Ryan Bonate ,&nbsp;Musaddiq J. Awan ,&nbsp;Heather A. Himburg ,&nbsp;Stuart Wong ,&nbsp;Monica Shukla ,&nbsp;Sergey Tarima ,&nbsp;Joseph Zenga ,&nbsp;Eric S. Paulson","doi":"10.1016/j.phro.2024.100693","DOIUrl":"10.1016/j.phro.2024.100693","url":null,"abstract":"<div><h3>Background and purpose</h3><div>Quantitative MRI (qMRI) has been explored for detecting tumor changes during radiation therapy (RT) in head and neck squamous cell cancer (HNSCC). Clinical trials show prolonged survival with PD-1 targeted immune checkpoint inhibition. Hypofractionated radiation regimens are being studied to counteract radioresistant clonogen formation. This study aims to use daily qMRI monitoring in these therapies. The objective of this exploratory study was to investigate if qMRI can detect tumor microenvironment changes during hypofractionated RT in a phase I trial of Dose-Escalated Hypofractionated Adaptive Radiotherapy (DEHART).</div></div><div><h3>Materials and methods</h3><div>Seventeen subjects with advanced HNSCC underwent MR-guided RT with daily qMRI using a 15-fraction regimen to a cumulative dose of 50, 55, or 60 Gy. A 1.5 T MRI-Linac collected daily intravoxel incoherent motion (IVIM), T₁, and T₂ mappings. Median primary tumor ADC, D, D*, f, T₁, and T₂ were calculated, using paraspinal muscle as a control. qMRI parameters were analyzed by treatment condition and length using linear mixed effect models and nonparametric tests.</div></div><div><h3>Results</h3><div>Significant (p &lt; 0.05) increases in ADC, D, f, and T2 were observed over treatment duration for multiple conditions. Daily monitoring enhanced result significance compared to weekly collection.</div></div><div><h3>Conclusions</h3><div>Daily qMRI effectively monitors tumor response over short periods and varying treatment conditions. Further studies on radiation and systemic therapy combinations in HNSCC could benefit from daily qMRI data collection.</div></div>","PeriodicalId":36850,"journal":{"name":"Physics and Imaging in Radiation Oncology","volume":"33 ","pages":"Article 100693"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11772986/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143059435","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Recommendations for reporting and evaluating proton therapy beyond dose and constant relative biological effectiveness","authors":"Armin Lühr ,&nbsp;Dirk Wagenaar ,&nbsp;Daniëlle B.P. Eekers ,&nbsp;Lars Glimelius ,&nbsp;Steven J.M. Habraken ,&nbsp;Semi Harrabi ,&nbsp;Miranda C.A. Kramer ,&nbsp;Ranald I. Mackay ,&nbsp;Ana Vaniqui ,&nbsp;Alexandru Dasu ,&nbsp;Damien C. Weber","doi":"10.1016/j.phro.2024.100692","DOIUrl":"10.1016/j.phro.2024.100692","url":null,"abstract":"<div><h3>Background and purpose</h3><div>In proton therapy, a relative biological effectiveness (RBE) of 1.1 is used to convert proton dose into an equivalent photon dose. However, RBE varies with tissue type, fraction dose, and beam quality parameters beyond dose such as linear energy transfer (LET) raising concerns about increased local effectiveness and potential toxicity. This work aims to harmonize quantities used for clinical consideration of variable RBE for proton therapy.</div></div><div><h3>Materials and methods</h3><div>A survey was distributed to proton centres to determine agreement on RBE-related concerns and clinical implementations. A subsequent clinical expert meeting facilitated by the European Particle Therapy Network was held to achieve consensus and to make clinical recommendations how to prescribe and report beyond using dose and constant RBE.</div></div><div><h3>Results</h3><div>The survey was answered by 17 out of 23 centres contacted (74%). For proton RBE, most concerns existed regarding toxicity in serial organs, while the assumption of an RBE of 1.1 was considered valid for targets. Most physicists intended to consider a physical quantity beyond dose in clinical decision making.</div></div><div><h3>Conclusions</h3><div>A constant RBE of 1.1 was the consensus for prescribing dose. However, current practice of recording and reporting dose in proton therapy must be complemented: the recommended quantity beyond dose was the dose-averaged LET in water from primary and secondary protons, normalized to unit density. This will facilitate analyses of treatment data on effectiveness beyond dose and between centres. No consensus on a single variable RBE model was found. More clinical training on proton RBE is needed.</div></div>","PeriodicalId":36850,"journal":{"name":"Physics and Imaging in Radiation Oncology","volume":"33 ","pages":"Article 100692"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11750264/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143013292","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"First dosimetric evaluation of clinical raster-scanned proton, helium and carbon ion treatment plan delivery during simultaneous real-time magnetic resonance imaging","authors":"Sebastian Klüter ,&nbsp;Karolin Milewski ,&nbsp;Wibke Johnen ,&nbsp;Stephan Brons ,&nbsp;Jakob Naumann ,&nbsp;Stefan Dorsch ,&nbsp;Cedric Beyer ,&nbsp;Katharina Paul ,&nbsp;Kilian A. Dietrich ,&nbsp;Tanja Platt ,&nbsp;Jürgen Debus ,&nbsp;Julia Bauer","doi":"10.1016/j.phro.2025.100722","DOIUrl":"10.1016/j.phro.2025.100722","url":null,"abstract":"<div><div>This work presents an experimental dosimetric evaluation of raster-scanning particle beam delivery during simultaneous in-beam magnetic resonance (MR) imaging. Using an open MR scanner at an experimental treatment room, radiochromic film comparisons for protons, helium and carbon ions, each with and without simultaneous in-beam cine MR imaging, yielded 2D gamma pass rates ≥ 98.8 % for a 3 % / 1.5 mm criterion, and ≥ 99.9 % for 5 % / 1.5 mm. These results constitute a first experimental confirmation that time varying magnetic fields of MR gradients do not result in clinically relevant additional dose perturbations.</div></div>","PeriodicalId":36850,"journal":{"name":"Physics and Imaging in Radiation Oncology","volume":"33 ","pages":"Article 100722"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143387666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Out-of-field dose assessment for pencil beam scanning proton radiotherapy versus photon radiotherapy for breast cancer in pregnant women","authors":"Menke Weessies,&nbsp;Murillo Bellezzo,&nbsp;Britt J.P. Hupkens,&nbsp;Frank Verhaegen,&nbsp;Gloria Vilches-Freixas","doi":"10.1016/j.phro.2025.100721","DOIUrl":"10.1016/j.phro.2025.100721","url":null,"abstract":"<div><h3>Background and purpose</h3><div>Cancer affects 1 in 1000–2000 pregnancies annually worldwide, creating challenges in balancing cancer treatment and fetal safety. This study compares out-of-field radiation doses between two treatment modalities: 6MV external photon radiotherapy (XRT) and pencil beam scanning proton-therapy (PBS-PRT) for breast cancer, including imaging, to evaluate PBS-PRT as a potential new treatment option.</div></div><div><h3>Materials and methods</h3><div>For breast cancer involving lymph node levels 1–4 and the intramammary lymph nodes, treatment plans were created for XRT (with Flattening Filter (FF) and FF-Free (FFF)) and PBS-PRT, prescribing 15 × 2.67 Gy(RBE). Measurements were conducted using an adapted anthropomorphic phantom representing 20- and 30-week pregnancy. Bubble detectors placed in the phantom’s abdomen assessed neutron dose from PBS-PRT, while a Farmer ion chamber was used for imaging and XRT dose.</div></div><div><h3>Results</h3><div>At 20 weeks, PBS-PRT including imaging delivered 22.4 mSv, reducing dose 3.4-fold versus 6FF XRT and 2.5-fold versus 6FFF XRT. At 30 weeks, the PBS-PRT dose was 25.4 mSv, resulting in 7.6-fold and 6.3-fold reductions compared to 6FF and 6FFF XRT, respectively.</div></div><div><h3>Conclusions</h3><div>This study presents the first one-by-one comparison between PBS-PRT and different XRT modalities for pregnant breast cancer patients with an adapted anthropomorphic phantom. PBS-PRT measurements showed that the total equivalent dose was below the 100 mSv threshold outlined in AAPM Task Group Report No. 36 for a 30-week pregnancy, even under a worst-case scenario, maintaining treatment goals. These findings support the adoption of PBS-PRT as the preferred approach for treating pregnant breast cancer patients, should radiotherapy be required.</div></div>","PeriodicalId":36850,"journal":{"name":"Physics and Imaging in Radiation Oncology","volume":"33 ","pages":"Article 100721"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143349615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimal timing of organs-at-risk-sparing adaptive radiation therapy for head-and-neck cancer under re-planning resource constraints","authors":"Fatemeh Nosrat ,&nbsp;Cem Dede ,&nbsp;Lucas B. McCullum ,&nbsp;Raul Garcia ,&nbsp;Abdallah S.R. Mohamed ,&nbsp;Jacob G. Scott ,&nbsp;James E. Bates ,&nbsp;Brigid A. McDonald ,&nbsp;Kareem A. Wahid ,&nbsp;Mohamed A. Naser ,&nbsp;Renjie He ,&nbsp;Aysenur Karagoz ,&nbsp;Amy C. Moreno ,&nbsp;Lisanne V. van Dijk ,&nbsp;Kristy K. Brock ,&nbsp;Jolien Heukelom ,&nbsp;Seyedmohammadhossein Hosseinian ,&nbsp;Mehdi Hemmati ,&nbsp;Andrew J. Schaefer ,&nbsp;Clifton D. Fuller","doi":"10.1016/j.phro.2025.100715","DOIUrl":"10.1016/j.phro.2025.100715","url":null,"abstract":"<div><h3>Background and purpose</h3><div>Prior work on adaptive organ-at-risk (OAR)-sparing radiation therapy has typically reported outcomes based on fixed-number or fixed-interval re-planning, which represent one-size-fits-all approaches and do not account for the variable progression of individual patients’ toxicities. The purpose of this study was to determine the personalized optimal timing of re-planning in adaptive OAR-sparing radiation therapy, considering limited re-planning resources, for patients with head and neck cancer (HNC).</div></div><div><h3>Materials and methods</h3><div>A novel Markov decision process (MDP) model was developed to determine optimal timing of re-planning based on the patient’s expected toxicity, characterized by normal tissue complication probability (NTCP), for four toxicities. The MDP parameters were derived from a dataset comprising 52 HNC patients treated between 2007 and 2013. Kernel density estimation was used to smooth the sample distributions. Optimal re-planning strategies were obtained when the permissible number of re-plans throughout the treatment was limited to 1, 2, and 3, respectively.</div></div><div><h3>Results</h3><div>The MDP (optimal) solution recommended re-planning when the difference between planned and actual NTCPs (ΔNTCP) was greater than or equal to 1%, 2%, 2%, and 4% at treatment fractions 10, 15, 20, and 25, respectively, exhibiting a temporally increasing pattern. The ΔNTCP thresholds remained constant across the number of re-planning allowances (1, 2, and 3).</div></div><div><h3>Conclusion</h3><div>In limited-resource settings that impeded high-frequency adaptations, ΔNTCP thresholds obtained from an MDP model could derive optimal timing of re-planning to minimize the likelihood of treatment toxicities.</div></div>","PeriodicalId":36850,"journal":{"name":"Physics and Imaging in Radiation Oncology","volume":"33 ","pages":"Article 100715"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143350636","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Critical assessment of knowledge-based models for craniospinal irradiation of paediatric patients","authors":"Paolo Caricato ,&nbsp;Francesca Cavagnetto ,&nbsp;Silvia Meroni ,&nbsp;Salvina Barra ,&nbsp;Laura Brambilla ,&nbsp;Enrica Bovo ,&nbsp;Samuele Cavinato ,&nbsp;Alessio Cirone ,&nbsp;Flavio Giannelli ,&nbsp;Marta Paiusco ,&nbsp;Emilia Pecori ,&nbsp;Emanuele Pignoli ,&nbsp;Margherita Pollara ,&nbsp;Giovanni Scarzello ,&nbsp;Alessandro Scaggion","doi":"10.1016/j.phro.2025.100703","DOIUrl":"10.1016/j.phro.2025.100703","url":null,"abstract":"<div><h3>Background and purpose</h3><div>Knowledge-Based Planning (KBP) is increasingly used to standardize and optimize radiotherapy planning. This study aims to develop, refine, and compare multicentric KBP models for craniospinal irradiation (CSI) in pediatric patients.</div></div><div><h3>Materials and methods</h3><div>A total of 113 CSI treatments from three Italian centers were collected, comprising Computed Tomography scans, target and organ contours, and treatment plans. Treatment techniques included Helical Tomotherapy (HT) and Volumetric Modulated Arc Therapy (VMAT). Three KBP models were developed: a full model (F-model) using data from 87 patients, a reduced model (R-model) based on a subset of the same sample, and a replanned model (RP-model) using KBP re-optimized plans. Models’ quality was evaluated using goodness-of-fit and goodness-of-prediction metrics, and their performance was assessed on a validation set of 26 patients through dose-volume histogram (DVH) comparisons, prediction bias, and variance analysis.</div></div><div><h3>Results</h3><div>The F-model and R-model exhibited similar quality and predictive ability, reflecting the variability of the original dataset and resulting in broad prediction intervals in low to mid-dose ranges. The RP-model achieved the highest quality, with narrower prediction bands. The RP-model is preferable for standardizing planning across centers, while the F-model is better suited for quality assurance as it captures clinical variability.</div></div><div><h3>Conclusions</h3><div>KBP models can effectively predict DVHs despite extreme geometric variability. However, models trained on highly variable datasets cannot simultaneously achieve high precision and accuracy. Comparing KBP models is essential for establishing benchmarks to meet specific clinical goals, particularly for complex pediatric CSI treatments.</div></div>","PeriodicalId":36850,"journal":{"name":"Physics and Imaging in Radiation Oncology","volume":"33 ","pages":"Article 100703"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143130346","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Adaptive treatment margins to reduce organs at risk dose in patients with no or minimal anatomical changes in radiotherapy of non-small cell lung cancer","authors":"Djoya Hattu,&nbsp;Daisy Emans,&nbsp;Janine Bouten,&nbsp;Richard Canters,&nbsp;Judith van Loon,&nbsp;Dirk De Ruysscher","doi":"10.1016/j.phro.2025.100699","DOIUrl":"10.1016/j.phro.2025.100699","url":null,"abstract":"<div><h3>Background and purpose</h3><div>In non-small cell lung cancer (NSCLC) a significant portion of the planning target volume (PTV) margin accommodates for anatomical changes during treatment. Patients with no or minimal anatomical changes might therefore benefit from a reduced PTV margin, resulting in lower organ at risk (OAR) doses. We evaluated a plan of the day approach using different PTV margins to quantify its effect on OAR and clinical target volume (CTV) dose.</div></div><div><h3>Materials and methods</h3><div>Twenty NSCLC patients were included in this retrospective study. CBCTs of all fractions were evaluated using an image-guided radiotherapy (IGRT) protocol to classify fractions into two groups: no or minimal anatomical changes to which reduced PTV margin plans (5 or 2 mm) were assigned, or with anatomical changes that received the reference treatment plan (8 mm PTV margin). OAR doses were investigated and CTV coverage was evaluated using CBCT dose recalculations.</div></div><div><h3>Results</h3><div>All plans showed decreased OAR dose when the PTV margin was reduced from 8 mm to 5 mm or 2 mm. The IGRT protocol selected 254/600 fractions in 19/20 patients, that could be treated with a smaller margin. CTV V_95% remained ≥95% in 94% of the 5 mm plans and 87% of the 2 mm plans, compared to 98% of the reference 8 mm plans.</div></div><div><h3>Conclusion</h3><div>The IGRT protocol could identify fractions with no or minimal anatomical changes allowing a plan of the day approach to reduce PTV margins. Target coverage remained adequate in the majority of patients, while reducing OAR doses.</div></div>","PeriodicalId":36850,"journal":{"name":"Physics and Imaging in Radiation Oncology","volume":"33 ","pages":"Article 100699"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143130457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development and comprehensive clinical validation of a deep neural network for radiation dose modelling to enhance magnetic resonance imaging guided radiotherapy","authors":"Moritz Schneider ,&nbsp;Simon Gutwein ,&nbsp;David Mönnich ,&nbsp;Cihan Gani ,&nbsp;Paul Fischer ,&nbsp;Christian F. Baumgartner ,&nbsp;Daniela Thorwarth","doi":"10.1016/j.phro.2025.100723","DOIUrl":"10.1016/j.phro.2025.100723","url":null,"abstract":"<div><h3>Background and purpose</h3><div>Online adaptive magnetic resonance imaging (MRI)-guided radiotherapy requires fast dose calculation algorithms to reduce intra-fraction motion uncertainties and improve workflow efficiency. While Monte-Carlo simulations are precise but computationally intensive, neural networks promise fast and accurate dose modelling in strong magnetic fields. This study aimed to train and evaluate a deep neural network for dose modelling in MRI-guided radiotherapy using a comprehensive clinical dataset.</div></div><div><h3>Materials and methods</h3><div>A dataset of 6595 clinical irradiation segments from 125 1.5 T MRI-Linac radiotherapy plans for various tumors sites was used. A 3D U-Net was trained with 3961 segments using 3D imaging data and field parameters as input, Root Mean Squared Error and a custom loss function, with full Monte-Carlo simulations as ground truth. For 2656 segments from 50 patients, gamma pass rates (γ-PR) for 3 mm/3%, 2 mm/2%, and 1 mm/1% criteria were calculated to assess dose modelling accuracy. Performance was also tested in a standardized water phantom to evaluate basic radiation physics properties.</div></div><div><h3>Results</h3><div>The neural network accurately modeled dose distributions in both patient and water phantom settings. Median (range) γ-PR of 97.7% (87.5–100.0%), 89.1% (69.7–99.4%), and 60.8% (38.5–82.1%) were observed for treatment plans, and 97.1% (55.5–100.0%), 88.8% (38.8–99.7%), and 61.7% (17.9–94.4%) for individual segments, across the three criteria.</div></div><div><h3>Conclusion</h3><div>High median γ-PR and accurate modelling in both water phantom and clinical data demonstrate the high potential of neural networks for dose modelling. However, instances of lower γ-PR highlight the need for comprehensive test data, improved robustness and future built-in uncertainty estimation.</div></div>","PeriodicalId":36850,"journal":{"name":"Physics and Imaging in Radiation Oncology","volume":"33 ","pages":"Article 100723"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143480620","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}