Medical physics最新文献

{"title":"A proof-of-concept study of multitask learning for cranial synthetic CT generation across heterogeneous MRI field strengths.","authors":"Zhuoyao Xin, Yiren Zhang, Christopher Wu, Dong Liu, Chunming Gu, Elena Greco, Erik H Middlebrooks, Jun Hua, Jia Guo","doi":"10.1002/mp.70429","DOIUrl":"10.1002/mp.70429","url":null,"abstract":"<p><strong>Background: </strong>Accurate synthesis of computed tomography (CT) images from magnetic resonance imaging (MRI) is clinically valuable for cranial applications such as attenuation correction, radiotherapy planning, and image-guided interventions. However, the heterogeneity of MRI data across magnetic field strengths and pulse sequences limits the generalizability of existing methods, posing a barrier to clinical translation.</p><p><strong>Purpose: </strong>This study aims to reformulate cranial CT synthesis as a modular, structurally coupled problem using a deep learning approach, enhancing adaptability across heterogeneous MRI conditions, including different field strengths and sequence protocols.</p><p><strong>Methods: </strong>We implemented a cascaded multitask pipeline that jointly models skull segmentation and Hounsfield Unit (HU) regression in anatomically targeted regions. A 3D patch-based training paradigm shifts the modeling focus from global image translation to localized bone feature extraction. The backbone leverages a residual Mamba-based state space model within a 3D U-Net structure to improve spatial representation, while a Transformer U-Net serves as a widely recognized reference baseline comparison. The model was trained and evaluated in terms of multi-modal (T1-weighted and T2-FLAIR) and cross-domain transferability from a public 1.5 Tesla brain dataset (n = 37) to an independent 7 Tesla clinical brain dataset (n = 44). Performance was assessed using Dice and Jaccard indices for segmentation accuracy and mean absolute error (MAE) for HU regression.</p><p><strong>Results: </strong>Quantitative analysis showed that our multitask pipeline, incorporating both morphological and HU map prediction stages, significantly outperformed conventional direct MRI-to-CT mapping across all metrics on the public 1.5T dataset (p < 0.05). Consistent performance on an external 7T clinical dataset (p < 0.001) further demonstrated the method's robustness and adaptability across field strengths.</p><p><strong>Conclusion: </strong>These findings support that task-structured training for modality transformation markedly improves both accuracy and generalizability of cranial CT synthesis across heterogeneous MRI conditions. The observed consistency across field strengths validates the robustness of the proposed methodology.</p>","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":"53 5","pages":"e70429"},"PeriodicalIF":3.2,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13167256/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147793482","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 transformer-based patient-agnostic tumor motion tracking model (TransTracking): A feasibility study in radiotherapy for lung cancer.","authors":"Maidina Abuduxiku, Ningning Chai, Xiaoyan An, Yazhou Li, Jian Wang, Ying Luo, Wei Wu, Yinuo Liu, Yang Hu, Qiang Li, Pengbo He","doi":"10.1002/mp.70449","DOIUrl":"10.1002/mp.70449","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Accurate and real-time localization of thoracic tumor targets is essential for effective radiation therapy. Recently, Transformer architectures have demonstrated strong global reasoning capabilities across multiple frames by leveraging both self-attention and cross-attention mechanisms. Transformers have therefore been applied to object tracking with great success. By combining Image Guided Radiation Therapy (IGRT) technologies and deep learning-based object tracking architecture, it is possible to deliver radiation doses to the target area with high accuracy.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;This study develops a transformer-based patient-agnostic tracking model (TransTracking) for surface and markerless internal target tracking in thoracic tumor radiotherapy.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We trained the TransTracking model using the training splits of publicly available object tracking datasets. Subsequently, for internal target tracking, the model is fine-tuned using 10,000 digitally reconstructed radiograph (DRR) images generated from the actual 4DCT datasets of 25 patients. The DRR images are annotated with bounding boxes of the moving tumor. Our method learns to directly predict the target classification and bounding-box regression weights through end-to-end training, enabling accurate target localization in each frame for both surface and internal target tracking sequences. The tracking performance of the trained model was evaluated in 20 volunteers for surface tracking and using DRR images generated from 20 4DCT datasets for internal tumor tracking. To address the limited availability of medical images for training, we conducted the data augmentation procedure to 4DCT datasets and expanded the data scale 40-fold in total.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;For the surface marker tracking, the mean absolute deviation (MAD) ± standard deviation (SD) between the model-predicted and the actual positions for 20 volunteers was 0.07 ± 0.06 mm, 0.12 ± 0.13 mm, and 0.29 ± 0.20 mm in left-right, superior-inferior, and anterior-posterior directions, respectively. In each directional axis, over 85% of frames exhibited a model-predicted target position within 0.5 mm of the corresponding ground-truth position. For the internal tumor tracking, the MAD ± SD between the predicted and annotated center positions of the tumor bounding boxes is 1.49 ± 1.39 mm, with a mean Intersection over Union (IoU) value of 0.83 and an area under curve (AUC) score of 82% for 20 patients. Additionally, our transformer-based model can extract the target position in 81 ms after an image is acquired.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;This study proposed a novel Transformer-based deep learning method aimed at training a patient-agnostic tumor motion tracking model in radiotherapy. The model enables real-time, high-precision tracking of surface markers using vision cameras, offering a cost-effective and compact solution. Additionally, we demonstra","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":"53 5","pages":"e70449"},"PeriodicalIF":3.2,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147793240","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Uncertainty-aware gamma interaction localization and reconstruction in PET.","authors":"Julian Thull, Jan Remennik, David Schug, Bjoern Weissler, Yannick Kuhl, Volkmar Schulz","doi":"10.1002/mp.70471","DOIUrl":"10.1002/mp.70471","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Precise localization of gamma-ray interactions inside scintillation detectors is essential for high-resolution positron emission tomography (PET) imaging. Although machine learning methods have demonstrated strong performance in gamma interaction positioning, most existing approaches do not quantify event-level uncertainty, leaving valuable information unused.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;This study quantifies event-wise positional uncertainties in gamma interaction localization and demonstrates its utility for improving PET image quality through uncertainty-aware filtering and weighting.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We employ machine learning to perform gamma interaction positioning in a semi-monolithic scintillation detector block with overall dimensions of &lt;math&gt; &lt;semantics&gt;&lt;mrow&gt;&lt;mn&gt;48.4&lt;/mn&gt; &lt;mo&gt;×&lt;/mo&gt; &lt;mn&gt;48.3&lt;/mn&gt; &lt;mo&gt;×&lt;/mo&gt; &lt;mn&gt;10&lt;/mn&gt; &lt;mspace&gt;&lt;/mspace&gt; &lt;msup&gt;&lt;mi&gt;mm&lt;/mi&gt; &lt;mn&gt;3&lt;/mn&gt;&lt;/msup&gt; &lt;/mrow&gt; &lt;annotation&gt;$48.4 times 48.3 times 10 ,mathrm{mm}^3$&lt;/annotation&gt;&lt;/semantics&gt; &lt;/math&gt; , corresponding to the planar-segmented, planar-monolithic, and depth-of-interaction (DOI) directions. We train multilayer perceptrons (MLPs) and convolutional neural networks (CNNs) using silicon photomultiplier (SiPM) light spread measurements as network inputs. Regression models were trained using a Gaussian negative log-likelihood to jointly estimate gamma interaction coordinates and event-wise positional variance. Classifiers were evaluated, which inferred event-level uncertainty from the variance of the predicted spatial probability distribution. These uncertainty estimates primarily reflect variability in detector response and photon statistics, corresponding to aleatoric uncertainty. Regression and classification objectives were explored through task-specific hyperparameter optimizations. All 24 semi-monolithic detectors of a proof-of-concept PET system were calibrated in the segmented, monolithic, and depth-of-interaction (DOI) dimensions using collimated fan-beam irradiations of each detector. For each detector, a dataset of 880,000 events per spatial dimension was acquired and randomly split into 60% training, 20% validation, and 20% test sets, with models trained and evaluated independently across seven detectors. The resulting calibration models were then applied to reprocess measurements of imaging phantoms. Performance was evaluated using the mean absolute error (MAE) on the fan-beam dataset and the median distance between reconstructed lines of response (LORs) and known point-source locations measured within the scanner. Predicted variances were further integrated into the time-of-flight ordered-subsets expectation maximization (TOF-OSEM) reconstruction via event-level filtering and LOR weighting to assess spatial resolution and noise propagation.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Across architectures, uncertainty-aware models achieved high positioning accuracy. CNN classifiers provided the best planar performa","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":"53 5","pages":"e70471"},"PeriodicalIF":3.2,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13176420/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147944404","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":"Dose prediction for head-and-neck cancer using a fusion of lightweight 3D multi-scale feature enhancement modules.","authors":"Long Wei, Yishen Yue, Haijiao Shang, Zhitao Dai, Xile Zhang","doi":"10.1002/mp.70481","DOIUrl":"10.1002/mp.70481","url":null,"abstract":"<p><strong>Background: </strong>Radiotherapy is a cornerstone of head-and-neck cancer (HNC) treatment, but traditional radiation therapy planning remains time-consuming, experience-dependent, and prone to inconsistent quality. Deep learning-based dose prediction has emerged as a promising solution, yet existing models struggle with insufficient long-range spatial correlation capture and imbalanced prediction accuracy across high- and low-dose regions. Thus, there is an urgent need for a tailored framework to address these clinical challenges.</p><p><strong>Purpose: </strong>This study proposes a U-Net-based encoder-decoder model fused with lightweight 3D multi-scale feature enhancement modules for 3D HNC radiotherapy dose prediction, aiming to improve target dose precision and organs-at-risk (OARs) sparing.</p><p><strong>Methods: </strong>A public OpenKBP dataset consisting of 340 HNC patients (200 for training, 40 for validation, 100 for testing) undergoing 6MV IMRT was utilized. Data were preprocessed (including CT truncation, normalization, multimodal integration) and augmented (random flipping, translation, rotation) to enhance generalization. The proposed model integrates an eight-layer Transformer for global feature extraction, 3D multi-scale convolutional blocks (MSCBs) for fine-grained feature capture, and an efficient multi-scale convolutional attention decoding (EMCAD) module for optimized feature fusion. A baseline DOSE-PYFER model, GAN-based models, and a cascade Transformer-based model were used for comparative analysis. Performance was evaluated using Dose score, DVH score, and gamma passing rate.</p><p><strong>Results: </strong>The proposed model exhibited superior accuracy compared to comparative models. It achieved a Dose score of 2.704 Gy and a DVH score of 1.611 Gy, outperforming the baseline and cascade Transformer-based model. The gamma passing rate reached 92.76%, indicating excellent spatial consistency with ground truth. No overfitting was observed, with stable training and validation loss curves. The model efficiently generates 3D dose distributions, supporting rapid clinical workflow.</p><p><strong>Conclusion: </strong>The proposed model, integrating Transformer-driven global feature extraction and EMCAD-based multi-scale attention decoding, outperforms conventional models in HNC dose prediction. It effectively resolves long-range correlation capture and dose region balance issues, and can be clinically deployed to streamline radiotherapy planning, enhance plan consistency, and support timely adaptive radiotherapy.</p>","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":"53 5","pages":"e70481"},"PeriodicalIF":3.2,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147944429","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards quality control and harmonization of deep learning CT radiomics: An in-silico feasibility study with virtual colorectal liver metastases.","authors":"Mamatha Venugopal, Sathish Ramani, Jacob J Peoples, Richard K G Do, Amber L Simpson, Ge Wang, Bruno De Man","doi":"10.1002/mp.70500","DOIUrl":"10.1002/mp.70500","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Radiomic imaging biomarkers are increasingly studied in oncology as a means to support disease prognosis and personalized treatment planning. While deep learning (DL) offers scalable alternatives to handcrafted radiomic features, DL-derived biomarkers are sensitive to variations in image acquisition protocols and scanner hardware-even within a single imaging modality. To ensure reliable and reproducible biomarker estimation, it is essential to (1) provide clinicians with quantitative uncertainty estimates associated with biomarker predictions, and (2) address acquisition-induced variability through harmonization strategies that render consistent performance across diverse imaging conditions.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;To evaluate uncertainty estimation as a reliability metric for biomarker prediction and to assess the role of image harmonization in improving cross-scanner inference, we develop deep learning radiomics (DLR) models for joint estimation of biomarkers and associated uncertainties and apply linear harmonization filters to standardize imaging conditions.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We constructed hybrid digital phantoms by embedding 20,000 virtual colorectal liver metastases into 20 clinical CT liver images to generate metastases-laden simulated scans. The proposed DLR models jointly estimated a selected set of biomarkers and their associated aleatoric uncertainties from simulated images. Models were trained, validated, and tested using a 70:20:10 data split. Variability in CT image acquisition was modelled using two scanner types, two reconstruction kernels, and three x-ray tube current settings. DLR performance was evaluated under direct inference (matched training and test scanners), cross-scanner inference, and harmonized inference using linear harmonization filters. Quantitative evaluation was based on correlation coefficients and root mean squared errors (RMSEs).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The estimated uncertainties were consistently higher for biomarker predictions with larger deviations from ground truth. Excluding high-uncertainty predictions improved concordance between predictions and ground truth biomarkers. The proportion of uncertain predictions increased under cross-scanner inference relative to direct inference, indicating reduced biomarker reliability under heterogeneous imaging conditions. Furthermore, application of harmonization filters reduced RMSEs by an average of 43% across biomarkers and experiments during cross-scanner inference, demonstrating improved cross-scanner consistency.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;In this controlled in-silico study, uncertainty estimation provided a practical reliability metric for DL-based radiomic biomarker prediction, while image harmonization improved reproducibility across heterogeneous acquisition conditions. These findings demonstrate methodological feasibility within a simulation-based framework but have not yet been valid","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":"53 5","pages":"e70500"},"PeriodicalIF":3.2,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147990458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Magnetic resonance imaging-guided large-scale lesion patterning with high-intensity focused ultrasound in homogeneous phantoms.","authors":"Anastasia Antoniou, Nikolas Evripidou, Leonidas Georgiou, Antonis Christofi, Jufeng Zhao, Liyang Yu, Wenjun Li, George C Kagadis, Christakis Damianou","doi":"10.1002/mp.70503","DOIUrl":"10.1002/mp.70503","url":null,"abstract":"<p><strong>Background: </strong>In typical high-intensity focused ultrasound (HIFU) therapy, energy is delivered sequentially to multiple focal points to collectively shape the overall treatment zone. However, the behavior of large-scale grid sonication patterns in homogeneous media under magnetic resonance imaging (MRI) guidance has not been fully characterized.</p><p><strong>Purpose: </strong>To characterize the spatial fidelity and MRI appearance of large-scale HIFU sonication grids in an agar-silica tissue-mimicking phantom and to describe their observed behavior relative to biological tissue.</p><p><strong>Methods: </strong>Rectangular and irregular large-scale (∼100-point) grids were executed in agar-silica phantoms using an MRI-guided HIFU robotic system with a single-element transducer under 3T MRI guidance. T2-weighted (T2-w) Turbo Spin Echo (TSE) images were acquired after each sonication row for lesion monitoring and post-sonication quantification. Each grid configuration was evaluated in a separate phantom preparation. Descriptive metrics were used to characterize lesion morphology and spatial fidelity, quantified using lesion diameter and Euclidean localization error when individual lesions were distinguishable, and region-level descriptors (area and centroid position) for merged lesion regions. One of the grid protocols was also applied to freshly excised porcine skeletal muscle, with lesion formation monitored using the same MRI-based approach.</p><p><strong>Results: </strong>In the phantom, lesions appeared as hyperintense regions at the intended focal depth and closely reproduced the prescribed grid geometry. For a standardized 10 × 10 grid (4 mm spacing) in the ablative regime, individual lesions exhibited a mean diameter of 3.0 ± 0.7 mm and a mean Euclidean localization error of ∼2.5 ± 0.7 mm relative to the planned coordinates. When inter-point spacing approached the lesion diameter, adjacent sonications produced merged regions that preserved the overall grid footprint while forming contiguous thermal coverage. In the irregular configuration, grid coverage was 96%, with a centroid offset of ∼1 mm. Under the tested conditions, ex vivo porcine muscle exhibited incomplete lesion formation and a shift of ∼15 mm from the prescribed depth, producing an irregular hypointense region displaced toward the tissue surface.</p><p><strong>Conclusions: </strong>This study documented the successful generation and MRI visualization of large-scale HIFU grid patterns in an agar-silica phantom using T2-w imaging. Under the tested conditions, the observed hyperintense regions showed spatial correspondence with the prescribed grid geometry. Observations in freshly excised porcine skeletal muscle recorded variations in lesion formation in the presence of biological heterogeneity, while the agar-silica phantom provided a standardized platform for system-level testing as configured in this study.</p>","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":"53 5","pages":"e70503"},"PeriodicalIF":3.2,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148007745","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development and validation of an automated, accurate in-house treatment planning system for pencil-beam scanning carbon ion radiotherapy.","authors":"Sac Lee, Yongdo Yun, Min Cheol Han, Changhwan Kim, Chae-Seon Hong, Ho Lee, Dong Wook Kim, Dong Min Jung, Jin Sung Kim, Hojin Kim","doi":"10.1002/mp.70456","DOIUrl":"10.1002/mp.70456","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Carbon ion radiotherapy (CIRT) offers superior physical and biological advantages over photon and proton radiotherapy (PRT). However, due to a need for accurate modeling of the relative biological effectiveness (RBE), currently available treatment planning systems (TPS) for CIRT remain limited. This has constrained clinical and research applications by compelling reliance on a narrow set of systems and restricting flexibility for broader application.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;This study aimed to develop and validate a new efficient and accurate CIRT TPS for pencil-beam scanning (PBS)-based CIRT, tailored for the Heavy-Ion Therapy Center (HITC) at Yonsei Cancer Center (YCC), incorporating modified microdosimetric kinetic model (mMKM)-based RBE-weighted dose calculation and spot weight optimization.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;The proposed TPS was designed to automate the workflow from CT image import to CIRT plan optimization. Key refinements included 3D Siddon's ray tracing for spot trajectory tracking, air-gap modeling for beam profile adaptation, and 3D stopping power ratio (SPR) calculation. Gaussian beam modeling was performed to calibrate in-air measurements and Monte Carlo (MC)-driven 3D dose kernels in water. These procedures produced dose influence matrices (DIMs) and MC-estimated &lt;math&gt; &lt;semantics&gt;&lt;msubsup&gt;&lt;mi&gt;z&lt;/mi&gt; &lt;mrow&gt;&lt;mn&gt;1&lt;/mn&gt; &lt;mi&gt;D&lt;/mi&gt;&lt;/mrow&gt; &lt;mo&gt;∗&lt;/mo&gt;&lt;/msubsup&gt; &lt;annotation&gt;$z_{1D}^*$&lt;/annotation&gt;&lt;/semantics&gt; &lt;/math&gt; matrices required for spot weight optimization with mMKM-based RBE-weighted dose calculation, which was further accelerated by Adam optimizer. Validation was performed by generating plans for a water phantom (10 Gy-RBE), a lung case (15 Gy-RBE), and a prostate case (4.3 Gy-RBE), using single-field and multi-field optimizations (MFO). The resulting plans were recalculated in RayStation (v.2025) and TOPAS MC, and compared using dose-volume histogram (DVH) and gamma passing rate (GPR) at 2%/2 mm.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The proposed TPS successfully automated the CIRT plan optimization within approximately 1 min for ∼1000 spots, which produced uniform RBE-weighted dose coverage in virtual water phantom, the lung patient, and the prostate patient cases. When the spot weights optimized by the proposed framework were recalculated using a commercial TPS and TOPAS MC, the resulting dose distributions closely matched those of the proposed TPS. Quantitatively, GPRs exceeding 98% were achieved for both physical and RBE-weighted dose at the 2%/2 mm criterion, except for the prostate single-field case, which yielded a GPR of 97.37% for the physical dose, relative to TOPAS MC, and a GPR of 96.28% for the RBE-weighted dose, relative to RayStation. These findings confirmed strong agreement between the proposed TPS, MC simulations, and a commercial TPS.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusion: &lt;/strong&gt;A new, fully automated in-house CIRT TPS was developed and validated, demonstrating high dosimet","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":"53 5","pages":"e70456"},"PeriodicalIF":3.2,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147793243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Triple collaborative consistency with Mamba for semi-supervised 3D medical image segmentation.","authors":"Yufei Gao, Bingning Liu, Qing Li, Yameng Zhang, Yujie Shi","doi":"10.1002/mp.70390","DOIUrl":"10.1002/mp.70390","url":null,"abstract":"<p><strong>Background: </strong>Current semi-supervised segmentation methods face the following challenges: (1) Cross-branch collaboration: Existing methods typically rely on single-branch pseudo-label generation or simple multi-view fusion strategies, failing to fully exploit the interaction between local details and global structures. This limitation leads to suboptimal performance in the boundary segmentation of complex anatomical structures. (2) Inefficiency in long-range modeling: While Transformer-based methods can capture global dependencies, they suffer from quadratic growth in computational complexity and the risk of overfitting when applied to high-resolution data (e.g., 3D medical images), making it difficult to balance efficiency and accuracy.</p><p><strong>Purpose: </strong>To address the above challenges, this article proposes a Tri-branch Collaborative Consistency model based on Mamba long-range modeling (TCC-Mamba), which aims to reduce reliance on annotations while improving segmentation accuracy in complex regions of medical images.</p><p><strong>Methods: </strong>TCC-Mamba consists of a shared encoder and a tri-branch decoder. Specifically, a tri-branch collaborative supervision mechanism is introduced where three decoders form a closed-loop learning system through cross-pseudo-label supervision, enabling collaborative optimization and information sharing. Additionally, a geometric consistency loss function is incorporated to enhance boundary awareness. Furthermore, we integrate the SpatialTriMamba module, leveraging the efficient long-range dependency modeling of state-space models to achieve dynamic fusion of global context and local features, thereby improving segmentation accuracy for complex boundaries.</p><p><strong>Results: </strong>We conducted experiments on three public datasets: Left Atrium (LA), Pancreas CT, and ACDC, using 10%, 20%, and 30% labeled data. The results demonstrate that our method outperforms the six current advanced semi-supervised methods, achieving better segmentation performance.</p><p><strong>Conclusions: </strong>The TCC-Mamba introduces novel methodologies in medical image segmentation tasks. This model combines the SpatialTriMamba module to capture long-range features and utilizes signed distance maps to enhance the use of geometric information, leading to exceptional results in handling complex anatomical structures. It provides an efficient and reliable solution for semi-supervised medical image segmentation.</p>","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":"53 5","pages":"e70390"},"PeriodicalIF":3.2,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147936709","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correction to \"Experimental investigation of real-time 3D beam's eye view image-guided radiotherapy for prostate SBRT\".","authors":"","doi":"10.1002/mp.70417","DOIUrl":"10.1002/mp.70417","url":null,"abstract":"","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":"53 5","pages":"e70417"},"PeriodicalIF":3.2,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147936741","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Phantom evaluation of spectral performance in photon-counting CT for breast cancer imaging.","authors":"Liqiang Ren, Yin Xi, Lakshmi Ananthakrishnan, Todd Soesbe, Matthew Lewis, Stephen Seiler, Afrouz Ataei, Hong Liu, Yuhua Li, Richard W Ahn","doi":"10.1002/mp.70472","DOIUrl":"10.1002/mp.70472","url":null,"abstract":"<p><strong>Background: </strong>Contrast enhancement is the most sensitive indicator for detecting breast malignancies. Computed tomography (CT) has had a limited role for the locoregional staging of breast tumors due to low soft tissue contrast.</p><p><strong>Purpose: </strong>To evaluate and optimize the performance of clinical photon-counting computed tomography (PCCT) for breast cancer imaging using a contrast-enhanced mammography (CEM) phantom and to compare its imaging performance with dual-source dual-energy CT (DS-DECT).</p><p><strong>Methods: </strong>A CEM phantom containing simulated breast lesions was positioned on an anthropomorphic thoracic phantom and scanned using a clinical PCCT system at 120 kV in multi-energy mode and a DS-DECT system with two kV pairs of 70/Sn150 kV and 90/Sn150 kV. PCCT scanner variables included scan mode [standard resolution (SR) and ultra-high resolution (UHR)], field of view (FOV) size (large and small), matrix size (512 and 1024), and type of image used for analysis [low-energy threshold images, virtual monoenergetic images (VMIs) at 50, 60, and 70 keV, and iodine maps]. Quantitative analysis was performed using circular regions of interest (ROIs) placed on iodine-containing lesions and background within the phantom. For each ROI, mean CT numbers or iodine concentrations and standard deviations were measured across the central five slices and three independent scans. Contrast-to-noise ratio (CNR) and circularity were evaluated across all PCCT configurations and image types and compared with those obtained from DS-DECT.</p><p><strong>Results: </strong>Among all PCCT configurations, the UHR mode with a small FOV and either a 512 or 1024 matrix at 50 keV VMI achieved the highest combined CNR across all iodine concentrations. Additionally, the UHR mode with a 512 matrix and either small or large FOV yielded the highest combined circularity values. The optimal PCCT configuration achieved higher CNR, and higher or comparable circularity compared with 50 keV VMIs derived from DECT scans.</p><p><strong>Conclusions: </strong>This phantom study demonstrated that optimal spectral performance for potential breast cancer imaging with PCCT is achieved using UHR mode, low-keV VMIs, a regular matrix size, and dedicated reconstruction FOVs, outperforming DECT.</p>","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":"53 5","pages":"e70472"},"PeriodicalIF":3.2,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147936703","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}