Medical physics最新文献

筛选
英文 中文
Dosimetric impact of positional uncertainties and a robust optimization approach for rectal intensity-modulated brachytherapy. 位置不确定性的剂量学影响以及直肠强度调制近距离治疗的稳健优化方法。
Medical physics Pub Date : 2025-03-31 DOI: 10.1002/mp.17800
Björn Morén, Alana Thibodeau-Antonacci, Jonathan Kalinowski, Shirin A Enger
{"title":"Dosimetric impact of positional uncertainties and a robust optimization approach for rectal intensity-modulated brachytherapy.","authors":"Björn Morén, Alana Thibodeau-Antonacci, Jonathan Kalinowski, Shirin A Enger","doi":"10.1002/mp.17800","DOIUrl":"https://doi.org/10.1002/mp.17800","url":null,"abstract":"<p><strong>Background: </strong>Intensity-modulated brachytherapy (IMBT) employs rotating high-Z shields during treatment to decrease radiation in certain directions and conform the dose distribution to the target volume. Prototypes for dynamic IMBT have been proposed for prostate, cervical, and rectal cancer.</p><p><strong>Purpose: </strong>We considered two shielded applicators for IMBT rectal cancer treatment and investigated how rotational uncertainties in the shield angle and translational uncertainties in the source position affect plan evaluation criteria.</p><p><strong>Methods: </strong>The effect of rotational errors of <math> <semantics><msup><mn>3</mn> <mo>∘</mo></msup> <annotation>$3^circ$</annotation></semantics> </math> , <math> <semantics><msup><mn>5</mn> <mo>∘</mo></msup> <annotation>$5^circ$</annotation></semantics> </math> and <math> <semantics><msup><mn>10</mn> <mo>∘</mo></msup> <annotation>$10^circ$</annotation></semantics> </math> , and translational errors of 1, 2 and 3 mm on evaluation criteria were investigated for shields with <math> <semantics><msup><mn>180</mn> <mo>∘</mo></msup> <annotation>${rm 180}^circ$</annotation></semantics> </math> and <math> <semantics><msup><mn>90</mn> <mo>∘</mo></msup> <annotation>${rm 90}^circ$</annotation></semantics> </math> emission windows. Further, a robust optimization approach based on quadratic penalties that includes scenarios with errors was proposed. The extent to which dosimetric effects of positional errors can be mitigated with this model was evaluated compared to a quadratic penalty model without scenarios with errors. A retrospective rectal cancer data set of ten patients was included in this study. Treatment planning was performed using the Monte Carlo-based treatment planning system, RapidBrachyMCTPS.</p><p><strong>Results: </strong>For the largest investigated rotational error of <math> <semantics><mrow><mo>±</mo> <msup><mn>10</mn> <mo>∘</mo></msup> </mrow> <annotation>$pm 10^circ$</annotation></semantics> </math> , the clinical target volume  <math> <semantics><msub><mi>D</mi> <mn>90</mn></msub> <annotation>${rm D}_{90}$</annotation></semantics> </math> remained, on average, within <math> <semantics><mrow><mn>5</mn> <mo>%</mo></mrow> <annotation>$5%$</annotation></semantics> </math> of the result without error, while the contralateral healthy rectal wall experienced an increase in the mean <math> <semantics><msub><mi>D</mi> <mrow><mn>0.1</mn> <mi>c</mi> <mi>c</mi></mrow> </msub> <annotation>${rm D}_{0.1cc}$</annotation></semantics> </math> , <math> <semantics><msub><mi>D</mi> <mrow><mn>2</mn> <mi>c</mi> <mi>c</mi></mrow> </msub> <annotation>${rm D}_{2cc}$</annotation></semantics> </math> , and <math> <semantics><msub><mi>D</mi> <mn>50</mn></msub> <annotation>${rm D}_{50}$</annotation></semantics> </math> of <math> <semantics><mrow><mn>26</mn> <mo>%</mo></mrow> <annotation>$26%$</annotation></semantics> </math> , <math> <semantics><mrow><mn>9</mn> <mo>%</mo></mrow> <annotat","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143756819","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}
引用次数: 0
Improving realism in abdominal ultrasound simulation combining a segmentation-guided loss and polar coordinates training. 结合分割引导损失和极坐标训练提高腹部超声模拟的逼真度。
Medical physics Pub Date : 2025-03-30 DOI: 10.1002/mp.17801
Santiago Vitale, José Ignacio Orlando, Emmanuel Iarussi, Alejandro Díaz, Ignacio Larrabide
{"title":"Improving realism in abdominal ultrasound simulation combining a segmentation-guided loss and polar coordinates training.","authors":"Santiago Vitale, José Ignacio Orlando, Emmanuel Iarussi, Alejandro Díaz, Ignacio Larrabide","doi":"10.1002/mp.17801","DOIUrl":"https://doi.org/10.1002/mp.17801","url":null,"abstract":"<p><strong>Background: </strong>Ultrasound (US) simulation helps train physicians and medical students in image acquisition and interpretation, enabling safe practice of transducer manipulation and organ identification. Current simulators generate realistic images from reference scans. Although physics-based simulators provide real-time images, they lack sufficient realism, while recent deep learning-based models based on unpaired image-to-image translation improve realism but introduce anatomical inconsistencies.</p><p><strong>Purpose: </strong>We propose a novel framework to reduce hallucinations from generative adversarial networks (GANs) used on physics-based simulations, enhancing anatomical accuracy and realism in abdominal US simulation. Our method aims to produce anatomically consistent images free from artifacts within and outside the field of view (FoV).</p><p><strong>Methods: </strong>We introduce a segmentation-guided loss to enforce anatomical consistency by using a pre-trained Unet model that segments abdominal organs from physics-based simulated scans. Penalizing segmentation discrepancies before and after the translation cycle helps prevent unrealistic artifacts. Additionally, we propose training GANs on images in polar coordinates to limit the field of view to non-blank regions. We evaluated our approach on unpaired datasets comprising 617 real abdominal US images from a SonoSite-M turbo v1.3 scanner and 971 artificial scans from a ray-casting simulator. Data was partitioned at the patient level into training (70%), validation (10%), and testing (20%). Performance was quantitatively assessed with Frechet and Kernel Inception Distances (FID and KID), and organ-specific <math> <semantics><msup><mi>χ</mi> <mn>2</mn></msup> <annotation>$chi ^2$</annotation></semantics> </math> histogram distances, reporting 95% confidence intervals. We compared our model against generative methods such as CUT, UVCGANv2, and UNSB, performing statistical analyses using Wilcoxon tests (FID and KID with Bonferroni-corrected <math> <semantics><mrow><mi>α</mi> <mo>=</mo> <mn>0.01</mn></mrow> <annotation>$alpha = 0.01$</annotation></semantics> </math> , <math> <semantics><msup><mi>χ</mi> <mn>2</mn></msup> <annotation>$chi ^2$</annotation></semantics> </math> with <math> <semantics><mrow><mi>α</mi> <mo>=</mo> <mn>0.008</mn></mrow> <annotation>$alpha =0.008$</annotation></semantics> </math> ). A perceptual realism study involving expert radiologists was also conducted.</p><p><strong>Results: </strong>Our method significantly reduced FID and KID by 66% and 89%, respectively, compared to CycleGAN, and by 34% and 59% compared to the leading alternative UVCGANv2 ( <math> <semantics><mrow><mi>p</mi> <mo>≪</mo> <mn>0.01</mn></mrow> <annotation>$p ll 0.01$</annotation></semantics> </math> ). No significant differences ( <math> <semantics><mrow><mi>p</mi> <mo>></mo> <mn>0.008</mn></mrow> <annotation>$p>0.008$</annotation></semantics> </math> ) in echogenicity distrib","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143756863","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}
引用次数: 0
Hybrid method for estimating lung ventilation from CT by combining intensity and motion information. 结合强度和运动信息从 CT 估算肺通气量的混合方法。
Medical physics Pub Date : 2025-03-30 DOI: 10.1002/mp.17787
Paris Tzitzimpasis, Mario Ries, Bas W Raaymakers, Cornel Zachiu
{"title":"Hybrid method for estimating lung ventilation from CT by combining intensity and motion information.","authors":"Paris Tzitzimpasis, Mario Ries, Bas W Raaymakers, Cornel Zachiu","doi":"10.1002/mp.17787","DOIUrl":"https://doi.org/10.1002/mp.17787","url":null,"abstract":"<p><strong>Background: </strong>Functional lung imaging modalities allow for capturing regional lung ventilation information. Computed Tomography based ventilation imaging (CTVI) has been proposed as a surrogate modality that relies on time-resolved anatomical data and image processing. However, generating accurate ventilation maps using solely computed tomography (CT) image information remains a challenging task, due to the need to derive functional information of ventilation from anatomical observations.</p><p><strong>Purpose: </strong>We introduce the hybrid estimation of computed tomography obtained respiratory function (HECTOR) method that consists of two components: a volume- and a density-based ventilation estimate. For the first component, a deformable image registration (DIR)-based solution for accurate volumetric CTVI generation is proposed, integrating the physical characteristics of the lung deformations in its design. For the second component, an already established air-tissue density model is used. Furthermore, a novel method is developed for combining the two components.</p><p><strong>Methods: </strong>The proposed method consists of four principal steps: (1) Application of a specially tailored DIR algorithm to estimate respiratory motion between inhale and exhale phases. (2) Conversion of the motion information to volumetric change maps using a variation of the Jacobian determinant method. (3) Computation of a HU-based method that estimates the local product of air-tissue densities. (4) Combination of the metrics estimated in steps 2 and 3 by means of a smooth minimum function. The proposed approach is validated using the publicly available VAMPIRE dataset consisting of two subgroups: 25 subjects scanned with Galligas 4DPET/CT and 21 subjects scanned with DTPA-SPECT. Another dataset of 18 patients available at The Cancer Imaging Archive (TCIA) was used for further validation. All datasets contain inhale/exhale CT scans paired with ground-truth ventilation images (RefVIs). The CTVIs generated by the proposed HECTOR method were tested against the RefVIs using the Spearman correlation coefficient and Dice overlap of low- and high-function lung (DSC-low and DSC-high, respectively).</p><p><strong>Results: </strong>The proposed method achieved mean Spearman, DSC-high and DSC-low coefficients of 0.62, 0.55, and 0.59 on the Galligas PET subgroup and 0.49,0,48, and 0.50 on the DTPA-SPECT subgroup of the VAMPIRE dataset. This performance was better than the highest performing method reported in the original challenge. The same metrics for the TCIA dataset were 0.66, 0.60, and 0.60. The proposed hybrid ventilation method achieved higher Spearman correlation scores than the individual volume- and density-based components in all datasets. Additionally, the use of the specially tailored DIR algorithm was found to achieve higher scores than previously reported volume-based methods.</p><p><strong>Conclusions: </strong>Our work provides a novel pr","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143756822","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}
引用次数: 0
Proton dose calculation with transformer: Transforming spot map to dose.
Medical physics Pub Date : 2025-03-29 DOI: 10.1002/mp.17794
Xueyan Tang, Hok Wan Chan Tseung, Mark D Pepin, Jed E Johnson, Doug J Moseley, David M Routman, Jing Qian
{"title":"Proton dose calculation with transformer: Transforming spot map to dose.","authors":"Xueyan Tang, Hok Wan Chan Tseung, Mark D Pepin, Jed E Johnson, Doug J Moseley, David M Routman, Jing Qian","doi":"10.1002/mp.17794","DOIUrl":"https://doi.org/10.1002/mp.17794","url":null,"abstract":"<p><strong>Background: </strong>Conventional proton dose calculation methods are either time- and resource-intensive, like Monte Carlo (MC) simulations, or they sacrifice accuracy, as seen with analytical methods. This trade-off between computational efficiency and accuracy highlights the need for improved dose calculation approaches in clinical settings.</p><p><strong>Purpose: </strong>This study aims to develop a deep-learning-based model that calculates dose-to-water (D<sub>W</sub>) and dose-to-medium (D<sub>M</sub>) using patient anatomy and proton spot map (PSM), achieving approaching MC-level accuracy with significantly reduced computation time. Additionally, the study seeks to generalize the model to different treatment sites using transfer learning.</p><p><strong>Methods: </strong>A SwinUNetr model was developed using 259 four-field prostate proton stereotactic body radiation therapy (SBRT) plans to calculate patient-specific D<sub>W</sub> and D<sub>M</sub> distributions from CT and projected PSM (PPSM). The PPSM was created by projecting PSM into the CT scans using spot coordinates, stopping power ratio, beam divergence, and water-equivalent thickness. Fine-tuning was then performed for the central nervous system (CNS) site using 84 CNS plans. The model's accuracy was evaluated against MC simulation benchmarks using mean absolute error (MAE), gamma analysis (2% local dose difference, 2-mm distance-to-agreement, 10% low dose threshold), and relevant clinical indices on the test dataset.</p><p><strong>Results: </strong>The trained model achieved a single-field dose calculation time of 0.07 s on a Nvidia-A100 GPU, over 100 times faster than MC simulators. For the prostate site, the best-performing model showed an average MAE of 0.26 ± 0.17 Gy and a gamma index of 92.2% ± 3.1% in dose regions above 10% of the maximum dose for D<sub>W</sub> calculations, and an MAE of 0.30 ± 0.19 Gy with a gamma index of 89.7% ± 3.9% for D<sub>M</sub> calculations. After transfer learning for CNS plans, the model achieved an MAE of 0.49 ± 0.24 Gy and a gamma index of 90.1% ± 2.7% for D<sub>W</sub> computations, and an MAE of 0.47 ± 0.25 Gy with a gamma index of 85.4% ± 7.1% for D<sub>M</sub> computations.</p><p><strong>Conclusions: </strong>The SwinUNetr model provides an efficient and accurate method for computing dose distributions in proton therapy. It also opens the possibility of reverse-engineering PSM from D<sub>W</sub>, potentially speeding up treatment planning while maintaining accuracy.</p>","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143744656","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}
引用次数: 0
Characterization and inter-scanner reproducibility of geometric distortion on a small footprint, high-performance, head-specific 0.5 T scanner.
Medical physics Pub Date : 2025-03-29 DOI: 10.1002/mp.17789
Curtis N Wiens, Chad T Harris, Ian R O Connell
{"title":"Characterization and inter-scanner reproducibility of geometric distortion on a small footprint, high-performance, head-specific 0.5 T scanner.","authors":"Curtis N Wiens, Chad T Harris, Ian R O Connell","doi":"10.1002/mp.17789","DOIUrl":"https://doi.org/10.1002/mp.17789","url":null,"abstract":"<p><strong>Background: </strong>Magnetic resonance imaging (MRI) offers superior soft tissue contrast and essential imaging capabilities for modern medicine. MRI is increasingly being used in applications that require a high degree of spatial fidelity; however, distortions are a well-known limitation of the modality. The mid-field (0.3 T ≤ B<sub>0</sub> < 1 T) has advantages in this respect due to being less susceptible to patient-induced distortions.</p><p><strong>Purpose: </strong>The purpose of this work was to characterize the geometric fidelity of a short-bore, head-specific, 0.5T MRI system.</p><p><strong>Methods: </strong>Assessment of spatial fidelity was performed using a 3D gradient recalled echo (GRE) acquisition on a commercial distortion phantom using the validated distortion analysis software provided. B<sub>0</sub>-induced distortions were measured using a 3D field map. Inter-scanner reproducibility was assessed across four distinct systems of identical make and model, while intra-scanner repeatability was assessed at one site over six repeat measurements.</p><p><strong>Results: </strong>Inter-scanner reproducibility measured an average 95th percentile distortion over 100 and 180 mm DSV of 0.15 ± 0.03  and 0.33 ± 0.05 mm. Average 95th percentile distortions due to B<sub>0</sub> field inhomogeneities over 100 and 180 mm DSV were 0.02 ± 0.01  and 0.07 ± 0.02 mm. Intra-scanner repeatability measured the uncertainty in distortion values to be 0.020 ± 0.005 mm.</p><p><strong>Conclusion: </strong>The total residual distortions measured in this phantom study were less than half the recommended value required for radiosurgery and significantly better than data published from other MR systems. This demonstrates that in addition to the compact footprint of the Synaptive 0.5T scanner, it exceeds current standards for geometric accuracy.</p>","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143744673","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}
引用次数: 0
Postal dosimetry audit for scanning proton beam using radiophotoluminescence glass dosimeter: A multicenter pilot study.
Medical physics Pub Date : 2025-03-29 DOI: 10.1002/mp.17790
Keisuke Yasui, Miuna Hayashi, Shiryu Otsuka, Toshiyuki Toshito, Chihiro Omachi, Masaya Ichihara, Riki Oshika, Yuki Tominaga, Hiromi Baba, Hidetoshi Shimizu, Naoki Hayashi
{"title":"Postal dosimetry audit for scanning proton beam using radiophotoluminescence glass dosimeter: A multicenter pilot study.","authors":"Keisuke Yasui, Miuna Hayashi, Shiryu Otsuka, Toshiyuki Toshito, Chihiro Omachi, Masaya Ichihara, Riki Oshika, Yuki Tominaga, Hiromi Baba, Hidetoshi Shimizu, Naoki Hayashi","doi":"10.1002/mp.17790","DOIUrl":"https://doi.org/10.1002/mp.17790","url":null,"abstract":"<p><strong>Background: </strong>Accurate dosimetry is important in radiotherapy, and all equipment used for radiotherapy shoud be audited by an independent external dose audit. Radiophotoluminescence glass dosimeter (RPLD) has excellent characteristics and is widely used for postal dose audit; however, postal dose audit for proton therapy using RPLD has not been established.</p><p><strong>Purpose: </strong>This study aims to develop a postal dose audit procedure for scanning proton beams using RPLD, estimate uncertainties, and conduct a multicenter pilot study to validate the methodology.</p><p><strong>Methods: </strong>A postal toolkit was developed and a postal dose audit procedure for RPLD measurements of scanning proton beams was established in cooperation with several facilities that employ various accelerators, irradiation equipment, and treatment planning systems (TPS) for clinical use. Based on basic and previous studies, an uncertainty budget was developed for estimating relative uncertainty and pilot studies were conducted at each site. A method for postal dose audits was developed in a multicenter collaboration to develop an approach suitable for implementation across multiple facilities.</p><p><strong>Results: </strong>The relative response of 60 RPLDs for scanning proton beam examined in this study was 1.00 ± 1.28% mean ± standard deviation. The combined relative standard uncertainty of postal dosimetry for scanning proton beams using the RPLD was 2.97% (k = 1). Under the reference condition, the maximum differences between the ionization chamber measurement (IC) and TPS, RPLD and TPS, and RPLD and IC were 0.97, 1.88, and 2.12%, respectively. The maximum differences between the RPLD and ionization chamber for plateau measurements at 3 cm depth using single-energy and non-reference conditions were 11.31 and 4.02%, respectively.</p><p><strong>Conclusion: </strong>We established a procedure for the postal dose audits of proton beams using RPLD and presented the results of a multicenter pilot study. By standardizing the reference conditions, the dosimetry uncertainty was estimated at 2.92%. The results demonstrated the feasibility of performing an independent third-party dose audit of scanning proton beams using RPLD, and for such postal dose audits for proton beams, the irradiation conditions should be standardized to reduce uncertainties. These results are expected to contribute to the development of proton beams.</p>","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143744674","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}
引用次数: 0
A segment anything model-guided and match-based semi-supervised segmentation framework for medical imaging.
Medical physics Pub Date : 2025-03-29 DOI: 10.1002/mp.17785
Guoping Xu, Xiaoxue Qian, Hua-Chieh Shao, Jax Luo, Weiguo Lu, You Zhang
{"title":"A segment anything model-guided and match-based semi-supervised segmentation framework for medical imaging.","authors":"Guoping Xu, Xiaoxue Qian, Hua-Chieh Shao, Jax Luo, Weiguo Lu, You Zhang","doi":"10.1002/mp.17785","DOIUrl":"https://doi.org/10.1002/mp.17785","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Semi-supervised segmentation leverages sparse annotation information to learn rich representations from combined labeled and label-less data for segmentation tasks. The Match-based framework, by using the consistency constraint of segmentation results from different models/augmented label-less inputs, is found effective in semi-supervised learning. This approach, however, is challenged by the low quality of pseudo-labels generated as intermediate products for training the network, due to the lack of the ''ground-truth'' reference.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;This study aims to leverage the foundation model, segment anything model (SAM), to assist unsupervised learning of Match-based frameworks. Trained with an extremely large dataset, SAM-based methods generalize better than traditional models to various imaging domains, allow it to serve as an assistant to Match-based frameworks to improve the quality of intermediate pseudo-labels for semi-supervised learning.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We propose SAM-Match, a SAM-guided and Match-based framework for semi-supervised medical image segmentation. Our approach involves two main steps: First, we use pretrained Match-based models to extract high-confidence predictions for prompt generation. Second, these prompts and unlabeled images are input into a fine-tuned SAM-based method to produce high-quality masks as pseudo-labels. And the refined pseudo-labels are further fed back to train the Match-based framework. SAM-Match can be trained in an end-to-end manner, facilitating interactions between the SAM- and Match-based models.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;SAM-Match demonstrates robust performance across multiple medical imaging datasets, including the ACDC cardiac MRI dataset, the BUSI breast ultrasound dataset, and an in-house liver MRI dataset (MRLiver). We partitioned the datasets into training, validation, and test sets (70%, 10%, and 20% for ACDC; 60%, 9%, and 31% for BUSI; and 62%, 12%, and 25% for MRLiver). On ACDC, with only 3 labeled cases, we achieved a Dice score of 89.36% ± 0.06% on 20 test cases. For BUSI, using just 30 labeled samples for training, we attained a Dice score of 59.35% ± 0.12% on 170 test samples. On MRLiver, training with only 3 labeled cases resulted in a Dice score of 80.04% ± 0.11% on 12 test scans. Wilcoxon signed-rank tests with Bonferroni corrections between the SAM-Match framework and the other comparison methods further demonstrated the statistical significance of SAM-Match's improvement in segmentation accuracy.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;Our SAM-Match framework shows promising results in semi-supervised semantic segmentation, effectively tackling the challenges of automatic prompt generation for SAM and high-quality pseudo-label generation for Match-based models. It can help accelerate the adoption of semi-supervised learning in segmentation tasks, particularly in data-scarce scenarios. Our data and code will be m","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143744672","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}
引用次数: 0
3D segmentation combining spatial and multi-scale features for intracranial aneurysm. 结合空间和多尺度特征的颅内动脉瘤三维分割。
Medical physics Pub Date : 2025-03-28 DOI: 10.1002/mp.17783
Xinfeng Zhang, Jie Shao, Xiangsheng Li, Xiaomin Liu, Hui Li, Maoshen Jia
{"title":"3D segmentation combining spatial and multi-scale features for intracranial aneurysm.","authors":"Xinfeng Zhang, Jie Shao, Xiangsheng Li, Xiaomin Liu, Hui Li, Maoshen Jia","doi":"10.1002/mp.17783","DOIUrl":"https://doi.org/10.1002/mp.17783","url":null,"abstract":"<p><strong>Background: </strong>Traditionally, the diagnosis of intracranial aneurysms has relied on the experience of the doctor in assessing the scanning results of radiological imaging technology, which is subjective and inefficient, and it is also limited by the physical strength and energy of the doctor.</p><p><strong>Purpose: </strong>In order to improve the diagnostic efficiency of doctors and reduce the rate of misdiagnosis and missed diagnosis as much as possible.</p><p><strong>Methods: </strong>We propose a 3D segmentation network, SMNet, based on the U-Net architecture that combines spatial and multi-scale features. The network can better solve the problem of intracranial aneurysm segmentation on magnetic resonance angiography (MRA) scanning sequences. Specifically, semantic information of different dimensions is extracted at each stage of the encoder by the multi-scale feature extraction block (MSE Block) and the strip volumetric pooling block (SVP Block), respectively. Then, after the fusion of adjacent scale features extracted by the decoder, the weight of features is further redistributed by the quaternary spatial attention block (QSA Block). While focusing on the important features, the ability to discriminate different foregrounds is improved.</p><p><strong>Results: </strong>Experiments show that the proposed three modules improve the segmentation performance to different degrees. Dice and MIoU have increased by 16.7% and 28% compared to the baseline in the private dataset, and the results of the Aneurysm Detection And segMentation (ADAM) public dataset are 0.482 and 0.389, respectively. It has shown better segmentation quality than 3D medical image segmentation mainstream models.</p><p><strong>Conclusion: </strong>Our model greatly improves the segmentation results of intracranial aneurysms with MRA images, and makes a certain contribution to the clinical intervention of computer-assisted diagnosis and treatment in this field.</p>","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143744671","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}
引用次数: 0
Evaluation of novel detector readout modes for MV scatter reduction on kV images.
Medical physics Pub Date : 2025-03-27 DOI: 10.1002/mp.17776
Ha Nguyen, Mathias Lehmann, Daniel Morf, Jason Luce, Andrew Keeler, Hyejoo Kang, John C Roeske
{"title":"Evaluation of novel detector readout modes for MV scatter reduction on kV images.","authors":"Ha Nguyen, Mathias Lehmann, Daniel Morf, Jason Luce, Andrew Keeler, Hyejoo Kang, John C Roeske","doi":"10.1002/mp.17776","DOIUrl":"https://doi.org/10.1002/mp.17776","url":null,"abstract":"<p><strong>Background: </strong>There is increased interest in concurrent kilovoltage (kV) imaging during megavoltage (MV) irradiation as a means of monitoring intra-fraction motion. However, scatter introduced by the MV beam degrades kV image quality, potentially making tumor visualization more challenging.</p><p><strong>Purpose: </strong>To implement novel imager readout modes on the on-board imager (OBI) of a Varian TrueBeam to reduce the effect of MV scatter on the kV detector.</p><p><strong>Methods: </strong>New readout strategies (vertical 3 × 2 binning, vertical region of interest, kV pulse drop functionality) were incorporated to achieve high frame rates. A series of 120 kV images were acquired on the OBI while simultaneously irradiating a phantom with an MV beam (6xFFF and 10xFFF) with field sizes ranging from 2 × 2 cm<sup>2</sup> to 6 × 6 cm<sup>2</sup> and a 10 × 10 cm<sup>2</sup> field for reference. Two datasets, with and without MV beam delivery, were collected at four frame rates varying from 7 to 45 fps, at four gantry angles. The effect of the frame rates was evaluated by the percent change with respect to the kV-only data within selected regions on the images. Image quality was evaluated using contrast-to-noise ratio (CNR).</p><p><strong>Results: </strong>The amount of MV scatter was directly proportional to the square field size and the detector integration time. At their respective maximum dose rates, 6xFFF and 10xFFF produced comparable amounts of MV scatter for considered frame rates and field sizes. Increasing the frame rates from 7 to 45 fps decreased MV scatter by up to 87%. Over the same frame rates, the CNR was improved by 20% for the 2 × 2 cm<sup>2</sup> field up to 85% for the 10 × 10 cm<sup>2</sup> field. These findings were consistent for both energies considered.</p><p><strong>Conclusions: </strong>MV scatter onto the kV detector can be significantly reduced with the implementation of the new detector readout modes. These strategies increase the CNR and thus improve the image quality.</p>","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143723046","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}
引用次数: 0
Accuracy of a helium-beam radiography system based on thin pixel detectors for an anthropomorphic head phantom.
Medical physics Pub Date : 2025-03-26 DOI: 10.1002/mp.17786
Margareta Metzner, Friderike K Longarino, Benjamin Ackermann, Annika Schlechter, Maike Saphörster, Yanting Xu, Julian Schlecker, Patrick Wohlfahrt, Christian Richter, Stephan Brons, Jürgen Debus, Oliver Jäkel, Mária Martišíková, Tim Gehrke
{"title":"Accuracy of a helium-beam radiography system based on thin pixel detectors for an anthropomorphic head phantom.","authors":"Margareta Metzner, Friderike K Longarino, Benjamin Ackermann, Annika Schlechter, Maike Saphörster, Yanting Xu, Julian Schlecker, Patrick Wohlfahrt, Christian Richter, Stephan Brons, Jürgen Debus, Oliver Jäkel, Mária Martišíková, Tim Gehrke","doi":"10.1002/mp.17786","DOIUrl":"https://doi.org/10.1002/mp.17786","url":null,"abstract":"<p><strong>Background: </strong>Ion-beam radiography is a promising technique to verify the range of ion-beam radiotherapy treatments regularly. To detect and quantify the water-equivalent thickness (WET) of potential anatomical changes, ion-beam radiographs must provide a sufficient WET accuracy on the level of 1%.</p><p><strong>Purpose: </strong>In this work, we show an energy-painted helium-beam radiograph of an anthropomorphic head phantom acquired with thin silicon pixel detectors for the first time. Furthermore, we determine the WET accuracy of our helium-beam radiography system for the especially heterogeneous skull base region, which is highly relevant for the treatment of head and neck and skull base tumors.</p><p><strong>Methods: </strong>With a detection system based on pixelated semiconducting Timepix detectors, we track single ions upstream and downstream of the head phantom. Furthermore, we measure their energy deposition in a thin Timepix detector behind the anthropomorphic phantom. To ensure a high precision of the image, we acquired a radiograph by using helium beams with five initial energies between 146.84 and 188.07 MeV/u following the energy painting algorithm. With a Siemens SOMATOM Confidence CT scanner, a single- and dual-energy CT were acquired with clinical protocols and translated to relative stopping power (RSP) values. After projecting these scans, the resulting WET maps were compared to the helium-beam radiograph. To evaluate the accuracy of all three modalities, a reference data set based on range-pullback measurements and a segmentation of a high-resolution CT scan was taken into account.</p><p><strong>Results: </strong>The mean absolute percentage error (MAPE) of all modalities was determined to be between 0.95% and 1.16%. Also, the root-mean-square percentage error (RMSPE) was similar for all modalities ranging from 1.19% to 1.46%. These deviations from the reference scan were found to mainly stem from an overestimation of air and sinus tissue and underestimation of cortical bone.</p><p><strong>Conclusions: </strong>The helium-beam radiograph was shown to achieve a WET accuracy competitive with that of clinically used imaging methods. If certain technical aspects are addressed, helium-beam radiography may emerge as an auspicious imaging modality for on-couch range verification of ion-beam radiotherapy treatments allowing for regular detection and quantification of anatomical changes.</p>","PeriodicalId":94136,"journal":{"name":"Medical physics","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143723033","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}
引用次数: 0
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信