Journal of Medical Imaging and Radiation Sciences最新文献

{"title":"Dosimetric comparison in treatment of nasopharyngeal carcinoma (NPC) using two different dose calculation algorithms (AAA & AXB) in Ring-based LINAC and C-arm LINAC","authors":"Chun Ho Cheung, Chung Kee Cheong, Ching Chiang, Chi Kiu Chan, Charlotte Cheung, Ming Chun Ricky Chau","doi":"10.1016/j.jmir.2026.102530","DOIUrl":"10.1016/j.jmir.2026.102530","url":null,"abstract":"<div><h3>Purpose</h3><div>The purpose of this retrospective study is to compare the dosimetric performance of c-arm and ring-based accelerators in conjunction with two dose calculation algorithms. The primary objective is to identify the optimal combination of accelerators and dose calculation algorithms for precise and effective treatment delivery in Nasopharyngeal carcinoma (NPC) treatment.</div></div><div>The purpose of this retrospective study is to compare the dosimetric performance of c-arm and ring-based accelerators in conjunction with two dose calculation algorithms. The primary objective is to identify the optimal combination of accelerators and dose calculation algorithms for precise and effective treatment delivery in Nasopharyngeal carcinoma (NPC) treatment.</div><div><h3>Method</h3><div>Twenty NPC patients treated at Kiang Wu Hospital were recruited. Target prescriptions and delineations were based on the latest IG-2024 guidelines, while Organ at Risk (OAR) were contoured according to RTOG-0225, RTOG-0615, and relevant protocols. For each patient, four Volumetric Modulated Arc Therapy (VMAT) plans were generated using three full arcs: Trilogy-AAA, Trilogy-AXB, Halcyon-AAA, and Halcyon-AXB. Dosimetric evaluation was based on target coverage, homogeneity index (HI), conformity index (CI), and various dose parameters of OARs. Delivery efficiency was assessed based on the total Monitor Units (MU), dose calculation time, and the estimated beam-on time. Two-way ANOVA was used for statistical analysis.</div></div><div>Twenty NPC patients treated at Kiang Wu Hospital were recruited. Target prescriptions and delineations were based on the latest IG-2024 guidelines, while Organ at Risk (OAR) were contoured according to RTOG-0225, RTOG-0615, and relevant protocols. For each patient, four Volumetric Modulated Arc Therapy (VMAT) plans were generated using three full arcs: Trilogy-AAA, Trilogy-AXB, Halcyon-AAA, and Halcyon-AXB. Dosimetric evaluation was based on target coverage, homogeneity index (HI), conformity index (CI), and various dose parameters of OARs. Delivery efficiency was assessed based on the total Monitor Units (MU), dose calculation time, and the estimated beam-on time. Two-way ANOVA was used for statistical analysis.</div><div><h3>Results</h3><div>All treatment plans had met the acceptance criteria in terms of OARs sparing and targets coverage. There was no statistically significant difference in terms of OARs sparing, targets coverage, HI, and CI. AXB was expected to yield the most favourable outcomes in dose prediction accuracy, while GPU-driven AXB achieved the shortest time in dose calculation. However, a higher efficiency in treatment delivery was anticipated when using a ring-based linear accelerator.</div></div><div>All treatment plans had met the acceptance criteria in terms of OARs sparing and targets coverage. There was no statistically significant difference in terms of OARs sparing, targets coverage, HI, and CI. AXB w","PeriodicalId":46420,"journal":{"name":"Journal of Medical Imaging and Radiation Sciences","volume":"57 5","pages":"Article 102530"},"PeriodicalIF":2.3,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149003687","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":"4D Monte Carlo simulation of proton therapy: Effects of tumor motion and tissue inhomogeneity","authors":"Taitacha Thongkam, Isara Israngkul Na Ayuthaya, Mintra Keawsamur","doi":"10.1016/j.jmir.2026.102526","DOIUrl":"10.1016/j.jmir.2026.102526","url":null,"abstract":"<div><h3>Introduction</h3><div>Proton therapy relies on the Bragg peak to deliver the dose at a defined depth. In lung cancer treatment, tissue inhomogeneity and respiratory motion are major challenges. Inhomogeneous tissue can cause proton scattering and dose uncertainty, while breathing motion may produce interplay effects, causing non-uniform dose distribution. Repainting is one strategy to improve target dose coverage. This study aimed to determine the optimal number of repaintings required to mitigate these effects in a CT-derived thorax phantom using 4D Monte Carlo simulation.</div></div><div>Proton therapy relies on the Bragg peak to deliver the dose at a defined depth. In lung cancer treatment, tissue inhomogeneity and respiratory motion are major challenges. Inhomogeneous tissue can cause proton scattering and dose uncertainty, while breathing motion may produce interplay effects, causing non-uniform dose distribution. Repainting is one strategy to improve target dose coverage. This study aimed to determine the optimal number of repaintings required to mitigate these effects in a CT-derived thorax phantom using 4D Monte Carlo simulation.</div><div><h3>Methods</h3><div>Monoenergetic proton pencil beams were simulated in a water phantom using GEANT4 and compared with commissioning data for nominal energies of 70–220 MeV. After beam model verification, CT images of a dynamic thorax phantom were converted into voxelized GEANT4 geometry for dose calculation. Respiratory tumor motion was then incorporated, and proton beams were delivered with a prescribed reference RBE dose of 200 cGy under varying target diameters (1, 2, and 3 cm), motion amplitudes (0.5, 1.0, 1.5, and 2.0 cm), and breathing patterns (symmetric and asymmetric). The number of repaintings (1–4) was varied to improve dose coverage. Dosimetric effects were evaluated using 2D gamma analysis (3%/2 mm, 10% dose threshold) and 1D dose profiles.</div></div><div>Monoenergetic proton pencil beams were simulated in a water phantom using GEANT4 and compared with commissioning data for nominal energies of 70–220 MeV. After beam model verification, CT images of a dynamic thorax phantom were converted into voxelized GEANT4 geometry for dose calculation. Respiratory tumor motion was then incorporated, and proton beams were delivered with a prescribed reference RBE dose of 200 cGy under varying target diameters (1, 2, and 3 cm), motion amplitudes (0.5, 1.0, 1.5, and 2.0 cm), and breathing patterns (symmetric and asymmetric). The number of repaintings (1–4) was varied to improve dose coverage. Dosimetric effects were evaluated using 2D gamma analysis (3%/2 mm, 10% dose threshold) and 1D dose profiles.</div><div><h3>Results</h3><div>In the static tumor case, GEANT4 and TPS showed good agreement, achieving a 2D gamma passing rate of 97.6% and a maximum 1D dose profile deviation of 3.8%. With respiratory tumor motion, the large tumor maintained better dose coverage than the small tumor under the s","PeriodicalId":46420,"journal":{"name":"Journal of Medical Imaging and Radiation Sciences","volume":"57 5","pages":"Article 102526"},"PeriodicalIF":2.3,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149004099","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":"Lost in Transition: Improving IV Contrast Safety in CT Simulation Through Epic Hyperspace‑Enabled Workflow Design","authors":"Xin Min Lee, Weiguang Guo, Ye Lin Chiau, Chiou Yen Koh","doi":"10.1016/j.jmir.2026.102536","DOIUrl":"10.1016/j.jmir.2026.102536","url":null,"abstract":"<div><h3>Background</h3><div>Singapore’s public healthcare system is progressively adopting the Next Generation Electronic Medical Record, powered by Epic Hyperspace, to consolidate and standardize clinical documentation across public healthcare clusters. Within Radiation Oncology, gaps in the documentation of intravenous contrast (IVC) safety checks became evident. Although contraindication screening, patient counselling, and pre‑injection checks were routinely performed by nursing and Radiation Therapy (RT) staff, documentation was inconsistent. As a result, safety‑critical information was often communicated verbally, creating an unowned clinical boundary and increasing the risk of missed contraindications during contrast‑enhanced CT simulation. Epic onboarding created an opportunity to redesign workflows to improve ownership, clarity, and visibility across teams.</div></div><div>Singapore’s public healthcare system is progressively adopting the Next Generation Electronic Medical Record, powered by Epic Hyperspace, to consolidate and standardize clinical documentation across public healthcare clusters. Within Radiation Oncology, gaps in the documentation of intravenous contrast (IVC) safety checks became evident. Although contraindication screening, patient counselling, and pre‑injection checks were routinely performed by nursing and Radiation Therapy (RT) staff, documentation was inconsistent. As a result, safety‑critical information was often communicated verbally, creating an unowned clinical boundary and increasing the risk of missed contraindications during contrast‑enhanced CT simulation. Epic onboarding created an opportunity to redesign workflows to improve ownership, clarity, and visibility across teams.</div><div><h3>Methods</h3><div>Institutional CT IVC contraindication guidelines were translated into a structured SmartPhrase embedded within the AVATAR assessment tool in Epic. AVATAR, typically used to document invasive devices such as IVC, was configured to prompt standardized screening at three care points: (1) nurse counselling and risk assessment performed 2–4 weeks before CT simulation, (2) pre‑cannulation nursing checks and (3) RT briefing prior to CT simulation. Documentation from all stages was consolidated into a single shared AVATAR assessment note to enhance continuity, transparency and cross‑disciplinary communication. The redesigned workflow was implemented at National University Hospital and subsequently proposed for extension to Tan Tock Seng Hospital as part of digital harmonisation efforts within the National Cancer Institute Singapore.</div></div><div>Institutional CT IVC contraindication guidelines were translated into a structured SmartPhrase embedded within the AVATAR assessment tool in Epic. AVATAR, typically used to document invasive devices such as IVC, was configured to prompt standardized screening at three care points: (1) nurse counselling and risk assessment performed 2–4 weeks before CT simulation, (2) p","PeriodicalId":46420,"journal":{"name":"Journal of Medical Imaging and Radiation Sciences","volume":"57 5","pages":"Article 102536"},"PeriodicalIF":2.3,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149004274","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":"Feasibility study for delivering breast VMAT radiation treatment with Simultaneous Integrated Boost (SIB) for early left-sided breast cancer following the Fast Forward (FF) protocol","authors":"Syazana Rashid, Tan Lloyd, Ng Yan Yee, Ho Bryan","doi":"10.1016/j.jmir.2026.102504","DOIUrl":"10.1016/j.jmir.2026.102504","url":null,"abstract":"<div><h3>Background and Purpose</h3><div>Adjuvant radiation therapy (RT) following primary surgery in women affected by early breast cancer (BC) plays a significant role in reducing local recurrences and overall mortality. The FAST-Forward (FF) trial demonstrated that 5# of hypofractionated RT is non-inferior to the standard 15# in terms of local relapse, cosmetic outcomes and toxicity. Additional radiation boost dose to the lumpectomy site has also been a standard of care to reduce local recurrences. Delivering the boost simultaneously with the Whole Breast (WB) will result in a shorter overall treatment duration for patients. Hence, the aim of this study is to establish the feasibility of using VMAT with Simultaneous Integrated Boost (SIB) for FF dose fractionation scheme in treating early Left Sided BC.</div></div><div>Adjuvant radiation therapy (RT) following primary surgery in women affected by early breast cancer (BC) plays a significant role in reducing local recurrences and overall mortality. The FAST-Forward (FF) trial demonstrated that 5# of hypofractionated RT is non-inferior to the standard 15# in terms of local relapse, cosmetic outcomes and toxicity. Additional radiation boost dose to the lumpectomy site has also been a standard of care to reduce local recurrences. Delivering the boost simultaneously with the Whole Breast (WB) will result in a shorter overall treatment duration for patients. Hence, the aim of this study is to establish the feasibility of using VMAT with Simultaneous Integrated Boost (SIB) for FF dose fractionation scheme in treating early Left Sided BC.</div><div><h3>Methods</h3><div>5 Patients affected by early left-sided BC scheduled to undergo adjuvant 3D-conformal radiation therapy (3D-CRT) to the WB with sequential boost were selected. A dose of 26Gy in 5# and a sequential boost of 10.5Gy in 3# was delivered for each patient. A VMAT plan with SIB was retrospectively generated for each patient. The total dose planned for was 30Gy in 5#(EQD2:40Gy) to the PTV_Boost and 26Gy in 5# to the PTV_WB. The planning outcomes of both plans were then compared against the planning constraints provided by the FF Planning Guidelines.</div></div><div>5 Patients affected by early left-sided BC scheduled to undergo adjuvant 3D-conformal radiation therapy (3D-CRT) to the WB with sequential boost were selected. A dose of 26Gy in 5# and a sequential boost of 10.5Gy in 3# was delivered for each patient. A VMAT plan with SIB was retrospectively generated for each patient. The total dose planned for was 30Gy in 5#(EQD2:40Gy) to the PTV_Boost and 26Gy in 5# to the PTV_WB. The planning outcomes of both plans were then compared against the planning constraints provided by the FF Planning Guidelines.</div><div><h3>Results</h3><div>A total of 19 outcomes were measured and of which 6 proved to be significant (p <0.05). VMAT provided better target coverage and homogeneity of PTV_WB with V95% = (95.7 ±) whereas 3D-CRT showed significantly lo","PeriodicalId":46420,"journal":{"name":"Journal of Medical Imaging and Radiation Sciences","volume":"57 5","pages":"Article 102504"},"PeriodicalIF":2.3,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149004364","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":"Does the application of Pharyngeal Constrictor Muscle (PCM) as an Organ at Risk (OAR) in tomotherapy planning for Nasopharyngeal Carcinoma (NPC) patients reduce the risk of dysphagia while maintaining sufficient target dose?","authors":"Yan Yee Ho, George Chiu, Joanna McNamara","doi":"10.1016/j.jmir.2026.102510","DOIUrl":"10.1016/j.jmir.2026.102510","url":null,"abstract":"<div><h3>Background</h3><div>Dysphagia is common for external beam radiotherapy (EBRT) of nasopharyngeal carcinoma (NPC) patients, and pharyngeal constrictor muscle (PCM) is an important swallowing-related structure. Although dosimetric constraints for volumetric modulated arc therapy (VMAT) are present, Hong Kong private hospitals are not applying them for Tomotherapy plan optimization. Therefore, this study aims at investigating whether contouring of PCM as organ at risk (OAR) during Tomotherapy planning can reduce its dose based on published dose constraints, while achieving target dose.</div></div><div>Dysphagia is common for external beam radiotherapy (EBRT) of nasopharyngeal carcinoma (NPC) patients, and pharyngeal constrictor muscle (PCM) is an important swallowing-related structure. Although dosimetric constraints for volumetric modulated arc therapy (VMAT) are present, Hong Kong private hospitals are not applying them for Tomotherapy plan optimization. Therefore, this study aims at investigating whether contouring of PCM as organ at risk (OAR) during Tomotherapy planning can reduce its dose based on published dose constraints, while achieving target dose.</div><div><h3>Methods</h3><div>This retrospective quantitative study involved an experimental group with PCM optimization and a comparison group without PCM optimization. Data was obtained from all NPC patients (26 patients) treated at Hong Kong Sanatorium & Hospital during 2021 to 2023 by an in-house oncologist. PCM was contoured, and its importance was set to be the same as larynx. Tomotherapy plan was then optimized with all other constraints unchanged. Other than doses to OAR, dosimetric parameters were analyzed, including maximum and minimum doses of the targets, the percentage of target volume receiving at least 95% of the prescribed dose TV95(%), homogeneity index (HI), and conformity index (CI).</div></div><div>This retrospective quantitative study involved an experimental group with PCM optimization and a comparison group without PCM optimization. Data was obtained from all NPC patients (26 patients) treated at Hong Kong Sanatorium & Hospital during 2021 to 2023 by an in-house oncologist. PCM was contoured, and its importance was set to be the same as larynx. Tomotherapy plan was then optimized with all other constraints unchanged. Other than doses to OAR, dosimetric parameters were analyzed, including maximum and minimum doses of the targets, the percentage of target volume receiving at least 95% of the prescribed dose TV95(%), homogeneity index (HI), and conformity index (CI).</div><div><h3>Results/Conclusion</h3><div>With the PCM contour, its dose met the requirement for minimizing dysphagia, while keeping similar nearby OAR dose. The outcomes also suggested that target clinical objectives were met when PCM was optimized. However, PCM plans generally had a higher maximum target dose, a lower minimum target dose, a higher HI and a higher CI, with p-value < 0.05. Th","PeriodicalId":46420,"journal":{"name":"Journal of Medical Imaging and Radiation Sciences","volume":"57 5","pages":"Article 102510"},"PeriodicalIF":2.3,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149004043","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":"Multi-omics Integration for Prognostication in Glioblastoma: A Comparative Analysis of MRI-based Radiomics and Multi-omics Models","authors":"Tsz Lung Wong, Hiu Yan Yeung, Sum Long Chan, Yi Yui Li, Chi Chung Siu","doi":"10.1016/j.jmir.2026.102520","DOIUrl":"10.1016/j.jmir.2026.102520","url":null,"abstract":"<div><div>Magnetic Resonance Imaging (MRI) is widely used in diagnosing Glioblastoma Multiforme (GBM) and patient survival prediction. Studies have used MRI radiomics to investigate survival of GBM patients, but they only relied on radiomics features which may hinder the accuracy of GBM prognosis prediction due to the ignorance of genomic and other clinical factors. As a result, it would lead to inaccurate prognostic prediction. Multi-omics offer a comprehensive view of the complicated GBM tumour status and survival outcomes because the genomic data and clinical factors can also be integrated with radiomics in the predictive model to improve the sensitivity. In this study, several multi-omics models were developed and compared with radiomics models to investigate the 1-year GBM survival predictive power of multi-omics models. The models were built with the use of MRI radiomics, genomics and clinical data from the University of Pennsylvania Glioblastoma (UPenn-GBM) dataset. 630 patients were included and the dataset was retrieved from the Cancer Imaging Archive. Radiomics features were selected using the Maximum Relevance Minimum Redundancy (mRMR) and the Least Absolute Shrinkage and Selection Operator (LASSO). Clinical features were chosen with Linear Regression and LASSO. Support vector machine (SVM) and Random Forest (RF) were used for building 16 radiomics-alone and multi-omics models in different combinations. The predictive performances of the 16 models were evaluated by receiver operating characteristic (ROC) curve with the area under the curve (AUC). Among the 16 models, the SVM model that combined radiomics feature selected by LASSO and clinical data selected by linear regression had achieved the best AUC which was 0.9033. The model offers a more robust predictive tool for health professionals and improves GBM prognostication. Future research can also focus on developing models predicting with different MR sequence combinations or data types.</div></div><div>Magnetic Resonance Imaging (MRI) is widely used in diagnosing Glioblastoma Multiforme (GBM) and patient survival prediction. Studies have used MRI radiomics to investigate survival of GBM patients, but they only relied on radiomics features which may hinder the accuracy of GBM prognosis prediction due to the ignorance of genomic and other clinical factors. As a result, it would lead to inaccurate prognostic prediction. Multi-omics offer a comprehensive view of the complicated GBM tumour status and survival outcomes because the genomic data and clinical factors can also be integrated with radiomics in the predictive model to improve the sensitivity. In this study, several multi-omics models were developed and compared with radiomics models to investigate the 1-year GBM survival predictive power of multi-omics models. The models were built with the use of MRI radiomics, genomics and clinical data from the University of Pennsylvania Glioblastoma (UPenn-GBM) dataset. 630 patients were inc","PeriodicalId":46420,"journal":{"name":"Journal of Medical Imaging and Radiation Sciences","volume":"57 5","pages":"Article 102520"},"PeriodicalIF":2.3,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149004098","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":"Artificial Intelligence in ultrasound training and education: A comprehensive literature review of applications, effectiveness, and future directions","authors":"Ryan Chou, Yee Fong Leung, Edward Hung Tat Chan","doi":"10.1016/j.jmir.2026.102501","DOIUrl":"10.1016/j.jmir.2026.102501","url":null,"abstract":"<div><h3>Background / Purpose</h3><div>Ultrasound proficiency is highly operator-dependent and requires intensive, supervised practice, creating significant educational bottlenecks in radiology, radiation therapy, and point-of-care ultrasound training programs. Artificial intelligence (AI) offers real-time probe guidance, automated feedback, and objective competency assessment that may democratize ultrasound training and alleviate faculty workload. This review synthesizes current AI applications in ultrasound training, evaluates their educational effectiveness, and highlights implementation challenges and future research priorities.</div></div><div>Ultrasound proficiency is highly operator-dependent and requires intensive, supervised practice, creating significant educational bottlenecks in radiology, radiation therapy, and point-of-care ultrasound training programs. Artificial intelligence (AI) offers real-time probe guidance, automated feedback, and objective competency assessment that may democratize ultrasound training and alleviate faculty workload. This review synthesizes current AI applications in ultrasound training, evaluates their educational effectiveness, and highlights implementation challenges and future research priorities.</div><div><h3>Methods</h3><div>A comprehensive search of PubMed, Web of Science, Google Scholar, and SciSpace identified English-language studies published from January 2015 to November 2024 that applied AI, machine learning, or deep learning to ultrasound training, education, or simulation. After systematic screening, 113 studies were included, with 61 high-impact articles undergoing detailed synthesis based on methodological quality, relevance to training outcomes, and citation impact.</div></div><div>A comprehensive search of PubMed, Web of Science, Google Scholar, and SciSpace identified English-language studies published from January 2015 to November 2024 that applied AI, machine learning, or deep learning to ultrasound training, education, or simulation. After systematic screening, 113 studies were included, with 61 high-impact articles undergoing detailed synthesis based on methodological quality, relevance to training outcomes, and citation impact.</div><div><h3>Results</h3><div>Across cardiac, lung, obstetric, abdominal, renal, and interventional ultrasound, AI guidance enabled novices to achieve diagnostic-quality image acquisition rates of approximately 87–99%, approaching the performance of experienced operators. Representative randomized controlled trials demonstrated faster acquisition (for example, 72 s versus 85 s, p = 0.01), higher image quality scores, and improved diagnostic accuracy (85% versus 50%, p = 0.02) compared with conventional training. Automated quality assessment systems achieved approximately 94% agreement with expert evaluation and reduced the expert review workload by up to five times.</div></div><div>Across cardiac, lung, obstetric, abdominal, renal, and interventional ultrasou","PeriodicalId":46420,"journal":{"name":"Journal of Medical Imaging and Radiation Sciences","volume":"57 5","pages":"Article 102501"},"PeriodicalIF":2.3,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149004361","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":"Shaping the Future of Our Profession: A Survey of Local Radiation Therapists’ Perceptions of the APRT Role in Hong Kong","authors":"Victor Chi Wing Tam, Vincent Wan Shun Leung","doi":"10.1016/j.jmir.2026.102534","DOIUrl":"10.1016/j.jmir.2026.102534","url":null,"abstract":"<div><h3>Background</h3><div>The Advanced Practice Radiation Therapist (APRT) role was first developed in the United Kingdom to address increasing service demands and improve efficiency in radiotherapy delivery, and has since been adopted in Europe, Canada, Australia, and Singapore. Previous studies have shown that adopting the APRT model improved service workflows, expanded system capacity, and enhanced patient education without compromising service quality. It also improved radiation therapists’ job satisfaction and perceived recognition and respect as healthcare professionals. Given the increasing service needs for radiation therapy in Hong Kong (HK), the role of RT has been evolving with emerging professional development programs, such as the Advanced Specialty Program in the public healthcare system. However, implementation pathways, scope of practice, and professional readiness regarding APRT remain unclear. This study aims to examine local radiation therapists’ perceptions of the APRT role in HK.</div></div><div>The Advanced Practice Radiation Therapist (APRT) role was first developed in the United Kingdom to address increasing service demands and improve efficiency in radiotherapy delivery, and has since been adopted in Europe, Canada, Australia, and Singapore. Previous studies have shown that adopting the APRT model improved service workflows, expanded system capacity, and enhanced patient education without compromising service quality. It also improved radiation therapists’ job satisfaction and perceived recognition and respect as healthcare professionals. Given the increasing service needs for radiation therapy in Hong Kong (HK), the role of RT has been evolving with emerging professional development programs, such as the Advanced Specialty Program in the public healthcare system. However, implementation pathways, scope of practice, and professional readiness regarding APRT remain unclear. This study aims to examine local radiation therapists’ perceptions of the APRT role in HK.</div><div><h3>Methods</h3><div>A cross-sectional population survey was conducted by recruiting all registered therapeutic radiographers in HK (N = 579). An anonymous online survey was distributed via the internal communication channels of the Hong Kong Association of Radiation Therapists (HKART), the official union for local RTs. Inclusion criteria included registered radiation therapists currently practicing clinically or involved in radiation therapy tertiary education in HK.</div></div><div>A cross-sectional population survey was conducted by recruiting all registered therapeutic radiographers in HK (N = 579). An anonymous online survey was distributed via the internal communication channels of the Hong Kong Association of Radiation Therapists (HKART), the official union for local RTs. Inclusion criteria included registered radiation therapists currently practicing clinically or involved in radiation therapy tertiary education in HK.</div><div><h3>Results</h","PeriodicalId":46420,"journal":{"name":"Journal of Medical Imaging and Radiation Sciences","volume":"57 5","pages":"Article 102534"},"PeriodicalIF":2.3,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149004454","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":"Evaluation of an online platform for automated technical assessment in prostate MRI quality","authors":"Anson CM Chau","doi":"10.1016/j.jmir.2026.102518","DOIUrl":"10.1016/j.jmir.2026.102518","url":null,"abstract":"<div><h3>Background</h3><div>Multiparametric prostate magnetic resonance imaging (MRI) serves as the primary investigation when prostate cancer is suspected, with its ability to identify clinically significant disease relying on the acquisition of high-quality images. Achieving consistent high-quality imaging outcomes necessitates meticulous attention to various factors, including patient preparation, scanner setup, pulse sequence optimisation, etc. The Prostate Imaging Quality (PI-QUAL) scoring system has been instrumental in standardising the quantification of image quality and its association with the ability to detect prostate cancer through MRI imaging. In this presentation, we explore the application of an online platform designed to evaluate the technical criteria outlined in PI-QUAL v2. We report on the platform’s accuracy and efficiency in automatically extracting relevant technical parameters from prostate MRI images.</div></div><div>Multiparametric prostate magnetic resonance imaging (MRI) serves as the primary investigation when prostate cancer is suspected, with its ability to identify clinically significant disease relying on the acquisition of high-quality images. Achieving consistent high-quality imaging outcomes necessitates meticulous attention to various factors, including patient preparation, scanner setup, pulse sequence optimisation, etc. The Prostate Imaging Quality (PI-QUAL) scoring system has been instrumental in standardising the quantification of image quality and its association with the ability to detect prostate cancer through MRI imaging. In this presentation, we explore the application of an online platform designed to evaluate the technical criteria outlined in PI-QUAL v2. We report on the platform’s accuracy and efficiency in automatically extracting relevant technical parameters from prostate MRI images.</div><div><h3>Methods</h3><div>The assessment focused on key PI-QUAL v2 criteria, including T2-weighted imaging (T2), diffusion-weighted imaging (DWI), and dynamic contrast-enhanced imaging (DCE). A spherical phantom was utilized to evaluate the platform’s accuracy in detecting parameters such as slice thickness, b-values, image weighting, and imaging plane orientation in T2 and DWI sequences. Imaging was performed on a 3T MRI scanner with a body coil, with deliberate modifications to standard clinical sequences to assess detection robustness. Additionally, anonymized clinical images were used to evaluate the platform’s ability to identify slice thickness, temporal resolution, and the use of fat saturation in DCE images. All images, exported in DICOM format from the scanner or archive, were uploaded to the platform for analysis.</div></div><div>The assessment focused on key PI-QUAL v2 criteria, including T2-weighted imaging (T2), diffusion-weighted imaging (DWI), and dynamic contrast-enhanced imaging (DCE). A spherical phantom was utilized to evaluate the platform’s accuracy in detecting parameters such as slice","PeriodicalId":46420,"journal":{"name":"Journal of Medical Imaging and Radiation Sciences","volume":"57 5","pages":"Article 102518"},"PeriodicalIF":2.3,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149003925","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":"Special paediatric CT Imaging for cardiac patients in Hong Kong Children’s Hospital","authors":"Wing Tak Lai","doi":"10.1016/j.jmir.2026.102511","DOIUrl":"10.1016/j.jmir.2026.102511","url":null,"abstract":"<div><div>Hong Kong Children’s Hospital (HKCH) is the only children's hospital in Hong Kong that provides advanced pediatric CT imaging. HKCH offers a full range of specialized CT examinations dedicated to patients suffering from cardiac failure and diseases which encompass a wide range of pathologies and morphologies arising from congenital heart diseases. Therefore, the related techniques used in pediatric CT will differ significantly from those used for adult patients. For patients suffering from cardiac issues, with or without simultaneous lung failure, an extracorporeal oxygenation membrane (ECMO) will be used to support their lives. To perform CT examinations for these types of patients, additional procedures must be followed to achieve diagnostic results. Moreover, patients with a univentricular heart must undergo a Fontan operation to ensure adequate blood flow for survival. CT cardiac examination for Fontan patients poses another challenge, as their circulation varies significantly from other patients; therefore, special examination techniques are required to obtain satisfactory results.Blood flow for patients on ECMO differs significantly from that of normal patients. When contrast is injected during the examination, it is drawn in by the ECMO machine for oxygenation. The diluted contrast will be pumped back to the patient for systemic circulation. The enhancement of different regions of the patient will depend on the types of ECMO insertion as well as the rate of the suction force. As a result, a preliminary investigation of the types of ECMO and a throughout discussion with the Radiologist and referring doctors is crucial to the development of the examination protocols leading to a successful CT examination. The purpose of this presentation is to describe the types of ECMO and patient suffered from special cardiac diseases and the related CT imaging techniques used for different kind of examinations as requested by the Pediatricians.</div></div><div>Hong Kong Children’s Hospital (HKCH) is the only children's hospital in Hong Kong that provides advanced pediatric CT imaging. HKCH offers a full range of specialized CT examinations dedicated to patients suffering from cardiac failure and diseases which encompass a wide range of pathologies and morphologies arising from congenital heart diseases. Therefore, the related techniques used in pediatric CT will differ significantly from those used for adult patients. For patients suffering from cardiac issues, with or without simultaneous lung failure, an extracorporeal oxygenation membrane (ECMO) will be used to support their lives. To perform CT examinations for these types of patients, additional procedures must be followed to achieve diagnostic results. Moreover, patients with a univentricular heart must undergo a Fontan operation to ensure adequate blood flow for survival. CT cardiac examination for Fontan patients poses another challenge, as their circulation varies significantly from oth","PeriodicalId":46420,"journal":{"name":"Journal of Medical Imaging and Radiation Sciences","volume":"57 5","pages":"Article 102511"},"PeriodicalIF":2.3,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149003934","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}