Journal of Medical Physics最新文献

{"title":"Characterization and Validation of MicroDiamond Detectors in Patient-specific Quality Assurance of Magnetic Resonance-guided Stereotactic Radiotherapy.","authors":"Sanjana Choudhary, Chithra Krishnamoorthy, Radhakrishnan B Nair, S D Sharma, Vinayaka Shenoy","doi":"10.4103/jmp.jmp_123_26","DOIUrl":"10.4103/jmp.jmp_123_26","url":null,"abstract":"<p><strong>Introduction: </strong>A dosimetric performance analysis of microDiamond (MD) detector in magnetic resonance imaging linear accelerators (MR-Linac) was performed to evaluate its utility for patient-specific quality assurance (PSQA) in stereotactic radiotherapy (SRT).</p><p><strong>Methods: </strong>Output factors (OFs), depth dose curves, and dose profiles for field sizes from 1 cm × 1 cm to 10 cm × 10 cm were measured in an Elekta Unity 1.5 T MR-Linac. Measurements were conducted using an MR-Safe three-dimensional (3D) water phantom with a PTW Semiflex 3D ionization chamber and a PTW MD detector. Angular response was assessed for orientations perpendicular to the beam and parallel to the magnetic field; correction factors were determined for gantry angles from 0° to 360° in 20° increments. Detectors were cross-calibrated against a 0.6 cc Farmer chamber, and patient-specific point-dose verification was performed for SRT plans.</p><p><strong>Results: </strong>Average <i>d</i> _max measured with the MD detector (1.28 ± 0.1 cm) agreed closely with the Semiflex 3D (1.27 ± 0.1 cm). The Semiflex 3D overestimated penumbra by 1.8 ± 0.6 mm and underestimated small-field OFs, while the MD detector showed slight overresponse (0.5%). The angular dependence remains substantial and highly sensitive to field size, requiring field-size-specific angular correction factors to achieve clinical accuracy better than 2%. After cross-calibration, the difference between the measured and planned doses for the 1.2 cc target was - 6.92% for Semiflex 3D and 1.36% for MD detector.</p><p><strong>Conclusion: </strong>The MD detector was successfully validated for MR-Linac. Asymmetric angular dependence due to the Lorentz force was confirmed, necessitating gantry angle-specific corrections for dose estimation. The MD detector demonstrated superior accuracy for small-field SRT targets, confirming its significance for precise PSQA.</p>","PeriodicalId":51719,"journal":{"name":"Journal of Medical Physics","volume":"51 3","pages":"618-627"},"PeriodicalIF":1.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13675055/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217201","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":"Toward End-to-end Artificial Intelligence Systems for Radiographic Fracture Assessment: A Critical Review of Detection, Explainability, and Prognosis.","authors":"Arshi Sheikh, Shailendra Singh","doi":"10.4103/jmp.jmp_67_26","DOIUrl":"10.4103/jmp.jmp_67_26","url":null,"abstract":"<p><p>This review synthesizes recent advances in fracture intelligence through a unified seven-stage end-to-end pipeline that includes preprocessing and (region of interest) ROI extraction, fracture detection, classification, localization and segmentation, prognostic modeling, explainability, and clinical deployment. By shifting the focus from isolated algorithms to complete clinical workflows, we expose a critical gap: current systems excel at basic detection but struggle with complex prognostic modeling and real-world deployment. Within this framework, we critically analyze methodological progress, highlight enduring research challenges, and demonstrate that fracture artificial intelligence continues to be detection focused with relatively less advancement in quantitative fracture characterization, prognostic intelligence, intrinsically interpretable learning, and prospective clinical validation. Based on these findings, we outline future research directions toward unified, anatomically informed, clinically deployable fracture intelligence systems.</p>","PeriodicalId":51719,"journal":{"name":"Journal of Medical Physics","volume":"51 3","pages":"486-493"},"PeriodicalIF":1.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13675012/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217256","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":"Is there Any Impact of Tumor Volume in Stereotactic Body Radiation Therapy for Lung Cancer? A Cohort for T1, 2 Non-small Cell Lung Cancer Cases.","authors":"Maria T Dimitropoulou, Kalliopi Platoni, Eleftheria Alevronta, Vassilis Kouloulias, Christina Armpilia, Anna Zygogianni","doi":"10.4103/jmp.jmp_161_26","DOIUrl":"10.4103/jmp.jmp_161_26","url":null,"abstract":"<p><strong>Purpose/aim: </strong>The aim of the study was to assess the impact of tumor size on dose-volume histograms (DVHs) and clinical outcomes in patients with early-stage non-small cell lung cancer (NSCLC) undergoing stereotactic body radiation therapy (SBRT) using the CyberKnife M6 System with a lung-optimized treatment module.</p><p><strong>Materials and methods: </strong>Retrospective analysis was performed on 12 patients diagnosed with NSCLC receiving SBRT treatment. Analysis of cumulative DVHs was done to evaluate dose distribution in the planning target volume (PTV) and surrounding organs at risk. The patient population was divided according to tumor size; one group was patients with PTV <12 cc and the other group was patients with PTV >12 cc. Statistical analysis was performed using SPSS software.</p><p><strong>Results: </strong>All treatment plans achieved adequate target coverage with sharp dose gradients. There was slight improvement in terms of dose distribution and response rate for smaller tumors (PTV <12 cc); however, these differences did not show statistical significance.</p><p><strong>Conclusions: </strong>There were no statistically significant differences between tumor volume and DVH parameters or clinical response in SBRT treatment of NSCLC. Most of the patients received biological effective dose (α/β =10) values above 100 Gy, with some of them receiving >120 Gy, which has been associated with improved treatment outcomes.</p>","PeriodicalId":51719,"journal":{"name":"Journal of Medical Physics","volume":"51 3","pages":"713-718"},"PeriodicalIF":1.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13675047/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217258","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":"<i>In vivo</i> Dose Measurement in Cervical Cancer Brachytherapy Using Dysprosium-doped Calcium Sulfate Thermoluminescent Dosimeters: A Phantom Study.","authors":"Prabhat Krishna Sharma, Saral Kumar Gupta, Devesh Gupta","doi":"10.4103/jmp.jmp_107_26","DOIUrl":"10.4103/jmp.jmp_107_26","url":null,"abstract":"<p><strong>Aims: </strong>High-dose-rate (HDR) brachytherapy (BT) is an important part of the treatment of cervical cancer. Although there have been advances in three-dimensional image-guided treatment planning systems (TPSs), uncertainties in applicator positioning and anatomical variations can produce discrepancies in planned and delivered doses. This study aimed to evaluate the clinical feasibility and reliability of dysprosium-doped calcium sulfate (CaSO₄:Dy) Teflon thermoluminescent dosimeters (TLDs) for independent dose verification in Co-60 HDR cervical BT.</p><p><strong>Materials and methods: </strong>For calibration, the CaSO₄:Dy Teflon TLD discs were held at a 5-cm depth in a phantom to closely mimic the conditions of dose measurements in a phantom. We specially constructed a phantom compatible with vaginal cylinders and Fletcher-Suit applicators to place TLDs at clinically significant sites. The measured doses were compared with TPS-calculated values by uncertainty analysis and statistical evaluation.</p><p><strong>Results: </strong>In-depth phantom measurements demonstrated that, for the same delivered dose, the response of CaSO₄:Dy TLDs at 1-cm and 5-cm depth in the phantom differed by only about 4%. The TLD system demonstrated a combined uncertainty of 11.80% (<i>c</i> = 2), with absolute mean deviation ranging from 6% to 9% across all measurement points. Bland-Altman analysis demonstrates that more than 95% of the paired observations were distributed within the limit.</p><p><strong>Conclusion: </strong>The results confirm that with proper calibration, CaSO₄:Dy Teflon TLD discs could also be a useful and trustworthy tool for independent dose verification in HDR BT. This approach can enhance routine quality assurance practices and improve confidence in dose delivery accuracy.</p>","PeriodicalId":51719,"journal":{"name":"Journal of Medical Physics","volume":"51 3","pages":"653-661"},"PeriodicalIF":1.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13675027/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217275","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":"Automated Radiotherapy Planning for Head-and-neck Cancer: Multi-institutional Development, Validation, and Clinical Implementation of Monaco Scripting.","authors":"Pichandi Anchineyan, Yuvarajan Sankaran, H Mahedran, Sam Skennet Das","doi":"10.4103/jmp.jmp_47_26","DOIUrl":"10.4103/jmp.jmp_47_26","url":null,"abstract":"<p><strong>Purpose: </strong>The purpose of this study was to develop, validate, and clinically implement a standardized automated treatment planning framework using Monaco scripting and to evaluate its dosimetric performance, consistency, and organ-at-risk (OAR) sparing compared with conventional manual planning for head-and-neck cancer (HNC) across a multi-institutional network.</p><p><strong>Materials and methods: </strong>A Pareto-based multi-criteria optimization template was iteratively developed, validated on >300 retrospective cases, and implemented across 14 centers. Forty-seven HNC patients were replanned using clinically approved plan (CA), AutoPlan (AP), and AutoPlan iteration (API) approaches and categorized into single-target (<i>n</i> = 11), two-target (<i>n</i> = 35), and three-target (<i>n</i> = 1) cohorts. Plan quality was analyzed in ProKnow using scorecard evaluation, boxplots, histograms, and scatter-plot correlations focusing on target coverage and doses to critical organs including brainstem, spinal cord, and bilateral parotids.</p><p><strong>Results: </strong>Automated workflows demonstrated superior consistency and OAR sparing compared with CA plans. AP achieved the highest overall scorecard performance, while API consistently produced the most stable dose distributions with lower variability and weaker correlations between PTV_High coverage and OAR doses. CA plans exhibited the widest dose variability and stronger OAR dose dependence. Clinically, the automated scripts achieved >70% automation and have been used in over 850 patients with minimal manual intervention.</p><p><strong>Conclusion: </strong>Automated Monaco scripting significantly improves plan quality, consistency, and OAR sparing in HNC radiotherapy, supporting routine clinical adoption - particularly of iterative API workflows - for standardized multi-center practice.</p>","PeriodicalId":51719,"journal":{"name":"Journal of Medical Physics","volume":"51 3","pages":"557-568"},"PeriodicalIF":1.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13675034/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217102","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 Lightweight Python-based Monte Carlo Framework for Visualizing Energy-dependent Dose Calibrator Response: A Nuclear Medicine Education Tool.","authors":"Anil Kumar Pandey, Shanshila Rayamajhi, Jagrati Chaudhary, Param Dev Sharma, Rakesh Kumar","doi":"10.4103/jmp.jmp_178_26","DOIUrl":"10.4103/jmp.jmp_178_26","url":null,"abstract":"<p><strong>Purpose: </strong>Trainees often have limited understanding of radionuclide dose-calibrator physics because access to radioactive materials is restricted by radiation-safety requirements, and the complexity of established Monte Carlo (MC) platforms can present a barrier to early learning. We present a lightweight, Python-based MC simulation framework designed to visualize energy-dependent photon transport and ionization-chamber response for nuclear medicine education.</p><p><strong>Materials and methods: </strong>A spherical gas-filled ionization chamber (radius 5 cm) was modelled. Photon free-path lengths were sampled from exponential attenuation distributions; deposited energy was assigned via a Beta-distributed stochastic transfer fraction (biased toward partial deposition). Ionization yield was estimated assuming 33.97 eV is required to produce one ion pair. Three clinically representative radionuclides - ^99mTc (140 keV, μ = 0.030 cm^-1), ¹³¹I (364 keV, μ = 0.015 cm^-1), and ¹⁸F (511 keV, μ = 0.011 cm^-1) - were each simulated with 100,000 independent photon histories.</p><p><strong>Results: </strong>Intrinsic efficiency decreased monotonically with photon energy: 13.65% (^99mTc), 7.01% (¹³¹I), and 5.25% (¹⁸F). Median deposited energy increased from 37.16 keV to 95.87 keV and 132.07 keV; mean ion-pair yield rose from 1180, 3046, and 4254. Detector response increased nonlinearly from 161 to 214 and 223 ion pairs per emitted photon. All trends were consistent with exponential attenuation theory.</p><p><strong>Conclusions: </strong>The framework reproduces first-order ionization-chamber physics without specialist software or radioactive sources. Its transparency, rapid execution, and interactive visualizations make it well suited for classroom demonstrations and self-directed nuclear medicine physics training, while its explicit limitations underscore the gap between educational modelling and clinical-grade Monte Carlo simulation.</p>","PeriodicalId":51719,"journal":{"name":"Journal of Medical Physics","volume":"51 3","pages":"680-687"},"PeriodicalIF":1.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13675019/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217310","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":"Independent Three-dimensional Dose Calculation from Log Files for Breast VMAT: A Measurement-free Quality Assurance Approach.","authors":"Rosa Angelica Petit Sevilla, Oscar Abel Apaza Blanco, Nayron Medeiros Soares, Maximiliano Musso, Ezequiel Agustín Cabrera Lacey, Luis C Bohorquez, Lucy Wolfsberger, Maria José Almada, Carlos Daniel Venencia","doi":"10.4103/jmp.jmp_38_26","DOIUrl":"10.4103/jmp.jmp_38_26","url":null,"abstract":"<p><strong>Purpose: </strong>The purpose of this study was to evaluate and validate a measurement-free patient-specific quality assurance (PSQA) approach for volumetric modulated arc therapy (VMAT) breast cancer treatments based on independent three-dimensional (3D) dose calculation from Digital Imaging and Communications in Medicine (DICOM) plans and linear accelerator log files.</p><p><strong>Materials and methods: </strong>Seventy clinical breast VMAT plans (36 left-sided and 34 right-sided) were retrospectively analyzed. Independent 3D dose calculations were performed using RadCalc v7.3 based on the DICOM treatment plan (independent dose calculation [IDC] RTPlan) and on delivery log files acquired during the first treatment fraction (IDC LOGs). Results were compared with portal dosimetry (two-dimensional [2D]) and fluence-based analysis. Gamma passing rates (GPRs) were evaluated using 3%/2 mm and 2%/2 mm criteria under action (90%) and tolerance (95%) limits. Receiver operating characteristic (ROC) analysis was performed to evaluate the discriminatory capability. Sensitivity to delivery errors was investigated by introducing controlled multileaf collimator (MLC) shifts (0.5-2.0 mm) and collimator rotation errors (1°-3°).</p><p><strong>Results: </strong>All PSQA methods met the TG-218 action limit (GPR ≥ 90%) at 3%/2 mm. Under the stricter 2%/2 mm criterion, statistically significant differences among quality assurance methods were observed (<i>P</i> < 0.001). ROC analysis demonstrated superior discriminatory performance for 3D dose-based methods, with area under the curve values of 0.9 for both IDC RTPlan and IDC LOGs, compared with 0.7 for portal dosimetry and 0.5 for fluence-based analysis. Error sensitivity analysis showed that stringent gamma criteria (1%/2 mm and 1%/1 mm) improved detection of subtle MLC and collimator deviations, with log file-based 3D reconstruction outperforming 2D approaches.</p><p><strong>Conclusions: </strong>Independent 3D dose calculation is a reliable PSQA approach for breast VMAT, showing higher sensitivity than conventional 2D methods.</p>","PeriodicalId":51719,"journal":{"name":"Journal of Medical Physics","volume":"51 3","pages":"569-577"},"PeriodicalIF":1.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13675032/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217322","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":"Comparative Analysis of Commercial Independent Dose Calculation Systems for Gamma Knife Stereotactic Radiosurgery.","authors":"Michael D Lowe, Jack C Nelson, Joe Caron, Joel J St-Aubin, Steven M Herchko, Deepak K Shrestha, Sridhar Yaddanapudi","doi":"10.4103/jmp.jmp_3_26","DOIUrl":"10.4103/jmp.jmp_3_26","url":null,"abstract":"<p><strong>Purpose: </strong>Independent dose calculation is recommended for stereotactic radiosurgery (SRS) quality assurance. This study evaluates the accuracy of three commercial secondary dose calculation software programs-MUCheck, RadCalc, and ClearCalc-against the Elekta Leksell Gamma Plan (LGP) treatment planning system (TPS) for Gamma Knife SRS.</p><p><strong>Materials and methods: </strong>Dose calculations for 133 patients (239 intracranial targets) planned with the LGP TPS using the TMR10 algorithm were retrospectively analyzed with MUCheck, RadCalc, and ClearCalc. Point dose calculations were compared between each software and LGP. Statistical analysis included analysis of variance (ANOVA) and paired <i>t</i>-tests, with clinical tolerance assessed against the American Association of Physicists in Medicine (AAPM) Task Group (TG)-178 recommendations.</p><p><strong>Results: </strong>Mean differences were 0.34% ± 1.10% (MUCheck), 0.53% ±0.53% (RadCalc), and 0.63% ± 0.16% (ClearCalc). All showed statistically significant differences from LGP but remained within clinical tolerance. Clinical tolerance analysis showed 97.9%, 100%, and 100% of calculations within ±3% for MUCheck, RadCalc, and ClearCalc, respectively. ANOVA revealed significant differences among platforms (<i>F</i> = 10.46, <i>P</i> < 0.001), although all deviations remained clinically acceptable.</p><p><strong>Conclusion: </strong>All three software platforms provide clinically acceptable independent dose verification for Gamma Knife SRS, with deviations well below AAPM TG-178 recommended tolerance levels. Selection of the appropriate secondary dose calculation platform should be based on institutional workflow integration and specific clinical requirements rather than statistical differences alone.</p>","PeriodicalId":51719,"journal":{"name":"Journal of Medical Physics","volume":"51 3","pages":"705-712"},"PeriodicalIF":1.1,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13675016/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217212","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":"Clinical Implementation and Accuracy Evaluation of a Novel Intrafraction Cone-beam Computed Tomography Technique for Stereotactic Radiosurgery and Radiotherapy.","authors":"Deeksha Jaiswal, Monika Goyal, C P Bhatt","doi":"10.4103/jmp.jmp_27_26","DOIUrl":"10.4103/jmp.jmp_27_26","url":null,"abstract":"<p><strong>Purpose: </strong>To evaluate the clinical implementation and setup accuracy of real-time intrafraction cone-beam computed tomography imaging during stereotactic radiosurgery and stereotactic radiotherapy.</p><p><strong>Materials and methods: </strong>Ten patients treated with intracranial SRS/SRT were retrospectively analyzed. All patients were immobilized using the Fraxion stereotactic head system and treated with volumetric-modulated arc therapy (VMAT) on a linear accelerator (Elekta Versa HD). A custom intrafraction cone-beam computed tomography (CBCT) preset was developed in X-ray volume imaging to acquire kV projections during each VMAT arc. Translational (X, Y, and Z) and rotational (pitch, roll, and yaw) errors were calculated by registering intrafraction CBCT to the planning CT using a 6D match and mean values were compared with SRS tolerance thresholds (0.5 mm and 1°).</p><p><strong>Results: </strong>The mean intrafraction translational errors were 0.4 mm (lateral), 0.3 mm (longitudinal), and 0.4 mm (vertical), while rotational errors were 0.3° in (pitch), 0.3° in (roll), and 0.4° in (yaw), all within SRS tolerances. Real-time intrafraction CBCT provided clinically acceptable volumetric images with minimal scatter, enabled continuous motion monitoring during VMAT delivery, eliminated additional CBCT scans and couch adjustments, and reduced treatment time by approximately 3-5 min per arc.</p><p><strong>Conclusion: </strong>Real-time intrafraction CBCT imaging enables highly accurate, efficient, and reproducible setup verification during SRS/SRT. The workflow maintains sub-millimeter precision, reduces treatment duration, and enhances clinical confidence in dose delivery. This approach represents a significant advancement in stereotactic image-guided radiation therapy and warrants the evaluation for extracranial SBRT applications.</p>","PeriodicalId":51719,"journal":{"name":"Journal of Medical Physics","volume":"51 2","pages":"326-332"},"PeriodicalIF":1.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13345032/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148413649","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":"Artificial Intelligence in Chest Computed Tomography Optimization: A Task-based Framework for Balancing Detectability and Radiation Dose.","authors":"Suhas Tivaskar, Anurag Luharia, Gaurav V Mishra, Aditya Shrivastav, Ravi Christian, Sarthak Das","doi":"10.4103/jmp.jmp_57_26","DOIUrl":"10.4103/jmp.jmp_57_26","url":null,"abstract":"<p><p>Computed tomography (CT) of the chest is essential for thoracic imaging because of its high spatial resolution and broad clinical applications. However, variability in acquisition parameters and reconstruction techniques contributes to inconsistent radiation dose and image quality, while conventional optimization approaches remain limited by complex parameter interactions and operator dependence. Optimization in chest CT is increasingly recognized as a task-based challenge in which diagnostic performance depends on lesion detectability rather than global image quality metrics alone. This review synthesizes artificial intelligence (AI)-assisted approaches for optimizing chest CT protocols within a task-based framework, with emphasis on lesion detectability, diagnostic efficacy, and radiation dose reduction, while examining how AI can address protocol variability and improve patient safety. A structured narrative review of PubMed-indexed literature published between January 2015 and January 2025 was conducted, focusing on studies evaluating AI-assisted acquisition, deep learning reconstruction, and radiation dose optimization in adult chest CT. The review highlights that the most effective outcomes are achieved when AI-based reconstruction is integrated with dose optimization strategies such as automatic exposure control and patient-adapted acquisition, rather than applied in isolation. Although the magnitude of benefit varies according to diagnostic task, study design, and evaluation metrics, AI consistently reduces operator variability and promotes protocol standardization. Overall, AI enables task-based optimization of chest CT by prioritizing lesion detectability and diagnostic performance over traditional image quality metrics alone. By integrating acquisition, reconstruction, and continuous feedback mechanisms, AI has the potential to support consistent, patient-specific optimization while maintaining diagnostic efficacy when appropriately validated.</p>","PeriodicalId":51719,"journal":{"name":"Journal of Medical Physics","volume":"51 2","pages":"215-226"},"PeriodicalIF":1.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13345074/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148413646","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}