https://ijmp.mums.ac.ir/ Iranian Journal of Medical Physics en daily 1 Mon, 01 Dec 2025 00:00:00 +0330 Mon, 01 Dec 2025 00:00:00 +0330 https://ijmp.mums.ac.ir/article_27443.html Introduction: Accurate verification of radiation dose delivery remains a major challenge in radiotherapy. Positron emission tomography (PET) imaging of megavoltage (MV)-induced positrons (MVIPET) has recently emerged as a potential in vivo dosimetry technique. In this study, we investigated the feasibility of enhancing MVIPET signals using high-Z nanoparticles (NPs), specifically platinum (Pt) and silver (Ag), to enable real-time dose monitoring during lung radiotherapy. Material and Methods: PET images arising from positrons induced by platinum and silver nanoparticles in a lung tumor were generated during radiotherapy with 6, 10, and 15 MV photon beams using the GATE Monte Carlo code. The resulting images were evaluated for both image quality and dose verification.Results: Results showed that positron production, absorbed dose, and PET signal intensity increased with both photon beam energy and NP concentration, with PtNPs producing significantly higher enhancement than AgNPs. High-quality MVIPET images with acceptable SBR and CNR, and low RMSE, were obtained for PtNP concentrations ≥8 wt% at 10 MV and ≥4 wt% at 15 MV. In contrast, AgNPs required higher concentrations and only yielded reliable monitoring at 15 MV. At 6 MV, image quality and dose–image correlation were insufficient for clinical feasibility.Conclusion: These findings demonstrate that MVIPET, particularly when combined with PtNPs and higher photon energies, is a promising strategy for real-time, non-invasive dose verification in lung radiotherapy. https://ijmp.mums.ac.ir/article_27444.html Introduction: This research aimed to evaluate the radiation-induced secondary cancer risks in normal tissues following Gamma Knife Radiosurgery (GKR) compared with the Volumetric-Modulated Arc Therapy (VMAT).Material and Methods: Eleven patients with meningioma (2 males, 9 females; median age 30 years) were analyzed. For each case, Gamma Knife radiosurgery (GKR) and volumetric modulated arc therapy (VMAT) plans were created and compared using dosimetric-metrics and radiobiological modeling. The Organ Equivalent Dose (OED) was estimated using linear, linear-exponential, and plateau dose–response models, and the Excess Absolute Risk (EAR) was calculated for organs at risk to estimate secondary cancer risk.Results: Dose coverage, conformity, and homogeneity in the Planning Target Volume (PTV) improved significantly ('p'< 0.05) with GKR and VMAT. Higher OED was observed in the optic nerve due to proximity to the target. While the EAR of the optic nerve increased by 11.23%, 13.17%, and 14.86% in GK compared to VMAT for the linear, plateau, as well as linear-exponential models, respectively, in GK plans compared to VMAT, the EAR for the brain stem increased by 54.4%, 17.14%, and 30% in VMAT. GKR had a considerably greater Tumor Control Probability (TCP) (95.76%±2.86) than VMAT (84.29%± 2.27, 'p'= 2.8×10-6).Conclusion: GKR provided better PTV dose coverage and conformity, whereas VMAT achieved superior dose homogeneity. According to EAR, VMAT had a greater second cancer risk than GK. For young patients, advanced radiotherapy techniques should be evaluated in consideration of dosimetric, radiobiological, and secondary cancer risks. https://ijmp.mums.ac.ir/article_27446.html Introduction: High-Dose-Rate (HDR) Vaginal Brachytherapy (VBT) is a standard adjuvant treatment for early-stage endometrial cancer following surgery. While the Total-Reference-Air-Kerma (TRAK) reflects the overall radiation output, its relationship with the actual irradiated tissue volumes and Organ-At-Risk (OAR) exposure remains underexplored in treatments using cylindrical applicators. The aim of this study is to evaluate the correlation between TRAK, irradiated volumes, and doses to OARs.Material and Methods: This retrospective study included 28 Iranian women with histologically confirmed endometrial carcinoma postoperatively treated with HDR brachytherapy using single-channel vaginal cylinder applicators. CT-based treatment planning was performed using Elekta Oncentra software. Prescribed doses were delivered over three weekly fractions. The volumes enclosed by the prescribed isodose surface (VISV) were quantified using 3D Slicer software. TRAK, dosimetric parameters (D90%, D2cc, D0.1cc), and VISV were analyzed. The geometric volume of the applicator was subtracted to determine the actual irradiated tissue volume. Statistical analysis involved Pearson and Spearman correlation coefficients.Results: The mean TRAK was calculated as 0.45 ± 0.06 cGy·m². A significant positive correlation (R²=0.981) was found between TRAK/dprescribed and the irradiated volume (VISV). No significant correlation was observed between TRAK and D2cc doses for the bladder and rectum (R²=0.075 and 0.46).Conclusion: This study demonstrates a strong association between TRAK and isodose volumes in HDR brachytherapy, suggesting TRAK could be a surrogate for overall dose delivery. However, its lack of correlation with OAR dose metrics emphasizes the need for comprehensive (Dose-Volume-Histogram) DVH assessments. Future research may explore incorporating TRAK into models for predicting toxicity and outcomes, enhancing brachytherapy planning and monitoring. https://ijmp.mums.ac.ir/article_27447.html Introduction: To ensure the accuracy of advanced radiotherapy techniques in the therapy of female pelvic malignancies using radiotherapy, comprehensive quality assurance protocols and patient-specific dosimetry was performed using a custom-designed female pelvic (CDFP) phantom and Arc-CHECK.Material and Methods: Volumetric modulated arc therapy (VMAT) and Intensity modulated radiation therapy (IMRT) plans originally developed for patients with cervical cancer were transferred to the CDFP phantom and Arc-CHECK. The doses for patient-specific quality assurance (PSQA) of advanced techniques were calculated using different algorithms in treatment planning system (TPS). To acquire dose measurements, a 0.600 cc ion chamber, Sun Nuclear 125c (SNC125c) and a pinpoint 0.015 cc chamber were employed. The percentage deviation between the computed and administered dose was calculated. Results: Percentage discrepancies between planned and administered doses were analyzed. The maximum percentage deviation, mean ± standard deviation(SD) differences for 0.6 cc ion chamber and a pin-point 0.015 cc chamber were found to be -3.6%, -0.89% ± 1.59  and 3.51%, 1.10% ± 1.31  for the IMRT treatment technique and -3.32%, -0.84% ± 1.53 and 3.6%, 1.25% ± 1.34 for the VMAT treatment technique using AAA algorithms using 6MV beam for CDFP phantom respectively. The maximum of percentage deviation, mean ± SD differences for 0.6 cc ion chamber and a SNC125c chamber were found to be -3.40%, -0.273% ± 2.161 and 3.107%, -0.028% ± 2.046  for CDFP phantom and -2.72%, -0.35% ± 1.41 and 2.74%, 0.197% ± 1.50 for Arc-CHECK using VMAT treatment technique using 6MV beam and Monte Carlo algorithm respectively.Conclusion: The inherent properties of anthropomorphic phantoms like the CDFP, such an anatomical fidelity and tissue equivalence, make them extremely useful in the quality assurance radiotherapy-programs. Their application increases the accuracy of dosimetric validation and therefore, these phantoms are applicable in the use of high-accuracy treatment modalities such as IMRT and VMAT https://ijmp.mums.ac.ir/article_27448.html Introduction: This study aimed to compare the treatment planning volumes and the integral doses (ID) during radiotherapy for prostate cancer using three-dimensional conformal radiotherapy (3D-CRT), intensity-modulated radiotherapy (IMRT), and volumetric-modulated arc therapy (VMAT) techniques.Material and Methods: The study included 10 patients treated in our clinic, diagnosed with intermediate risk prostate cancer at the T2bN0M0 stage, Gleason score ≤ 7 and prostate-specific antigen (PSA) levels between 10-20 ng/mL. For each patient, 6 MV photon energy was used for 3D-CRT, IMRT and VMAT planning.Results: The rectum, bladder, right and left femur, body IDs, the homogeneity index (HI), and conformity index (CI) values were found to be higher in 3D-CRT planning compared to IMRT and VMAT planning (p < 0.001). The body mean, and ID were lower in VMAT planning compared to IMRT and 3D-CRT plans (p < 0.001). The monitor unit (MU) value was significantly higher in the IMRT planning compared to VMAT and 3D-CRT plans, and higher in the VMAT plan compared to 3D-CRT plans (p < 0.001).Conclusion: According to our study, for organ protection, body ID doses and HI, as well as CI, the VMAT planning technique is mostly superior to IMRT and 3D-CRT. In terms of MU, the 3D-CRT planning technique is better than the others. Our findings indicate that VMAT could generate plans that deliver lower IDs to healthy tissues while providing relatively uniform doses to the target area; and lower MUs. https://ijmp.mums.ac.ir/article_27450.html Introduction: Accurate fetal head circumference (HC) estimation from ultrasound images is critical for prenatal assessment, yet current deep learning approaches are limited by scarce annotated training data and inherently low image contrast. These limitations compromise the model's capacity to reliably delineate fetal head boundaries from surrounding uterine structures, directly impacting clinical utility.Material and Methods: This study introduces an attention-based deep learning framework designed to optimize feature extraction by selectively emphasizing diagnostically relevant regions within ultrasound images. The attention mechanism guides the network to prioritize fetal head boundaries while suppressing irrelevant background information, thereby enhancing segmentation precision and feature discrimination during training.Results: Comprehensive evaluation on benchmark ultrasound datasets validates the clinical effectiveness of our approach. The proposed attention-based model achieves a 2% improvement in fetal head detection accuracy compared to current state-of-the-art methods, while simultaneously reducing overfitting probability by 50%. These gains translate to more robust and reliable HC measurements across diverse imaging conditions.Conclusion: Integration of attention mechanisms into deep neural networks substantially advances automated fetal biometry by addressing two critical challenges: measurement accuracy and model generalization. The demonstrated improvements in both detection performance and overfitting mitigation establish attention-guided learning as a viable pathway toward clinically deployable ultrasound analysis systems, with potential to enhance prenatal care quality and consistency. https://ijmp.mums.ac.ir/article_27449.html Introduction: This study evaluated the dosimetric performance and delivery efficiency of three volumetric modulated arc therapy (VMAT) techniques using different multileaf collimator (MLC) systems: MLCi2, Agility and HDMLC for the treatment of pharyngeal cancer.Material and Methods: A retrospective analysis of 21 pharyngeal cancer patients was performed. For each patient, three VMAT plans were generated using a simultaneous integrated boost (SIB) technique, delivering 66 Gy, 60 Gy and 54 Gy to PTV66, PTV60 and PTV54 respectively, over 30 fractions. Plans were created in Monaco treatment planning system (TPS) with MLCi2 (VMAT1) and Agility (VMAT2) and in Eclipse TPS with HDMLC (RapidArc). Plan quality was evaluated using conformity index (CI), homogeneity index (HI) and mean dose for target volumes. Organ-at-risk (OAR) doses were assessed according to RTOG criteria, while treatment efficiency was analyzed using monitor units (MUs) and beam-on time (BOT).Results: All techniques produced clinically acceptable plans. RapidArc demonstrated superior conformity for PTV66 and significantly improved homogeneity for PTV60 and PTV54. It also achieved reduced maximum doses to the spinal cord and decreased low- and medium-dose spillage to normal tissue compared with VMAT1 and VMAT2. RapidArc required fewer MUs and achieved shorter BOT (p < 0.05), indicating enhanced delivery efficiency.Conclusion: While all techniques met clinical requirements, RapidArc with HDMLC achieved superior conformity, better OAR sparing and improved delivery efficiency. These findings underscore the pivotal role of MLC design and configuration in optimizing VMAT plan quality and treatment performance for head and neck cancers. https://ijmp.mums.ac.ir/article_27451.html Introduction: This study evaluated the potential of the Varian Halcyon LINAC and its associated Portal Dosimetry (PD) software, using in vivo megavoltage portal images, as a tool for in vivo portal dosimetry constancy (IVDc) in head and neck cancer (HNC) patients receiving volumetric modulated arc therapy (VMAT).Material and Methods: A retrospective study included 100 HNC patients (mean age 58.4 ± 9.2 years; 72 males, 28 females) treated with three arc VMAT plans on the Halcyon LINAC. A prescription dose of 7000 cGy was delivered over 35 fractions to high risk volumes and 6300 cGy to intermediate risk volumes. Portal images from all fractions (3500) and arcs (10,500) were compared to first fraction baselines via gamma analysis with a 3%/2 mm criterion and 95% passing threshold. Phantom tests evaluated sensitivity to thickness changes.Results: Gamma passing rates exhibited a steady decline over the treatment course (R²=0.9724), with an average 93.9 ± 3.6% passing rate, dropping below 90% after 25 fractions. Failures correlated with thickness variations exceeding 1 cm, which were detected with high accuracy during phantom tests. Pretreatment patient-specific QA yielded 100% pass rates.Conclusion: This study demonstrates the potential of the Halcyon LINAC's integrated PD system and Varian PD software for IVDc in HNC patients treated with VMAT. The results confirm the feasibility of using in vivo MV portal images acquired on the Halcyon LINAC to assess dose consistency throughout treatment. https://ijmp.mums.ac.ir/article_26802.html Introduction The objective of this survey is to develop a deep-learning dose calculation framework that utilizes water dose distribution and the characteristics of heterogeneous phantoms as inputs. Methods Two hundred heterogeneous phantoms with identical geometry but varying thicknesses of bone and lung were simulated using the Dosxyznrc code. The thicknesses of the bone and lung inhomogeneity layers were randomly varied, ranging from 1 cm to 5 cm at different locations along the Z-axis of the heterogeneous phantom. Then, we developed the dose prediction framework based on the deep learning method, with five inputs and a single output channel, to predict dose distribution. Five input channels were included: a matrix representing water dose distribution, a matrix of mass density, a matrix of CT numbers, a matrix indicating the distance of each voxel from the central point of the radiation field, and a matrix containing values of zero or one that corresponds to the radiation field. The output consisted of the dose distribution for each heterogeneous phantom. Results The accuracy of predicted results by the DL-Dose Framework was compared with those obtained by the Monte Carlo method, utilizing the delta index, inspired by the global gamma index in the heterogeneous phantoms. The dose distribution results in the water medium, before the heterogeneities, indicated that 100% of the dose distribution of voxels with deviations of less than 1% from the maximum dose were comparable to and the same as the results measured using the Monte Carlo method. Additionally, 94.2% of the dose distribution for voxels in areas of lung heterogeneity and 98.1% for voxels in regions of bone heterogeneity are comparable to the results obtained using Monte Carlo simulations, with deviations of less than 1% from the maximum dose. conclusions The developed DL-Dose Framework can predict dose distribution in heterogeneous phantoms accurately. https://ijmp.mums.ac.ir/article_27678.html Purpose: The reproducibility of radiomic features could be a serious obstacle that limits further applications. This study aims to assess the reproducibility of MRI radiomic features by using a biological ovarian phantom and different acquisition parameters. Materials and Methods: Three 1.5 Tesla MRI scanners from different manufacturers as the first source, and also alterations in imaging parameters, including slice thickness, space between slices, image weight, and fat saturation sequence, as the second source of feature variations, were evaluated. In addition, to evaluate the effect of image normalization on feature reproducibility, all the images were normalized. Ninety-three radiomic features from 6 feature classes, including First-Order, GLCM, GLDM, GLRLM, GLSZM, and NGTDM, were calculated by the 3D-Slicer. Reproducibility of features was measured by COV, ICC, and CCC. Results: The significant impact of scanner and image weight variation on feature reproducibility is obvious when about 90% and 64% of features showed 20 %< COV, respectively. On the other hand, slice thickness was the least affected source, where 58.8% of features showed excellent reproducibility (COV ≤ 5%). GLRLM showed the best reproducibility against scanner variation (ICC=0.6996 and CCC=0.3503). Also image normalization has positively affected feature reproducibility in the scanner variation scheme. Additionally, good (5 %< COV≤10%) and intermediate (10 %< COV≤20%) COV groups have increased by normalization. Conclusion: MRI radiomic features are highly dependent on image acquisition scanner types and imaging parameters, and utilizing biological phantoms can lead to reliable outcomes that make the way of clinical translation of these results easier. Future works should be the priority in the robustness evaluation of radiomic features, and the inconsistent behavior of the image normalization filter needs higher attention.