Journal of Medical Physics最新文献

Introduction: The objective of the present study is to classify chest X-ray (CXR) images into COVID-positive and normal categories with the optimal number of features extracted from the images. The successful optimal feature selection algorithm that can represent images and the classification algorithm with good classification ability has been determined as the result of experiments.

Materials and methods: This study presented a framework for the automatic detection of COVID-19 from the CXR images. To enhance small details, textures, and contrast of the images, contrast limited adaptive histogram equalization was used. Features were extracted from the first-order statistics, Gray-Level Co-occurrence Matrix, Gray-Level Run Length Matrix, local binary pattern, Law's Texture Energy Measures, Discrete Wavelet Transform, and Zernikes' Moments using an image feature extraction tool "pyFeats. For the feature selection, three nature-inspired optimization algorithms, Grey Wolf Optimization, Particle Swarm Optimization (PSO), and Genetic Algorithm, were used. For classification, Random Forest classifier, K-Nearest Neighbour classifier, support vector machine (SVM) classifier, and light gradient boosting model classifier were used.

Results and discussion: For all the feature selection methods, the SVM classifier gives the most accurate and precise result compared to other classification models. Furthermore, in feature selection methods, PSO gives the best result as compared to other methods for feature selection. Using the combination of the SVM classifier with the PSO method, it was observed that the accuracy, precision, recall, and F1-score were 100%.

Conclusion: The result of the study indicates that with optimal features with the best choice of the classifier algorithm, the most accurate computer-aided diagnosis of CXR can be achieved. The approach presented in this study with optimal features may be utilized as a complementary tool to assist the radiologist in the early diagnosis of disease and making a more accurate decision.

Purpose: Rising cancer incidences, complex treatment techniques, and workflows have all impacted the radiotherapy scheduling process. Intelligent appointment scheduling is needed to help radiotherapy users adapt to new practices.

Materials and methods: We utilized van Herk's safety margin formula to determine the radiotherapy department's treatment scheduling window (TSW). In addition, we examined the influence of in-room imaging on linac occupancy time (LOT). Varian Aria™ software version 15.1 was used to collect retrospective data on LOT, treatment site, intent, techniques, special protocol, and in-room imaging.

Results: Treatment scheduling windows varied across treatment sites. The mean TSW using van Herk's formalism was 31.5 min, significantly longer than the current TSW of 15 min (P = 0.036), with the pelvic site having the longest (43.8 min) and the brain site having the shortest (12 min). 28% of patients exceeded the in-practice TSW of 15 min. 46.2% of patients had multiple images per fraction, with the proportion being highest in pelvic patients (33%). Patients treated with palliative intent, intensity-modulated radiotherapy, special protocols (bladder protocol and gating), and multiple in-room images per fraction had significantly higher LOT. High treatment time uncertainty was observed in the pelvic and thorax sites, indicating the impact of in-room imaging frequency and on-couch treatment decisions on overall treatment time and indicating that current treatment practices should be reviewed and modified if necessary.

Conclusions: The time margin recipe can customize the treatment scheduling window and improve treatment practices. This formalism can help manage the radiotherapy department's workload and reduce patient wait times.

Purpose: Volumetric modulated arc therapy (VMAT) has been increasingly used for cancer patients due to the fast delivery and improved dose conformity. Adaptive radiotherapy (ART) can significantly decrease dose to normal tissues and allow for dose escalation. However, current imaging techniques cannot provide four-dimensional (4D) patient anatomy or dose information during VMAT, which is critical for ART that involves respiratory motion. A novel imaging tool named VMAT-computed tomography (VMAT-CT) has the potential to reveal intra-fractional patient information. The goal of this study was to evaluate the feasibility of 4D adaptive VMAT based on 4D VMAT-CT.

Materials and methods: A commercial QUASAR respiratory phantom and an in-house deformable lung phantom were used in this study, and lung VMAT plans, including 4D union plan and 4D ART plan, were generated for the phantoms. A real lung patient's plan was also used in this feasibility study. ART plans based on 4D VMAT-CT were created for the phantoms and the real patient when planning goals were not met. Dose escalation plan based on 4D VMAT-CT was also created for the real patient.

Results: Planning target volume (PTV) coverage for the QUASAR phantom was 85.5% after breathing pattern being changed, and went up to 95% after adaptive re-planning. PTV coverage for the deformable phantom was 93% after deformation and breathing pattern being changed, and went up to 95% after re-planning. Re-planning and dose escalation were feasible and can spare normal tissues for the real patient. 4D ART plan based on 4D VMAT-CT required smaller margins than 4D union plan while maintaining the same prescription dose coverage.

Conclusions: ART based on 4D VMAT-CT is feasible and would potentially facilitate re-planning and PTV dose escalation for VMAT patients who have the motion issue.

Purpose: This study aims to investigate the use of the neutral comet assay to assess deoxyribonucleic acid (DNA) damage in lymphocytes exposed to high doses of radiation.

Materials and methods: The research was conducted by obtaining informed consent, after which blood samples were taken from seven healthy individuals and this study was approved by the institutional ethics committee. At first, for the determination of dose-effect curves, samples obtained from the first five individuals were irradiated for doses ranging from 0 to 35 Gy after which they were processed under neutral comet assay. In order to verify the determined dose-effect curves, a test dose of 15 Gy was delivered to the samples obtained from the sixth and seventh individuals. The amount of DNA damage from the obtained comet assay images was analyzed using four comet assay parameters namely % tail DNA, tail length, tail moment (TM), and Olive TM (OTM). The most suitable comet assay parameter was evaluated based on the obtained dose-effect curves. Furthermore, the distribution of individual cells for each dose point was evaluated for all the four comet assay parameters to find the optimal parameter.

Results: From our results, it was found that from 0 to 25 Gy all the four comet assay parameters fit well into a linear quadratic curve and above 25 Gy saturation was observed. Based on the individual cell distribution data, it was found that % tail DNA could be an optimal choice to evaluate DNA damage while using neutral comet assay for high-dose ionizing radiation.

Conclusion: The neutral comet assay could be a potential tool to assess DNA damage from high doses of ionizing radiation greater than 5 Gy.

Aim: The main objective of this work is to propose an efficient segmentation model for accurate and robust lung segmentation from computed tomography (CT) images, even when the lung contains abnormalities such as juxtapleural nodules, cavities, and consolidation.

Methodology: A novel deep learning-based segmentation model, pyramid-dilated dense U-Net (PDD-U-Net), is proposed to directly segment lung regions from the whole CT image. The model is integrated with pyramid-dilated convolution blocks to capture and preserve multi-resolution spatial features effectively. In addition, shallow and deeper stream features are embedded in the nested U-Net structure at the decoder side to enhance the segmented output. The effect of three loss functions is investigated in this paper, as the medical image analysis method requires precise boundaries. The proposed PDD-U-Net model with shape-aware loss function is tested on the lung CT segmentation challenge (LCTSC) dataset with standard lung CT images and the lung image database consortium-image database resource initiative (LIDC-IDRI) dataset containing both typical and pathological lung CT images.

Results: The performance of the proposed method is evaluated using Intersection over Union, dice coefficient, precision, recall, and average Hausdorff distance metrics. Segmentation results showed that the proposed PDD-U-Net model outperformed other segmentation methods and achieved a 0.983 dice coefficient for the LIDC-IDRI dataset and a 0.994 dice coefficient for the LCTSC dataset.

Conclusions: The proposed PDD-U-Net model with shape-aware loss function is an effective and accurate method for lung segmentation from CT images, even in the presence of abnormalities such as cavities, consolidation, and nodules. The model's integration of pyramid-dilated convolution blocks and nested U-Net structure at the decoder side, along with shape-aware loss function, contributed to its high segmentation accuracy. This method could have significant implications for the computer-aided diagnosis system, allowing for quick and accurate analysis of lung regions.

Objective: To examine the dosimetric characteristics of circular cones, the accuracy of dose modeling and overall treatment delivery of two radiosurgery systems integrated on a linear accelerator (Linac).

Materials and methods: The dosimetric characteristics of circular cones (4-17.5 mm) from Varian (VC) and BrainLAB (BLC) were measured for 6 MV flattening filter free beam from Edge linac using stereotactic field diode and 0.65 cc ionization chamber following established protocols. The Eclipse and iPlan modeled dose distribution for VCs and BLCs were validated with EBT3-film measurement. End-to-end tests were performed using stereotactic phantom having PTW 60008 diode connected to a Dose-1 electrometer.

Results: The depth at dose maximum, TRP²⁰₁₀ and dose at 10cm depth of the same size VC and BLC agree within ± 0.7 mm, ± 0.71% and ± 0.81% respectively. Full width at half maximum (FWHM) of any cone beyond 15 mm depth increases at 1% of nominal cone size per 10 mm depth. The penumbra of 4mm and 17.5mm VC at 15 mm depth was 1.1 mm and 1.50 mm. At 300 mm depth, penumbra increased by around 0.4 mm for 4 mm cone and up to 1 mm for cone size ≥12.5 mm. The VCs penumbra values were within ±1mm of the corresponding BLCs. Scatter factors for VCs varies from 0.609 to 0.841 and were within ± 1.0% of corresponding values of BLCs. Agreement between the Eclipse and iPlan computed dose fluence and the EBT3-film measured dose fluence was >98% (γ: 1%@1 mm), and the absolute dose difference was ≤ 2.2%, except for the 4 mm cone in which it was >96% and ≤4.83%. Target localization using cone-beam computed tomography was accurate within ± 0.8 mm and ± 0.3° in translation and rotation. The end-to-end dose delivery accuracy for both radiosurgery systems was within ± 3.62%.

Conclusion: The dosimetric characteristics of Varian and BLC cones of same diameter was comparable. Both Eclipse and iPlan cone planning system modeled dose fluences agree well with the EBT3 film measurement. The end-to-end tests revealed an excellent target localization accuracy of Edge linac with satisfactory and comparable absolute dose agreement between Varian and BLC radiosurgery systems and hence these can be interchanged on edge linac.

Purpose: In our institution, stereotactic radiosurgery of multiple brain metastases is performed with the CyberKnife® (CK) device, using fixed/Iris collimators. In this study, nineteen fixed/Iris plans were recalculated with the multileaf collimator (MLC), to assess if it is possible to produce plans with comparable dosimetric overall quality.

Materials and methods: For consistent comparisons, MLC plans were re-optimized and re-normalized in order to achieve the same minimum dose for the total planning target volume (PTV_tot). Conformation number (CN), homogeneity index (HI) and dose gradient index (DGI) metrics were evaluated. The dose to the brain was evaluated as the volume receiving 12 Gy (V₁₂) and as the integral dose (ID). The normal tissue complication probability (NTCP) for brain radionecrosis was calculated as a function of V₁₂.

Results: The reoptimized plans were reviewed by the radiation oncologist and were found clinically acceptable according to the The American Association of Physicists in Medicine (AAPM) Task Group-101 protocol. However, fixed/Iris plans provided significantly higher CN (+8.6%), HI (+2.2%), and DGI (+44.0%) values, and significantly lower ID values (-35.9%). For PTV_tot less than the median value of 2.58cc, fixed/Iris plans provided significantly lower NTCP values. On the other side, MLC plans provided significantly lower treatment times (-18.4%), number of monitor units (-33.3%), beams (-46.0%) and nodes (-21.3%).

Conclusions: CK-MLC plans for the stereotactic treatment of brain multi metastases could provide an important advantage in terms of treatment duration. However, to contain the increased risk for brain radionecrosis, it could be useful to calculate MLC plans only for patients with large PTV_tot.

An extended version of task group report (TG)-119 dosimetric tests was introduced and tested on the TrueBeam linear accelerator setup. Treatment plan results and quality assurance (QA) results of RapidArc (RA) and intensity-modulated radiotherapy (IMRT) were compared to understand the limitation and efficacy of the RA and IMRT system of the linear accelerator. Test structure sets were drawn on OCTAVIUS four-dimensional (4D) phantom computed tomography scan data for this study. We generated treatment plans based on the specified goal in the Eclipse™ treatment planning system using RA and IMRT in the study phantom. We used the same planning objectives for RA and IMRT techniques. Planar dose verification was performed using electronic portal imaging device and OCTAVIUS 4D phantom. The treatment log file was further analyzed using Pylinac (V2.4.0 (Open Source Code library available on Github, runs under Python programming language)) to compare the dosimetric outcome of RA and IMRT. Dose to the planning target volume (PTV) 1-5 and organ at risk (OAR) were analyzed in this study for the efficiency comparison of RA and IMRT. The primary objective was accomplished by adhering to the dose constraints associated with PTV 2 and the OAR. RA and IMRT also met the secondary objective. The tertiary goal of dose delivery to PTV 4 was met with RA but not IMRT. This study can be utilized to compare different institutions' planning and patient-specific QA (PSQA) procedures. The findings of this study were in line with the published works of the literature. A multi-institutional planning and delivery accuracy audit can be built using this structure and set of planning objectives having similar PSQA phantom. The TG-119 report incorporated test challenges that were combined in a single study set and a single plan. This reduces the complexity of performing the original TG-119 tests, whereas keeping the challenges as introduced in the TG-119 report. This study's planning and dosimetric results could be further utilized for dosimetry audit with any institute having a linear accelerator and OCTAVIUS 4D phantom for PSQA.

Background: The aim of the current study was to compare three different dose-calculating algorithms, i.e., superposition (SP), fast SP (FSP), and convolution (CV), for breast cancer patients treated with intensity-modulated radiotherapy (IMRT) and field-in-Field forward plan IMRT (FiF-FP-IMRT).

Materials and methods: The current retrospective study involved 100 postmastectomy breast cancer patients who were given radiotherapy using IMRT and FiF-FP-IMRT planning techniques. All the initially SP-calculated plans were recalculated with the same monitor units for FSP and CV algorithm without change in any of the other planning parameters. The isodose distribution and various plan evaluating parameters, for example, conformity index (CI), homogeneity index, and uniformity index target volume and normal structure doses were compared and analyzed for all the different algorithm calculated plans.

Results: In the IMRT plans, all the target and normal structure dose-volume parameters showed a significant difference between all the three different algorithms with P < 0.05. In the FiF-FP-IMRT plans, CV algorithm showed a significant difference in most of the target and normal structure dose-volume parameters. Among quality indexes, only CI showed a significant difference between all the algorithms in both the planning techniques. R₅₀ showed a significant difference with the CV algorithm in both the planning techniques.

Conclusion: The change in the dose calculation algorithm resulted in dosimetric changes which must be evaluated by the medical physicists and oncologists while evaluating treatment plans. In the current study with breast patients, the results obtained for target and normal structure doses using the CV algorithm are overestimated as compared to SP and FSP algorithms, producing variable results in air and bony normal structures. However, the ipsilateral lung V₅ parameter and the ipsilateral humeral head mean dose were found to be underestimated by the CV algorithm as compared to the SP and FSP algorithm in both the planning techniques.