Journal of Applied Clinical Medical Physics最新文献

The thermoluminescence dosimeter (TLD) sheet is a measurement device coated with manganese-doped LiB₃O₅ in sheet form. The sheet is 0.2-mm thick and flexible. Hence, it can fit and be installed on irregular surfaces. The current study aimed to evaluate the feasibility of measuring patient surface doses during radiation therapy using TLD sheets. Further, the surface doses when using a three-dimensional (3D) printed bolus were compared. First, calibration of the between TLD sheet measurements and the irradiation doses was performed. Second, dose calculations and TLD sheet measurements were compared to evaluate the error between treatment planning system (TPS) dose calculations and surface dose measurements. Finally, the TLD sheet was fitted to the head phantom to measure the surface dose at the uneven areas under two conditions (with a commercial [CM] bolus and a 3D-printed bolus). Based on the error from the dose calculation algorithm and the difference between the TLD sheet thickness and the dose grid size or evaluation structure volume, the differences between the TLD sheet measurements and the treatment plan doses were up to 68.6% for the X-ray collapsed cone convolution (CCC) method and up to 8.4% for the electron beam Monte Carlo (MC) method. Based on the results of the surface dosimetry of the head phantom, the application of surface dosimetry, particularly skin dosimetry, even on uneven body surfaces, can be feasible.

Purpose: To investigate the relationship between the stereographic and azimuthal equidistant projection (AEP) of the human retina for radiotherapy planning with OPTOS optomap wide-field fundus images (Optos, UK). Further, the geometric accuracy of an OPTOS fundus image is quantified.

Methods: The fundamental relationship between both projection modes was applied to transform images acquired with an OPTOS Silverstone camera to the azimuthal equidistant projection using MATLAB. Fundus images of four patients were used to quantitatively demonstrate the impact for neglecting the proper projection. For that purpose, a delineated contour for each patient was analyzed if created in a treatment planning system, which assumes AEP, and compared with an OPTOS image. Furthermore, an eye model with a novel 3D printed retina pattern was used to quantify the geometric accuracy for an OPTOS optomap image.

Results: The difference between both projections was found substantial, leading to a change in delineated contours of more than 5 mm in the investigated cases and a change of delineated area of more than 40%. The geometric accuracy of OPTOS images of a customized eye model was found to be 0.2 mm on average, increasing to at most ∼0.5 mm at eye angles of 81°.

Conclusion: The fundamental difference in the representation of the eye fundus needs to be accounted for in radiotherapy planning of uveal melanoma. The basic underlying relationship for transformation is known, but more research is required to quantify other aberrations. The novel use of 3D printed retina patterns with known dimensions is providing a flexible approach for further investigations.

Purpose: The dual-layer multileaf collimator (MLC) in Halcyon adds complexities to the dose calculation process owing to the variability of dosimetric characteristics with leaf motion. Recently, an enhanced leaf model (ELM) was developed to refine the MLC model in the Eclipse treatment planning system. This study investigates the performance of the Halcyon ELM by verifying doses for simple slit fields and volumetric modulated arc therapy (VMAT) plans.

Materials and methods: Dose calculations were performed with Acuros XB using the ELM. To commission the leaf-tip model, the dosimetric leaf gap (DLG) was calculated (referred to as DLG_ELM) and compared with Halcyon measurements. The DLGs were assessed under conditions both with and without leaf trailing between the MLC layers. The tongue-and-groove (TG) model was evaluated by comparing leaf-edge profiles and the outputs of the asynchronous sweeping gap. Furthermore, eleven VMAT plans were validated against chamber doses and Delta4 measurements.

Results: DLG_ELM demonstrated variation between layers, measuring 0.42 mm for the proximal layer and 0.23 mm for the distal layer, and showed a correspondence with the measured DLGs in 0.1 mm. Additionally, ELM reduced the discrepancy between calculated and measured DLGs when accounting for leaf trailing. In the TG model test, ELM calculations successfully mirrored the measured leaf-edge profiles. Moreover, the median dose difference between ELM calculations and chamber doses was -0.8% in asynchronous sweeping gaps. In the VMAT dose verification, the incorporation of ELM enhanced the target dose and resulted in a gamma pass rate (2%/2 mm) exceeding 95%.

Conclusion: Halcyon ELM considerably improved the accuracy of simulating actual leaf-tip transmission, both with and without leaf trailing, and it effectively accounted for the additional blocking caused by TG design. Furthermore, the introduction of ELM in Eclipse considerably enhanced the VMAT dose calculation. ELM addresses the limitations of traditional leaf models and reduces uncertainties in Halcyon dose calculations.

Purpose: To compare plan quality among photon volumetric modulated arc therapy (VMAT) and intensity-modulated proton therapy (IMPT) with robustness using three different proton beam delivery systems with various spot size (σ) ranges: cyclotron-generated proton beams (CPBs) (σ: 2.7-7.0 mm), linear accelerator proton beams (LPBs) (σ: 2.9-5.5 mm), and linear accelerator proton mini beams (LPMBs) (σ: 0.8-3.9 mm) for the treatment of head and neck (HN) cancer with bilateral neck irradiation.

Methods: Ten patients treated for oropharynx cancer with bilateral neck irradiation were planned using CPBs, LPBs, LPMBs, and VMAT. The homogeneity index (HI), mean body dose, and defined volumetric doses for selected critical organs-at-risk (OARs) were compared. Set-up uncertainties of ±3 mm and ± 3.5% range uncertainties were included in robust evaluation using V_95%Rx > 95% (Volume that covers 95% of the target volume at 95% of the prescription (Rx) dose) to high dose and low dose CTV volumes (CTV_70 Gy and CTV_56 Gy). VMAT and proton plans were compared in terms of OAR doses and mean body dose only. Homogeneity Indices were compared among IMPT plans in addition to OAR doses. The Wilcoxon signed-rank test was used to evaluate statistical differences between evaluation metrics for VMAT plans and all proton plan types.

Results: OAR dose metrics were improved by 2% to 30% from CPB plans to LPB or LPMB plans. Compared to photon VMAT plans, all OAR doses except for mandible dose metrics were improved by 2% to 53% for all proton plans. The mean body dose was also improved by 7.5% from CPB to LPB and by 10.8% from CPB to LPMB. In addition, the mean body dose was also improved by 44% from VMAT to CPB, by 48% from VMAT to LPB, and by 50% from VMAT to LPMB plans. Compared to CPB plans, HI was significantly better (p < 0.05) for the LPB and LPMB plans. HI also improved considerably from VMAT to CPB, LPB, and LPMB. For both CTV_70 Gy and CTV_56 Gy, average robust evaluation across all worst-case scenarios was slightly better for CPB plans, with an average of V_95%Rx of the CTV_70 Gy of 97.6% ± 1.22%, followed by 97.2% ± 1.31% and 97.2% ± 1.35% for LPB and LPMB plans, respectively. Robustness for CTV_56 Gy showed comparable robustness across all proton plan types, with an average V_95%Rx of 97.4% ± 0.87% for CPB, 97.4% ± 1.21%, and 97.5% ± 1.08% for CPB, LPB, and LPMB plans, respectively.

Conclusion: With decreased spot size, the LPB and LPMB are excellent alternatives to VMAT and CPB therapy and can significantly reduce the dose to normal tissue.

Purpose: Breath-hold techniques are widely used in radiation therapy to minimize respiratory-induced tumor or organ-at-risk motion. However, residual motion persists, necessitating a reliable daily evaluation method.

Methods: At our institution, fiducial markers serve as surrogates for target localization in pancreatic cancer treatment. We developed an automated method to detect fiducial markers in every projection image of cone-beam computed tomography (CBCT) scans acquired for patient setup and positioning verification. This method was retrospectively validated using data from nine pancreatic cancer patients.

Results: Residual motion was observed in all patients during breath-hold maneuvers. Intrafraction target motion in repeated breath-hold simulation CT scans averaged 1.9 ± 2.2 mm, with a maximum displacement of 8 mm in the superior-inferior direction. Within a single CBCT scan, residual motion reached up to 7.3 mm, with an average drifting range of 3.8 ± 1.1 mm across 94 CBCT scans. The average standard deviation of drift was 1.5 ± 0.5 mm. Significant drift (1.3 ± 1.2 mm) and inter-breath-hold gaps (2.6 ± 2.0 mm) were detected within the same CBCT scan.

Conclusion: Our method enables daily residual motion assessment without additional equipment or extra radiation exposure. This information is critical for refining planning margins in online adaptive radiation therapy, improving treatment precision and patient safety.

Lung cancer is one of the most common malignant tumors in the world. Early detection and precise treatment are of great significance to clinical decision-making and patient prognosis. As an emerging imaging technology, dual-energy computed tomography (DECT) has increasingly prominent advantages in multi-parameter and quantitative analysis in assessing the benign and malignant, classification, and prognosis of lung cancer. Radiomics uses an automated high-throughput method to extract a large number of quantitative features from medical images, quantify tumor heterogeneity, monitor tumor development and prognosis, and provide new ideas for the diagnosis and identification of lung cancer. This article will review the application progress of DECT post-processing technology combined with radiomics in lung cancer diagnosis, identification, biomarker and gene prediction, and prognosis assessment.

Integrating artificial intelligence (AI) into radiation oncology has revolutionized clinical workflows, enhancing efficiency, safety, and quality. However, this transformation comes with a price of increased complexity and the emergence of unpredictable events. This letter proposes a framework based on high reliability organization (HRO) principles for managing real-time, unforeseen events. The framework emphasizes proactive risk assessment, adaptive teamwork at the situation assessment point, and reactive learning through incident analysis by placing human-centered decision-making at the core. Integrating cognitive diversity, psychological safety, and emotional intelligence fosters collective intelligence, enabling teams to navigate AI-driven complexities while safeguarding patient safety.

Purpose: This work describes a single institution experience of commissioning a real-time target tracking and beam control system, known as comprehensive motion management, for a 1.5 T Elekta MR-Linac.

Methods: Anatomical tracking and radiation beam control were tested using the MRI^4D Quasar motion phantom. Multiple respiratory breathing traces were modeled across a range of realistic regular and irregular breathing patterns ranging between 10 and 18 breaths per minute. Each of the breathing traces was used to characterize the anatomical position monitoring (APM) accuracy, and beam latency, and to quantify the dosimetric impact of both parameters during a respiratory-gated delivery using EBT3 film dosimetry. Additional commissioning tasks were performed to verify the dosimetric constancy during beam gating and to expand our existing quality assurance program.

Results: It was determined that APM correctly predicted the 3D position of a dynamically moving tracking target to within 1.5 mm for 95% of the imaging frames with no deviation exceeding 2 mm. Among the breathing traces investigated, the mean latency ranged between -21.7 and 7.9 ms with 95% of all observed latencies within 188.3 ms. No discernable differences were observed in the relative profiles or cumulative output for a gated beam relative to an ungated beam with minimal dosimetric impact observed due to system latency. Measured dose profiles for all gated scenarios retained a gamma pass rate of 97% or higher for a 3%/2 mm criteria relative to a theoretical gated dose profile without latency or tracking inaccuracies.

Conclusion: MRI-guided target tracking and automated beam delivery control were successfully commissioned for the Elekta Unity MR-Linac. These gating features were shown to be highly accurate with an effectively small beam latency for a range of regular and irregular respiratory breathing traces.

Purpose: Results of a prospective, randomized controlled trial at our institute demonstrate an association between the dose to the left hippocampus and neurocognitive decline post-radiotherapy for patients with glioblastoma. To minimize the dose to the left hippocampus, a left hippocampus sparing model was created using RapidPlan (RP) and multi-criteria optimization (MCO).

Materials and methods: For 147 patients with newly diagnosed glioblastoma treated with volumetric modulated arc therapy (VMAT), the left and right hippocampus were delineated. Ninety-seven of 147 VMAT plans were used to configure a RP model named HCS1. The remaining 50 VMAT plans were used for the model validation. All 97 plans were replanned with the HCS1 and further optimized using MCO (HCS1+MCO). MCO was used to explore the trade-off between reducing the left hippocampus mean dose and planning objectives for the targets and other organs-at-risk (OAR) for HCS1 plans. These plans were used to create a new model called HCS2. MCO and RP model configuration were done within the Eclipse treatment planning system.

Results: The final HCS2 model decreased the mean dose to the left hippocampus by 26% compared to clinically treated plans without reducing target coverage for 50 validation data. The mean dose to the left hippocampus decreased from 32.65 Gy in clinically treated plans, 30.45 Gy in HCS1-generated plans, and 24.04 Gy in HCS2-generated plans. The mean volume receiving 95% of the prescription dose (V95%) of the planning target volume was 99.08% ± 1.39% in clinically treated plans, 99.03% ± 1.37% in HCS1-generated plans, and 98.80% ± 1.48% in HCS2-generated plans. Mean dose to 0.1 cc of the brainstem improved from 45.91 Gy in clinically treated plans to 39.29 Gy in HCS2-generated plans.

Conclusions: The RP model and MCO helps to decrease left hippocampus mean dose while maintaining the target volume coverage and OAR sparing comparable to clinically treated plans for glioblastoma patients.

Background: Variation in imaging protocol, patient positioning, and the presence of artifacts can vary image quality in CT images used for radiotherapy planning. Automated methods for spatial resolution (SR) estimation exist but require further investigation and validation for wider adoption.

Purpose: To validated previously existing algorithm for SR estimation and introduce improvements that make it robust to patient positioning, CT protocol, site, and artifacts.

Method: A reference algorithm based on the previous gold standard was recreated and modified to improve robustness. The algorithms were tested on three different datasets: (1) a cylindrical ACR CT QC phantom scanned using a Siemens SOMATOM Definition Edge scanner and reconstructed using 61 different kernels, (2) a set of anthropomorphic phantoms scanned with the presence of artifacts common to clinical acquisitions such as blankets and immobilization devices, and (3) a clinical patient dataset of head and neck (HN) CT scans (nine patients) and spine/pelvis (10 patients). The robustness of both algorithms was tested on the clinical patient data.

Results: Over the range of tested kernels, both algorithms were accurate when the ground truth MTF f₅₀ was within the range 0.2-0.7 mm^-1 in the cylindrical phantom datasets with an RMS error of 10.3% and 3.8% for the reference and modified versions of the algorithm, respectively, as compared to the ground truth. In the anthropomorphic phantom datasets the reference algorithm showed an 8.4% and 30.0% difference from ground truth for the Pelvic and HN phantoms, respectively, while the modified algorithm showed 4.9% and 3.9% percent difference from ground truth. In the clinical dataset the reference algorithm estimated a mean f₅₀ value of 0.21 ± 0.03 mm^-1 and 0.25 ± 0.03 mm^-1 for pelvis/spine while the reference algorithm estimated mean of 0.28 ± 0.02 and 0.29 ± 0.01 mm^-1 for HN and pelvis/spine, respectively, as compared to the ground truth found to be 0.28 mm^-1 on the cylindrical phantom.

Conclusion: The SR algorithm was validated cylindrical/anthropomorphic phantoms and clinical CT scans. Further modifications were tested and showed improved accuracy in more challenging CT acquisitions.