Journal of Applied Clinical Medical Physics最新文献

Background: Whole bladder irradiation is an organ preservation treatment approach for muscle-invasive bladder cancer (MIBC). Conventional planning margins, typically 15-20 mm, increase normal tissue toxicity and limit possible dose escalation.

Purpose: The study aimed to develop a patient-specific adaptive margin recipe for whole bladder irradiation to minimize the planning target volume (PTV) while preserving adequate dose coverage.

Methods: Sixteen patients who received whole-bladder irradiation were retrospectively selected for this study. We proposed a patient-specific anisotropic adaptive margin recipe, derived from the first five fractions of kV-CBCTs, to account for inter-fractional bladder changes. This recipe was validated using kV-CBCTs from fractions six to ten and the final five fractions. The goal was to achieve a residual volume, defined as the percentage of daily bladder volume (V_daily) outside the PTV, of less than 5%. Adaptive and conventional plans were created using proposed and conventional margins, respectively. A dosimetric comparison of targets and organs-at-risk (OARs) was performed between the two approaches.

Results: (V_daily) decreased throughout the treatment course. The most notable inter-fractional bladder variations were in the superior and anterior directions. The patient-specific anisotropic adaptive margins, averaging 6 mm (± 2.9 mm), achieved a residual volume of less than 5%. Compared to conventional planning, the adaptive approach reduced PTV volume by an average of 135.3 cc (± 46.6 cc). A significant correlation (p < 0.05) was identified between residual volume and adaptive margins in the anterior, superior, left, and right directions. Using the proposed adaptive margins, the median residual volume was 0.71% (interquartile range 0.09%-3.55%), and the median (V_daily) receiving the prescribed dose was 99.1% (interquartile range 95.3%-99.9%). Adaptive plans demonstrated superior OAR sparing compared to conventional plans.

Conclusions: The proposed patient-specific adaptive margin recipe for whole bladder irradiation resulted in margins smaller than conventional ones, optimized normal tissue sparing, and maintained adequate PTV coverage.

Purpose: The goal of this study was to assess the feasibility of a cost-effective prototype of a laser-based respiratory motion detection system utilizing a Leuze LDS for breath monitoring through calibration and volunteer tests.

Methods: This study was performed using the Anzai AZ-773 V and computerized imaging reference systems (CIRS) motion phantoms for calibration tests. The calibration of the laser-based respiratory motion detection system involved spatial accuracy testing, amplitude calibration, and temporal accuracy. Volunteer testing was conducted on eight volunteers at the inferior end of the sternum and the abdomen area. The accuracy of the data recorded by the laser-based respiratory motion detection system was validated against established clinical reference tracking systems namely real-time position management (RPM) and Anzai AZ-733 V system.

Results: Calibration with an Anzai AZ-773 V and CIRS phantoms demonstrated an average error of 1.17% ± 0.64% and an average amplitude calibration correlation coefficient of 0.975 ± 0.004. Volunteer tests, compared to the Anzai AZ-733 V clinical system and RPM system, revealed average correlation coefficients for deep inspiration breath-hold are 0.931 ± 0.02 and 0.936 ± 0.03, respectively, and for free breathing are 0.85 ± 0.07 and 0.77 ± 0.1, respectively.

Conclusions: Overall, the data suggest that the in-house laser-based respiratory motion detection system performed well, with an error percentage below 10%. A reasonably good correlation coefficient was obtained, indicating that the readings obtained from the laser system are consistent with those set on the phantom and clinical respiratory motion detection systems. Although promising through the calibration process and volunteer tests, further studies are required to generate trigger data linked directly to computerized tomography and linear accelerator facilities, thereby advancing the clinical viability of this innovative laser-based respiratory motion detection system.

Purpose: This study introduced a novel 3D dosimetry system for radiotherapy in order to address the limitations of traditional quality assurance methods in precision radiotherapy techniques.

Methods: The research required the use of scintillation material, optical measurements, and a dose reconstruction algorithm. The scintillation material, which mimics human soft tissue characteristics, served as a both physical phantom and a radiation detector. The dose distribution inside the scintillator can be converted to light distributions, which were measured by optical cameras from different angles and manifested as pixel values. The proposed dose reconstruction algorithm, LASSO-TV, effectively reconstructed the dose distribution from pixel values, overcoming challenges such as limited projection directions and large-scale matrices.

Results: Various clinical plans were tested and validated, including a modified segment from the SBRT plan and IMRT clinical plan. The dosimetry system can execute full 3D dose determinations as a function of time with a spatial resolution of 1-2 mm, enabling high-resolution measurements for dynamic dose distribution. Comparative analysis with mainstream device MapCHECK2 confirmed the accuracy of the system, with a relative measurement error of within 5%.

Conclusions: Testing and validation results demonstrated the dosimetry system's promising potential for dynamic treatment quality assurance.

Purpose: Distributive stereotactic radiosurgery (dSRS) is a form of fractionation where groups of metastases are treated with a full single-fraction dose on different days. The challenge with dSRS is determining optimal target groupings to maximize the distance between targets treated in the same fraction. This study aimed to develop and validate an accessible optimization technique for distributing brain metastases into optimal treatment fractions using a genetic algorithm.

Methods: The Evolutionary Solver in Excel was used to optimize the grouping of target volumes for distributive SRS fractionation. The algorithm's performance was tested using three geometric test cases with known optimal solutions, 400 simulations with randomly distributed target volumes, and clinical data from five GammaKnife patients. The objective function was defined as the sum of average distances between target volumes within each fraction, with constraints ensuring 2-5 targets per fraction, each target being assigned to only one fraction, and a constraint on the minimum distance between any two targets in the same fraction.

Results: The Evolutionary Solver successfully identified optimal target groupings in all geometric test cases. Compared to random groupings, the mean distance between target volumes increased by 9%, from 68.1 ± 0.8  to 74.2 ± 1.1 mm post-optimization, while the minimum distance between targets increased by 57%, from 24.9 ± 5.9  to 39.1 ± 7.5 mm. In clinical test cases, the mean distances improved from 81.6 ± 11.9 mm for manual target grouping to 85.6 ± 14.5 mm for optimized target grouping. The minimum separation improved from 35.2 ± 14.5 mm with manual grouping to 51.6 ± 14.7 mm with optimized grouping, corresponding to a mean improvement of 16.4 ± 6.1 mm.

Conclusion: The Evolutionary Solver in Excel provides a systematic and reproducible method for optimizing distributive target groupings in SRS and enhances spatial separation.

Purpose: Cardiotoxicity is one of the major concerns in breast cancer treatment, significantly affecting patient outcomes. To improve the likelihood of favorable outcomes for breast cancer survivors, it is essential to carefully balance the potential advantages of treatment methods with the risks of harm to healthy tissues, including the heart. There is currently a lack of comprehensive, data-driven evidence on effective risk stratification strategies. The aim of this study is to investigate the prediction of cardiotoxicity using machine learning methods combined with radiomics, clinical, and dosimetric features.

Materials and methods: A cohort of 83 left-sided breast cancer patients without a history of cardiac disease was examined. Two- and three-dimensional echocardiography were performed before and after 6 months of treatment to evaluate cardiotoxicity. Cardiac dose-volume histograms, demographic data, echocardiographic parameters, and ultrasound imaging radiomics features were collected for all patients. Toxicity modeling was developed with three feature selection methods and five classifiers in four separate groups (Dosimetric, Dosimetric + Demographic, Dosimetric + Demographic + Clinical, and Dosimetric + Demographic + Clinical + Imaging). The prediction performance of the models was validated using five-fold cross-validation and evaluated by AUCs.

Results: 58% of patients showed cardiotoxicity 6 months after treatment. Mean left ventricular ejection fraction and Global longitudinal strain decreased significantly compared to pre-treatment (p-value < 0.001). After feature selection and prediction modeling, the Dosimetric, Dosimetric + Demographic, Dosimetric + Demographic + Clinical, Dosimetric + Demographic + Clinical + Imaging models showed prediction performance (AUC) up to 73%, 75%, 85%, and 97%, respectively.

Conclusion: Incorporating clinical and imaging features along with dose descriptors are beneficial for predicting cardiotoxicity after radiotherapy.

Purpose: In radiotherapy, body contour inaccuracies may compromise the delineation of adjacent structures and affect calculated dose. Here, we evaluate the un-editable body contours auto-generated by Ethos versions 1.0 (v1) and 2.0 (v2) treatment planning softwares for two simulated cases: weight-loss and bolus application, particularly important for head and neck radiotherapy patients.

Methods: A 3D-printed target structure was secured to the neck of an anthropomorphic phantom and sequentially covered with silicone boluses of uniform thickness, providing cases for bolus application (0.5 and 1 cm) and weight-loss (2.0, 1.5, 1.0, 0.5, and 0 cm). HyperSight CBCT images of the phantom were acquired to simulate the online adaptation process. Baseline body contours were manually produced and compared to those auto-generated in Ethos v1 (synthetic CTs) and Ethos v2 (synthetic CTs and direct calculation on HyperSight CBCTs). Additionally, the target volume D95% dose metric for weight-loss adapted plans generated by the Ethos v2 were analyzed as a function of surface layer thickness.

Results: The Ethos v1 body contour did not adapt adequately for the weight-loss image set [mean absolute volume deviation from baseline (MAD) = 205 cm³]. The weight-loss synthetic CT and HyperSight CBCT volumes in Ethos v2 were comparable to manually generated contours (MAD = 34 and 46 cm³ _, respectively); however, the bolus Hypersight CBCT body contour intersected the outer edge of the phantom (MAD = 157 cm³). The D95% deviation from the planned dose decreased by up to 10% when using the Ethos v2 adapted plan for the weight-loss scenario.

Conclusion: Contours in Ethos v1 rely on reference contours and deformable registration algorithms, whereas Ethos v2 does not. Hence, Ethos v2 is preferred for cases involving weight change. A tight-fitted air gap-free bolus is critical for achieving accurate body contours for Ethos v2 Hypersight CBCTs.

Tumor hypoxia significantly impacts the efficacy of radiotherapy. Recent developments in the technique of dose painting by numbers (DPBN) promise to improve the tumor control probability (TCP) in conventional radiotherapy for hypoxic cancer. The study initially combined the DPBN method with hypoxia-guided dose distribution optimization to overcome hypoxia for lung cancers and evaluated the effectiveness and appropriateness for clinical use of the DPBN plans. ¹⁸F-FMISO PET-CT scans from 13 lung cancer patients were retrospectively employed in our study to make hypoxia-guided radiotherapy. In the clinic, TCP and normal tissue complication probability (NTCP) derived from the DPBN plans in comparison to conventional intensity modulated radiation therapy (IMRT) plans were evaluated. Additionally, in order to investigate the improved clinical suitability, the robustness of DPBN plans in response to potential patient positioning errors and radiation resistance variations throughout the treatment course was assessed. The DPBN approach, employing voxelized prescription doses, led to an average increase of 24.47% in TCP, alongside a reduction of 1.83% in NTCP, compared to the conventional radiotherapy treatment plans. Regarding the robustness of the DPBN plans, it was observed that positional uncertainties were limited to 2 mm and radiosensitivity deviations were within 4%. The lung NTCP showed a 0.05% increase when the isocenter was moved by 3 mm in any direction, suggesting that the DPBN plan meets clinical acceptability criteria. Our study has shown that the DPBN technique has significant potential as an innovative approach to enhance the efficacy of radiotherapy for lung cancer with hypoxic regions.

Purpose: A novel proton beam delivery method known as DynamicARC spot scanning has been introduced. The current study aims to determine whether the partial proton arc technique, in conjunction with DynamicARC pencil beam scanning (PBS), can meet clinical acceptance criteria for bilateral head and neck cancer (HNC) and provide an alternative to full proton arc and traditional intensity-modulated proton therapy (IMPT).

Method: The study retrospectively included anonymized CT datasets from ten patients with bilateral HNC, all of whom had previously received photon treatment. The clinical target volumes (CTV) were categorized into three levels: CTV_7000, CTV_5950, and CTV_5600. IMPT plans included three beams, whereas DynamicARC plans included dual-partial-arcs (DPA), single-partial-arc (SPA), and single-full-arc (SFA). All plans underwent robust optimization considering setup (± 3 mm) and range (± 3%) uncertainties applied to the CTVs. DynamicARC plans were evaluated against the NRG-HN009 criteria and IMPT plans using various metrics.

Results: All four techniques-IMPT, DPA, SPA, and SFA-demonstrated substantial compliance with NRG-HN009 dosimetric criteria. DynamicARC produced superior dose conformity, lower hotspot, and improved homogeneity for high-risk CTV compared to IMPT, with comparable performance for intermediate- and low-risk CTVs. DynamicARC reduced the D_mean to the parotid glands by average differences of 14.5%-22.1% and to the oral cavity by an average difference of 15.75% compared to IMPT. DPA and SPA techniques achieved reductions in total integral dose of 3.7%-5.7% relative to IMPT. Overall, DPA yielded dosimetric results comparable to those of SFA while offering more conformal dose distributions and slightly better organ at risk sparing than SPA.

Conclusion: On the ProteusOne with a partial gantry system, DPA and SPA, in conjunction with DynamicARC PBS protons, provided clear dosimetric advantages over three-field IMPT. Future clinical implementation and further research into optimizing DynamicARC protocols are warranted to fully realize the benefits of these techniques in clinical settings.