Patient Cranial Angle and Intrafractional Stability in CyberKnife Robotic Radiosurgery: A Retrospective Analysis.

Purpose: The aim of this study was to investigate whether variations in cranial angles and treatment accuracy during CyberKnife robotic radiosurgery necessitate adjustment of the margins of the planning target volume. Patients and Methods: Data from 66 patients receiving CyberKnife treatment for brain tumors were retrospectively analyzed. Patients were immobilized using a thermoplastic mask and headrest. The cranial angle was measured on planning CT and patients were divided into 2 groups: ≤10° (Group A) and >10° (Group B). Intrafractional motion was recorded using the CyberKnife tracking system over 50 min. Translational and rotational errors were compared between groups, and planning target volume margins were calculated. Results: In Group A, significant translational error differences were found along with the X-axis over time (P < .02). In Group B, significant differences occurred along with the Z-axis (P < .03). No significant rotational or 3-dimensional vector differences were found in either group. Group A had significantly lower Y-axis (P < .045) and roll axis (P < .005) errors compared to Group B. Estimated planning target volume margins in Group A were 0.56 mm (X), 0.46 mm (Y), and 0.47 mm (Z). In Group B, margins were 0.62 mm (X), 0.48 mm (Y), and 0.46 mm (Z). Margins covering 95% of intrafraction motion were 0.49 to 0.50 mm (X, Y, Z) and 0.69 mm (3-dimensional vector) for Group A, and 0.48 to 0.60 mm and 0.79 mm for Group B. With a 1-mm margin, complete coverage was achieved in Group A while 2.1% of vectors in Group B exceeded 1 mm. Conclusion: Adjusting cranial angle to ≤10° during thermoplastic mask molding provided better or similar intrafractional stability compared to >10°.

Evaluation of the Position Error of Wearing Surgical Masks During Radiotherapy in Head and Neck Cancer Patients.

Purpose: Wearing a mask during the coronavirus disease 2019 epidemic (COVID-19) is a preventive way to reduce droplet and aerosol transmission. The purpose of this study was to evaluate the position error of wearing a surgical mask during radiotherapy in head and neck cancer patients. Patients and Methods: We collected and analyzed 2351 kV X-ray image records of 81 patients with head and neck cancer who underwent image-guided radiotherapy (IGRT). Patients with/without a surgical mask were divided into the head-neck (HN) mask group and head-neck-shoulder (HNS) mask group. The position error in the X (left-right), Y (superior-inferior), Z (anterior-posterior), 3D (three dimensional) vectors, as well as the pitch and yaw axes were compared between the four groups. Results: We found that patients wearing surgical masks in the HN mask group showed no significant differences in the mean position error of the different types of headrest (p>0.05). In the HNS mask group, only the type C headrest group showed significant differences (P < 0.05). The X axis values were -0.05±0.07 and -0.11± 0.01 cm (P = 0.04), and the pitch axis values were 0.34±0.29° and 0.83±0.08° (P = 0.01). Conclusion: The mean position error of most patients wearing surgical masks was not greater than patients without a surgical mask. Patients wearing while receiving treatment is a low-cost and easy-to-implement prevention method.

Using deep learning models to analyze the cerebral edema complication caused by radiotherapy in patients with intracranial tumor.

Using deep learning models to analyze patients with intracranial tumors, to study the image segmentation and standard results by clinical depiction complications of cerebral edema after receiving radiotherapy. In this study, patients with intracranial tumors receiving computer knife (CyberKnife M6) stereotactic radiosurgery were followed using the treatment planning system (MultiPlan 5.1.3) to obtain before-treatment and four-month follow-up images of patients. The TensorFlow platform was used as the core architecture for training neural networks. Supervised learning was used to build labels for the cerebral edema dataset by using Mask region-based convolutional neural networks (R-CNN), and region growing algorithms. The three evaluation coefficients DICE, Jaccard (intersection over union, IoU), and volumetric overlap error (VOE) were used to analyze and calculate the algorithms in the image collection for cerebral edema image segmentation and the standard as described by the oncologists. When DICE and IoU indices were 1, and the VOE index was 0, the results were identical to those described by the clinician.The study found using the Mask R-CNN model in the segmentation of cerebral edema, the DICE index was 0.88, the IoU index was 0.79, and the VOE index was 2.0. The DICE, IoU, and VOE indices using region growing were 0.77, 0.64, and 3.2, respectively. Using the evaluated index, the Mask R-CNN model had the best segmentation effect. This method can be implemented in the clinical workflow in the future to achieve good complication segmentation and provide clinical evaluation and guidance suggestions.

Comparison of Intrafractional Motion in Head and Neck Cancer Between Two Immobilization Methods During Stereotactic Ablative Radiation Therapy by CyberKnife.

PURPOSE: Maintaining immobilization to minimize spine motion is very important during salvage stereotactic ablative radiation therapy (SABR) for recurrent head and neck cancer. This study aimed to compare the intrafractional motion between two immobilization methods. PATIENTS AND METHODS: With a spine tracking system for image guiding, 9094 records from 41 patients receiving SABR by CyberKnife were obtained for retrospective comparison. Twenty-one patients were immobilized with a thermoplastic mask and headrest (Group A), and another 20 patients used a thermoplastic mask and headrest together with a vacuum bag to support the head and neck area (Group B). The intrafractional motion in the X (superior-inferior), Y (right-left), Z (anterior-posterior) axes, 3D (three-dimensional) vector, Roll, Pitch and Yaw in the two groups was compared. The margins of the planning target volume (PTV) to cover 95% intrafractional motion were evaluated. RESULTS: The translational movements in the X-axis, Y-axis, and 3D vector in Group A were significantly smaller than in Group B. The rotational errors in the Roll and Yaw in Group A were also significantly smaller than those in Group B; conversely, those in the Pitch in Group A were larger. To cover 95% intrafractional motion, margins of 0.96, 1.55, and 1.51 mm in the X, Y and Z axes, respectively were needed in Group A, and 1.06, 2.86, and 1.34 mm, respectively were required in Group B. CONCLUSION: The immobilization method of thermoplastic mask and head rest with vacuum bag did not provide better immobilization than that without vacuum bag in most axes. The clinical use of 2 mm as a margin of PTV to cover 95% intrafractional motion was adequate in Group A but not in Group B.

Comparison of Skull Motions in Six Degrees of Freedom Between Two Head Supports During Frameless Radiosurgery by CyberKnife.

Introduction: Maintaining immobilization to minimize skull motion is important during frameless radiosurgery. This study aimed to compare the intrafractional skull motions between two head supports. Methods: With 6D skull tracking system, 4,075 image records from 45 patients receiving radiosurgery by CyberKnife were obtained. Twenty-three patients used TIMO head supports (CIVCO) (Group A) and twenty-two patients used Silverman head supports (CIVCO) with MoldCare cushions (ALCARE) (Group B). The skull motions in X (superior-inferior), Y (right-left), Z (anterior-posterior) axes, 3D (three-dimensional) vector, Roll, Pitch and Yaw between the two groups were compared and the margins of planning target volume were estimated. Results: The translational motions in Group A were similar in three axes at initial but became different after 10 min, and those in Group B were less prominent in the Y axis. The rotational errors in Group A were most obvious in Yaw, but those in Group B were stationary in three axes. The motions in the X axis, 3D vector, Pitch and Yaw in Group B were significantly smaller than those in Group A; conversely, the motions in the Z axis in Group B were larger. To cover the 95% confidence intervals, margins of 0.77, 0.79, and 0.40 mm in the X, Y, and Z axes, respectively, were needed in Group A, and 0.69, 0.50, and 0.51 mm were needed in Group B. Conclusions: Both head supports could provide good immobilization during the frameless radiosurgery. Silverman head support with MoldCare cushion was better than TIMO head support in the superior-inferior direction, 3D vector, Pitch and Yaw axes, but worse in the anterior-posterior direction.