Evaluation of stability of stereotactic space defined by cone‐beam CT for the Leksell Gamma Knife Icon

Abstract The Gamma Knife Icon comes with an integrated cone‐beam CT (CBCT) for image‐guided stereotactic treatment deliveries. The CBCT can be used for defining the Leksell stereotactic space using imaging without the need for the traditional invasive frame system, and this allows also for frameless thermoplastic mask stereotactic treatments (single or fractionated) with the Gamma Knife unit. In this study, we used an in‐house built marker tool to evaluate the stability of the CBCT‐based stereotactic space and its agreement with the standard frame‐based stereotactic space. We imaged the tool with a CT indicator box using our CT‐simulator at the beginning, middle, and end of the study period (6 weeks) for determining the frame‐based stereotactic space. The tool was also scanned with the Icon's CBCT on a daily basis throughout the study period, and the CBCT images were used for determining the CBCT‐based stereotactic space. The coordinates of each marker were determined in each CT and CBCT scan using the Leksell GammaPlan treatment planning software. The magnitudes of vector difference between the means of each marker in frame‐based and CBCT‐based stereotactic space ranged from 0.21 to 0.33 mm, indicating good agreement of CBCT‐based and frame‐based stereotactic space definition. Scanning 4‐month later showed good prolonged stability of the CBCT‐based stereotactic space definition.


| INTRODUCTION
Stereotactic radiosurgery (SRS) delivers a high dose of radiation to a target while sparing healthy structures, and this mandates precise localization. Traditionally, Gamma Knife SRS treats intracranial lesions and involves localizing the target coordinates based on an invasive frame fixed to the patient skull. 1 With the advances in image-guided radiotherapy, the possibility of localizing targets using images allows for noninvasive frameless stereotactic radiosurgery, as well as for fractionated stereotactic radiotherapy. The new Gamma Knife model, Leksell Gamma Knife â Icon TM , has been recently introduced and includes a cone-beam CT (CBCT) which can be used to define the 3D stereotactic coordinate space without the need for an invasive frame system. The CBCT can be used to define the stereotactic space coordinates for either G-frame treatments or the new frameless thermoplas- ies have looked at the accuracy of Gamma Knife delivery when using a thermoplastic mask system for skull immobilization with an IR camera and CBCT; 2 and at quantifying translational and rotational shifts when using the invasive frame on a prototype CBCT image-guided Gamma Knife Perfexion unit. 3 The purpose of this work was to evaluate the stability of the CBCT-based stereotactic coordinate space and confirm it is in agreement with the standard frame-based stereotactic coordinate system throughout a partial volume of the defined stereotactic space.

2.A | Leksell Gamma Knife â Icon TM system
Leksell Gamma Knife â systems have been designed by the manufacturer (Elekta Instruments, A.B., Stockholm, Sweden) to precisely deliver stereotactic treatments to intracranial targets. The latest design, the Leksell Gamma Knife â Icon TM , is identical in the core radiation unit to its predecessor, the Leksell Gamma Knife â Perfexion TM (i.e., 192 Co-60 sources distributed over eight sectors that can be moved independently to deliver an isocenteric treatment). The new model Icon, however, comes with a cone-beam CT (CBCT) system for image guidance and a couch-mounted infrared camera for intrafraction motion management, allowing for frameless thermoplastic maskbased stereotactic radiosurgery and fractionated stereotactic radiotherapy treatments. The CBCT system is composed of a rotating anode X-ray tube (RTM 75H, Industria Applicazioni Elettroniche, Cormano MI, Italy) and a 34 cm 9 39 cm flat-panel x-ray detector (Pixium CBCT 2630, Thales Electron Devices SAS, France) mounted on an arm allowing a 210 degree rotation for scanning. CBCT can be used for obtaining a reference image and determining the Leksell stereotactic space coordinates (hereafter called CBCT-based stereotactic coordinates), and also can be used prior to treatment delivery for verifying the actual skull position and determining translational and rotation shifts based on coregistration with the reference CBCT image 4 so that the shot positions are adapted to the target and plan dose distribution is recalculated. 5 The intrafraction motion management system (IFMM) is composed of an infrared camera (Polaris Vicra, Northern Digital Inc., Waterloo, ON, Canada) mounted on the far end of the patient couch and is used to monitor the movement of a reflector marker placed on the patient nose tip when using thermoplastic mask for immobilization, with the option of automatically stopping the delivery if the movement exceeds a threshold that can be set from 0.5 mm up to 3 mm. 6,7 2.B | CBCT stereotactic space definition and CBCT precision QA A special calibration tool is used by the manufacturer's service engineer to find the CBCT to Leksell coordinate transform between the uncalibrated CBCT image and the radiation delivery unit ( Fig. 1(a)).
The tool consists of six steel ball-bearings with known Leksell coor- dinates. An algorithm running in service mode uses the projection images of a CBCT scan of this special tool to calculate the transform between CBCT image and the Leksell coordinates systems. 8 This calibration procedure was performed once by the manufacturer's service engineer at the time of treatment unit commissioning.
The manufacturer also provides a user QA tool (QA tool Plus) to enable each user to test the CBCT precision ( Fig. 1(b)). This CBCT precision QA test is performed daily as part of a comprehensive QA

2.C | In-house marker tool
A simple tool with fixed fiducial markers that can easily be localized in both CT and CBCT scans was designed and implemented, similar to the tool used in a previous work, 10  CBCT images of the tool (Fig. 3(b)) were taken every working day for 6 weeks using the two predefined scanning settings: CTDI 6.3 mGy (high quality) preset at 90 kV and 25 mA; and CTDI 2.5 mGy (low dose) preset at 90 kV and 10 mA. The CBCT images were automatically imported to the LGP software upon scanning.
The pixel sizes for both the CT and CBCT images were 0.5 mm (X) by 0.5 mm (Y). However, when projecting these images in the LGP software, the pixel sizes are interpolated and became 0.1 mm (X) by 0.1 mm (Y).
The LGP software was used to determine the (X, Y, Z) and (X 0 , Y 0 , Z 0 ) coordinates of each marker for each image set of the CT and CBCT scans, respectively. The geometric center of each marker was determined by a single observer looking for the center of each "fuzzy" enhancement in maximum zoomed-in and maximum contrasted images, as shown in Fig. 4. To examine the reproducibility of determining the marker center, the coordinates of each marker were read three times in separate instances by the observer, and the average of the three readings was reported. The magnitude of vector difference (r) between the mean coordinates of frame-based and CBCT-based was calculated as: Furthermore, an additional CBCT of the markers frame tool was scanned 4 months later to determine the long-term stability of the CBCT-based stereotactic coordinate system. The in-house markers tool was also scanned 4 months later to determine the prolonged stability of CBCT-based stereotactic definition. The coordinates of the five markers were within the range of measurements performed in the initial 6-week study period, indicating good CBCT-based coordinate definition prolonged stability.

| CONCLUSION
A simple in-house tool was used to test the stability of the CBCTdefined stereotactic space in Gamma Knife Icon and its agreement with the standard frame-defined stereotactic space, independently from the manufacturer provided tool and methodology. CBCT-based stereotactic space definition in Gamma Knife Icon was found to be stable over a period of 4 months, and in good agreement with the standard frame-based stereotactic space definition.

ACKNOWLEDG MENTS
Ismail AlDahlawi acknowledges the financial scholarship support from King Fahad Specialist Hospital-Dammam.

CONFLI CT OF INTEREST
The authors have no conflicts of interest to disclose.