Purpose: To develop an accurate, portable, efficient, and cost-effective system for SGRT-based patient alignment using mixed-reality (MixR) visualization.
Methods: The game engine Unity was used to develop HoloMatch, a fully featured MixR application for the Microsoft HoloLens 2 head-mounted display system. The application allows a user to simultaneously and directly view a patient and a reference hologram derived from their simulation CT scan. The hologram is registered to the RT device; Thus, by matching a patient to their hologram, they are registered to the RT device. A holographic menu system was designed to navigate the application using hand gestures and voice control. Furthermore, a novel technique was developed to assess patient-hologram registration using raycasting. Usability and safety factors were assessed in relation to clinical workflows by RT staff, and a comparative analysis was performed with current SGRT methods based on accuracy, cost, portability, ergonomics, and field-of-view (FOV). IRB approval was requested to pilot the application in a first-in-human clinical trial.
Results: Accuracy: MixR, 3-5mm; SGRT, sub-mm. Cost: MixR, $3.5K per system; SGRT, $150-300K per system. Portability: MixR, inherently portable; SGRT, requires fixed sensors. Ergonomics: MixR, allows user free navigation to align patient via natural viewing perspectives; SGRT, utilizes a traditional 2D display and requires gaze redirection. FOV: MixR, ability to view entire reference surface including immobilization devices, no obstruction issues; SGRT, limited to anterior/lateral surfaces visible to fixed sensors, difficulty extracting surface from certain skin tones, sensors may be obstructed during alignment. IRB approval granted for study to include 12 photon and 12 proton patients.
Conclusion: MixR visualization offers a new way to perform SGRT-based patient alignment that dramatically improves access to the technique by decreasing cost and improving portability. A fully-featured application was developed to clinically implement the method and is currently being piloted in a first-in-human clinical trial.
Funding Support, Disclosures, and Conflict of Interest: This work is funded by the Florida Breast Cancer Foundation.