Ultrageluid-gebaseerde bewegingsregistratie van het os scaphoideum voor computerondersteunde orthopedische chirurgie

Scaphoid fractures are the most common carpal fractures and are frequently treated using percutaneous scaphoid fixation (PSF), a minimally invasive procedure performed under fluoroscopic guidance. However, fluoroscopy provides only two-dimensional views of a complex three-dimensional wrist anatomy and exposes patients and staff to ionizing radiation. In addition, intraoperative manipulation can induce scaphoid motion relative to the wrist, degrading image-to-patient registration and challenging CAOS approaches that assume immobilization. These limitations motivate radiation-free guidance capable of tracking scaphoid motion in real time while maintaining reliable acoustic coupling on the curved, mobile wrist surface. This PhD project developed and validated a wearable, flexible ultrasound solution for motion-tolerant scaphoid tracking during PSF within a CAOS framework. A dual linear flexible array was designed via a parametric k-Wave simulation study using CT-derived wrist anatomy and implemented as a prototype with two parallel 64-element arrays. The system streams dual-plane ultrasound images, tracks the scaphoid using template matching at 28 Hz, reconstructs a 3D surface point cloud, and registers it to a CT-derived scaphoid model using iterative closest point. Validation against a synchronized 4D ultrasound reference registered to the same CT model yielded a median translation error of 0.48 mm and a median rotation error of 3.90 degrees, with limited drift. Finally, a synchronized multimodal dataset combining preoperative CT, wearable ultrasound, and multi-view near-infrared video acquired under a clinically inspired PSF workflow was introduced to support reproducible development and benchmarking of motion-aware CAOS algorithms. Together, these contributions establish the feasibility of wearable, radiation-free scaphoid tracking within CAOS and provide a foundation for future development of clinically deployable motion-aware navigation systems for wrist surgery