The personalization of medical devices has made considerable progress in recent years. In the orthopedic field, following a typical reverse engineering approach, 3D scanning of anatomical regions of interest is the starting point for various kind of customized manufacturing, followed by the design of the device and its fabrication, particularly with additive manufacturing techniques. In this context, particularly challenging issues emerge from the customization of hand orthotic devices. The present work focuses on the identification and comparison of suitable scanning procedures in order to acquire accurate hand and fingers 3D models. Two different types of structured light scanners (static and real-time) were compared and several configuration of the hand were studied and acquired, allowing to evidence strength and weakness of the various approaches, while keeping in consideration the target application. In particular, the issues related to the presence of involuntary movements during acquisition are considered and possible solving approaches indicated.