Title:
TouchIoT: Soft Tactile Sensing System for Application in Massage Training Process
Spekers:
Van Anh Ho
Massage techniques vary and rely on the therapist’s experience in applying suitable pressure to muscles through contact actions involving hands, fingers, and the patient’s skin to alleviate pain. Training individuals with empirical techniques and the therapist’s sensory perception is considered challenging. Therefore, there is a growing need for a database collecting contact information (location, force, depth) related to diverse massage actions. While vision can identify contact location, quantifying contact force or depth on the patient’s skin is challenging and costly due to the complexity of whole-tactile sensing. In response, we propose a cost-effective vision-based tactile system for outfitting patient phantoms (e.g., arms) with whole-body sensing, utilizing machine learning and VR techniques. Our system records therapist actions on the phantom skin, allowing for playback to trainees to enhance the massage training process. We conducted experiments with a large dataset to train massage actions, and a demonstration will be presented at the workshop.