Dec 3, 2024
4:30pm - 4:45pm
Hynes, Level 3, Room 302
Laura Ferrari1,2,Marina Galliani1,Michele Foggetti1,Lorenzo Amati1,Simona Crea1,Francesco Greco1,3
Scuola Superiore Sant’Anna1,INRIA2,Graz University of Technology3
Laura Ferrari1,2,Marina Galliani1,Michele Foggetti1,Lorenzo Amati1,Simona Crea1,Francesco Greco1,3
Scuola Superiore Sant’Anna1,INRIA2,Graz University of Technology3
Tattoo electrodes are made by printing PEDOT:PSS onto commercially available temporary tattoo paper. The result is an ultra-thin and conformable dry electrode that can record high-quality surface electrophysiological signals. Tattoo electrodes have been compared with standard Ag/AgCl electrodes in many applications, for biomonitoring, and their internal structure and signal transmission mechanism have been fully detailed. [1]<br/>We have started translating tattoo technology into real-life applications, showing their use in biomonitoring.<br/>We have interfaced tattoos with multiple wearable devices [2], and we have used a learning approach (an autoencoder trained for one-class classification) on EEG tattoo data to identify the optimal wearable setup for alpha wave detection. [3]<br/>Now we are moving a step further in translating this technology into biorobotics applications.<br/>We have applied tattoo electrodes under lower and upper-limb exoskeletons. The lower-limb exoskeleton used in this study is a robot that transfers assistive torques during hip flexion and extension through a human-robot interface (HRI) at the thigh level [4]. Whereas the upper-limb exoskeleton uses arm cuffs to transfer assistive torque to the user to help the shoulder flexion and extension [5]. Tattoos have been developed as bipolar electrodes and their interconnections have been optimized to reach the robot via a Serial Peripheral Interface (SPI) allowing synchronous and full-duplex data transmission. Thanks to the ultrathin and conformable nature of tattoos, EMG signals have been acquired in direct contact with the HRI, where the electrodes are subjected to high mechanical stress. Such an approach opens novel control algorithms that, by leveraging the information coming from muscles, can be adapted for locomotion recognition and generation of torque profiles. This allows for real-time, adaptive responses of exoskeletons and personalized assistance, crucial in rehabilitation, industrial tasks, and assistive devices. We have developed tattoo high-density electromyography (HD-EMG) matrices to study manipulation. The interconnections have been improved to enable a stable and reliable link with wearable electronics, guaranteeing high signal quality. Conformable HD-EMG matrices ensure a drastic reduction of movement artifacts, typical of other dry interfaces, and permit the reaching of complex anatomical regions. Thanks to these latest developments we investigate novel application scenarios of HD-EMG in free-moving humans for the understanding of finger movements and forces, related to manipulation.<br/><br/>References<br/><br/>[1] Ferrari, Laura M., et al. Multifunctional Materials 3.3 (2020): 032003.<br/>[2] Taccola, Silvia, et al. Sensors 21.4 (2021): 1197.<br/>[3] Ferrari, Laura M., et al. EMBC, IEEE, (2021).<br/>[4] H. Eken et al., IEEE RAS EMBS, BioRob (2024), accepted.<br/>[5] L. Grazi, et al. in IEEE Trans. Neural Syst. Rehabil. Eng. (2020).