A hybrid sensor for motor tics recognition based on piezoelectric and triboelectric design and fabrication

With the gradual development of various artificial electronic skins and smart patches and other wearable electronic products, the collection of biomechanical energy to achieve self-powered sensing is critical to achieving the efficient function and sustainability of the system. In this work, a study of a novel hybrid sensor fabricated based on the piezoelectric and triboelectric design for motor tics recognition will be presented. A plant bionic and flexible hybrid self-powered sensor (PBHS) for motor tics recognition is reported. By combining bionic polydimethylsiloxane (PDMS) triboelectric nanogenerator and a layered stacked porous polyvinylidene fluoride-trifluoroethylene (PVDF-TrFE) nanofiber piezoelectric nanogenerator in mixing through near-field electrospinning (NFES) process on the flexible printed circuit board (FPCB) substrate, this enables the sensor which is a layer-by-layer stacked porous PVDF-TrFE nanofiber (LPPN) mainly composed of about 2500 piezoelectric polymers as a height of 2.2 mm thin wall to enhance energy harvesting characteristics. Compared with the original PVDF-TrFE nanogenerator, the voltage output performance is nearly reached to 5 V as a ~200% improvement. Furthermore, a self-powered tics recognition system has been developed through deep learning to provide doctors to observe the status of patients with motor tics of Tourette syndrome. By using the deep learning model of long short-term memory (LSTM) of a type of recurrent neural network (RNN), the overall sequences hybrid signal recognition rate for tic recognition has been achieved to 88.1%. • The new sensor enables to enhance energy harvesting characteristics. • A self-powered personal tic state recognition system for patient identification is explored for big data analysis. • Combining the deep learning LSTM model, the recognition rate of personal sequence tic signals reached 88.1%. • Opening up new ways for self-powered wearable electronic systems and bring huge opportunities for medical big data analysis.