Deep Learning and Spiking Neural Networks for Neuromorphic Applications for Classifying Health Status Using Wearable and Wireless Systems

Advances in machine learning, such as deep learning and spiking neural networks that are representative of neuromorphic applications, are anticipated to substantially amplify the impact of wearable and wireless inertial sensor systems for the field of healthcare. Deep learning algorithms have already been successfully demonstrated with wearable systems for human activity recognition. In particular, discretized diagnostic states for healthcare may be discerned as distinctly recognizable human activities. Within the field of neuromorphic computing, which strives to realize more biologically representative processes that are more characteristic of neuronal activity, is spiking neural networks that constitutes the third generation of neural networks. The topics of deep learning and neuromorphic artificial intelligence respective of spiking neural networks are elucidated, in light of their potential to facilitate the advance of wearable and wireless systems for healthcare.