An Event Polarized Paradigm for ADL Detection in AAL Context

The possibility to detect activities of daily living (ADLs), with particular regards to fall detection, is mandatory to implement a rigorous remote monitoring of frail people. Actually, unintentional falls cause a lot of hospitalizations and may lead to serious consequence due to long-lie happenings. Moreover, the ability to properly classify different kinds of falls could represent a strategic diagnostic tool. The widespread use of smartphones equipped with many embedded sensors would represent a suitable solution for ADL monitoring, especially for the future generation of elderly. The activity presented through this paper focuses on the development of a fully smartphone-based ADL detector, which uses an advanced classification paradigm to discriminate between different kinds of falls. The sensitivity and specificity features of the system are in line with the real needs of the Ambient Assisted Living context.