An event polarized paradigm for ADL detection in AAL context

Abstract

The monitoring of Activities of Daily Living (ADL) is a fundamental task to implement a rigorous remote monitoring of weak users with particular regards to falls. Actually, unintentional falls cause a lot of hospitalizations and could produce serious consequence due to long-lie happenings. The widespread use of smartphones associated with the wide variety of embedded sensors would represent a suitable solution for ADL monitoring. This paper faces the development of fully smartphone based ADL detector which uses advanced signal processing to provide useful and reliable information for the efficient implementation of remote elderly monitoring. The developed strategy allows for ADL classification with sensitivity and specificity features in line with real applications in Ambient Assistive Living (AAL) context.

Is Technology Present in Frailty? Technology a Back-up Tool for Dealing with Frailty in the Elderly: A Systematic Review

“…Ando et al, 2015[62] carried out a laboratory study in which they used all possible capacities of a smartphone to not only detect the risk of falling but also to discriminate between the different types of fall in frail individuals. The fall detection methodology was based on an acceleration and orientation analysis, and the information gathered by the smartphone sensors was processed via threshold algorithms (TAS).…”

An Event Polarized Paradigm for ADL Detection in AAL Context

“…Andò et al [16] presented a smartphone-based platform aimed to provide effective solutions for ADL detection in AAL contexts. The main task of the assistive system is to acquire awareness of common ADLs such as stair negotiation, sitting, and falls to provide the frail user with a suitable degree of assistance.…”

“…With respect to [16], in this paper, an improvement of the classification methodology is described and assessed by a larger set of users. The aim of the methodologies proposed is the classification of an unknown ADL event as belonging to one of the following classes: forward fall (FF), backward fall (BF), lateral fall (LF), sitting down (SI), stair negotiation, and LD.…”

“…Actually, as it will be clarified in Section II-A, the use of a larger data set coming from a larger set of users (with respect to [16]) entails the generation of more reliable signatures.…”

A Novel Unsupervised Abnormal Event Identification Mechanism for Analysis of Crowded Scene

“…This process of event detection becomes further more challenging if the criticality of the event has to be determined. From application viewpoint, it is essential for the application to take decision based on the context of the scene, which is another challenging task [9]. It is because there are multiple forms of uncertainty associated in extracting contextual information from the event detection.…”