Inference of Hygiene Behaviours While Recognising Activities of Daily Living

Many health problems are generally caused by unhealthy behaviours that occur whilst conducting everyday Activities of Daily Living (ADL), such as poor use of sanitation and hygiene. This paper describes the development of an ADL inference engine, which is able to recognise natural hygiene behaviour patterns. As opposed to traditional ADL classification approaches, the developed inference engine employs a novel hierarchal structure for the modelling, representation and recognition of the ADLs, its associated tasks, objects, dependencies and their relationships. The organisation of this information in a contextual structure plays a key role in carrying out robust ADL recognition for the detection of hygiene behaviours. The proposed work also marks a shift in feature detection methodology, as it allows actual behaviour to be studied in its natural environment within actual households, with at least two individuals per household as opposed to a laboratory based controlled setting. This paper also presents experimental results that validate the performance of the inference engine.