Automatic dietary monitoring using on-body sensors: Detection of eating and drinking behaviour in healthy individuals

In this work a novel concept is introduced, called automatic dietary monitoring (ADM). ADM attempts to replace the error-prone manual reporting of eating behaviour that is currently used in diet coaching programs. ADM-based diet coaching solutions are supported by the constant trend in electronic miniaturisation, allowing to embed sensors and computers in everyday objects, including clothing, accessories, and buildings. Systems that leverage this potential of pervasive computing can support their user with personalised health status and diet coaching services. The essential functions of ADM-based solutions are sensing and recognition of the user's eating behaviour. This work evaluates on-body sensing and pattern recognition solutions for ADM.