The cost to the economy of mental health illnesses in England was estimated to be £77.4 billion a year in 2003, of which bipolar disorder (BD) accounts for around £2 billion. The prevalence of BD is estimated at between 0.4-1.6% of the population.
The illness consists of recurring episodes of mania and depression and is very disruptive as a sufferer's daily life can be severely affected. Many people with BD self-monitor their condition in order to try and keep the disturbances from affective episodes to a minimum.
The PAM project aims to develop a self-help system for people with BD that can, monitor their behaviour patterns and identify the onset of a manic or depressive episode. To date, the PAM project has developed a system that performs behavioural monitoring in an unobtrusive manner and can detect changes in a person’s behaviour. The system uses a variety of discreet sensors to gather data on the parson’s behaviour and this data is processed to extract behavioural patterns and detect changes in those patterns.
Here we present one method of data processing that takes 24hr long data-streams from the sensors, over multiple days of recordings, pre-processes them and uses the Continuous Profile Model to align and extract the underlying 24hr patterns from the data-streams. We present some preliminary results from a technical trial.