Vicomtech is developing a machine learning (ML) based system for the detection of anomalies in the wellbeing variables of older persons participating in SHAPES project.

The system is based on unsupervised ML techniques and its main objective is to serve as a tool for aiding health professionals or caregivers in the detection of irregularities in the activity of daily living of these individuals.

The systems will incorporate different functionalities for:

i) giving a risk value to each of the individuals in order to detect health priorities,
ii) creating alerts for both low- or high-risk anomalies, or
iii) comparing a detected anomaly with other individual’s attributes in time.

For the visualisation of these anomalies, different types of graphs will be used, most of them giving the option to see the evolution of each variable in time, and to add more variables to the same graph to compare them and thus, i) better understand the origin, behaviour, or trend of the detected anomaly (see c) graph in the image), or ii) see the consequences of these irregularities (graph d) in the image).