student profile: Mr Claudio Diaz Cifuentes


Map

Thesis work

Thesis title: Personalised Health Education using Human Centred Technologies Multimodal Data

Supervisors: Kalina YACEF , Judy KAY

Thesis abstract:

Obesity and sedentarity in children has increased in the last three decades (Ng et al., 2014). In order to reverse this trend, countries and organisations worldwide implement health education programs for seniors, adults and children, in order to promote behaviour changes and raise awareness with regards to diet and physical activity, two major factors linked to obesity and non-communicable diseases. Health education programs are likely to lead to positive health outcomes, especially when done in childhood (Cecchini et al., 2010), but effectiveness and efficiency varies greatly and the problems are still subsiding. .

The use of technology in health education programs regarding physical activity and dietary behaviours has shown that computer tailored education seems to have a better effect than other type of non-computer tailored interventions (Kroeze, Werkman, & Brug, 2006), but the subtle mechanisms underlying successful computer tailoring remains unknown. Other types of successful health interventions, for example smoke cessation, have identified that these particular mechanisms in tailored messages are personalization, adaptation or feedback (Dijkstra, 2005).

Furthermore, health education programs in physical activity behaviours increasingly use unobtrusive and ubiquitous human centered technologies, especially accelerometers, to measure interventions effectiveness. These technologies provide a new source of data about physical activity that is objective, continuous and precise (rather than self-reported, discrete and approximate). , This new source of data can be used as evidence of learning (or change of behaviour) and used for a more refined personalisation, especially when combined with other sources of learning data for example to measure more complex learning tasks (Blikstein & Worsley, 2016) and discover (complex) learning behaviours (Hung & Zhang, 2008).

The challenges to work with these data are how capture different modalities, how process them and then how integrate and analyze them to have effectiveness and impact on learning (Ochoa, 2017).

In this research we want to answer the question, how can multiple sources of data analytics leverage personalization in physical activity eHealth programs?

Selected publications

Download citations: PDF RTF Endnote

Conferences

  • Diaz, C., Yacef, K. (2018). Detecting behaviour changes in accelerometer data. 3rd International Workshop on Knowledge Discovery in Healthcare Data (KDH@IJCAI-ECAI 2018), Stockholm: CEUR-WS.

2018

  • Diaz, C., Yacef, K. (2018). Detecting behaviour changes in accelerometer data. 3rd International Workshop on Knowledge Discovery in Healthcare Data (KDH@IJCAI-ECAI 2018), Stockholm: CEUR-WS.

Note: This profile is for a student at the University of Sydney. Views presented here are not necessarily those of the University.