Changing behaviours with AI

 

RESEARCH: APA Innovation Advisor Barry Nguyen discusses a research collaboration investigating whether artificial intelligence can help to change human behaviour to achieve better health outcomes.

One of the biggest healthcare problems we are seeing today is the cost of managing chronic diseases and its associated economic burden. Many of these chronic diseases such as heart disease, lung cancer and type II diabetes can often be prevented or reversed by developing healthy habits and a positive mindset, for example, by cessation of smoking, eating well and exercising.

Physiotherapists are deeply involved in lifestyle and behavioural change as part of their daily clinical practice, and such interventions can be very challenging because a complex range of factors need to be considered.

It is important to note that scientific literature on behaviour change is diverse but also accumulating at a very fast rate. However, the literature is fragmented and not reported well and it can take many years to synthesise the data; so, advances in computing power can enable artificial intelligence to be applied to organise the data and produce new, useful insights about behaviour change.

A collaboration between a group of University College London behavioural scientists and IBM AI researchers are attempting to answer the question of whether AI can help change human behaviour to help achieve health goals and develop good habits.

The Human Behavioural Change Project is an online knowledge system that uses artificial intelligence, in particular natural language processing and machine learning, to extract information from intervention evaluation reports to answer key questions about the evidence. Systems such as these can help policymakers, researchers and practitioners get answers to questions related to ‘what works, compared with what, how well, with what exposure, with what behaviours (for how long), for whom, in what settings, and why?’

The challenge, noted in the research, is to find the right behavioural change techniques to help motivate individuals to develop better habits to improve their health. Examples of such techniques are goal-setting, and self-monitoring; therefore, if AI could assist in determining the best behavioural change techniques for specific scenarios, it could assist individuals, clinicians and policymakers to make more informed decisions on how to tackle this issue.

The AI system IBM is developing for behavioural change consists of:

  • information extraction algorithms that automatically key information from behaviour change intervention reports
  • machine learning and reasoning algorithms that generate suggested interventions for specific scenarios, and predictions of the likely outcomes of an untested interventions.

Both supervised and unsupervised learning algorithms are being applied to extract information from the research reports. Despite this project being in its early stages, there is confidence that the findings from this work will advance society’s understanding of behavioural change at both individual and whole-population levels, leading to better health outcomes for generations to come.

Implications for physiotherapists:

  • AI systems have the potential to be applied to physiotherapy research where scientific literature is vast and accelerating at a fast rate 
  • physiotherapists should be mindful of the advances in computing power, and how it can enable artificial intelligence to assist them in evidence-based practice and enhance their academic and clinical research capabilities.

For more information about this research visit ibm.com/blogs/research/2018/12/ ai-help-change-behaviour.

Email barry.nguyen@australian.physio with any comments or queries regarding this article.

Disclaimer: This material is intended for general information purposes only and does not constitute legal advice or meet the specific needs of your clinical context.

 

© Copyright 2018 by Australian Physiotherapy Association. All rights reserved.