Machine learning (ML) is the fastest growing field in computer science and healthcare, providing future benefits in improved medical diagnoses, disease analyses and prevention. In this paper, we introduce an application of interactive machine learning (iML) in a telemedicine system, to enable automatic and personalised interventions for lifestyle promotion. We first present the high level architecture of the system and the components forming the overall architecture. We then illustrate the interactive machine learning process design. Prediction models are expected to be trained through the participants' profiles, activity performance, and feedback from the caregiver. Finally, we show some preliminary results during the system implementation and discuss future directions. We envisage the proposed system to be digitally implemented, and behaviourally designed to promote healthy lifestyle and activities, and hence prevent users from the risk of chronic diseases.
According to the World Health Organisation (WHO) poor diet and physical inactivity are public health issues and their associated health problems are getting beyond the healthcare capabilities [25]. These issues are strong contributors to overweight and obesity epidemic, which escalates to chronic diseases in the long-term. Research shows that the risk of developing chronic conditions can be reduced by adhering to a healthy lifestyle (e.g., a balanced diet and sufficient physical activity). There is a shift towards scalable solutions to promote healthier lifestyles outside the clinical settings. User adherence to the assigned plan is an indicator of the effectiveness of healthy lifestyle program. Yet, promoting deliberate lifestyle is not straightforward and maintaining a change in behaviour is a hard task to achieve. On the other hand, relatively few diet and physical activity applications have been tested in research environment to determine their effectiveness in health promotion [7]. Moreover, a small segment of such programs consider the delivery of practical and empathic health behaviour change support by considering cognitive, emotional and behavioural aspects of behaviour change. There is a need for a tailored system feedback that goes in parallel and fulfils users' preferences, while keeping the interaction simple. Healthy lifestyle promotes optimal health and prevents health problems such as obesity and eating disorders [25]. This could also prevent long-term health problems, such as heart disease, cancer and stroke. There is a need to reinforce the adaption of long-term healthy eating behaviour. With this research we investigate formulating health promotion techniques with prioritisation based on user data. To create sustainable healthy lifestyle, personalised feedback is always favoured over a one-size-fits all approach. Using modern sensor technology and proper algorithms, we can detect if a user is active, neutral or passive, and show dimensions of data about their activities. However, to guide users along their journey and create awareness, commitment to lifestyle goals it is necessary to offer interactive coaching support. Thus, based on the user activity data acquired, the caregiver entails the delivery of practical and empathic health behaviour change support, which is more personal and responds better to user feelings. A human in the loop can be effective in all lifestyle promotion domains, including physical activity exercise and food intake [31]. This paper provides an overview of an interactive machine learning to classify lifestyle promotion data. We first consider the state of the art view of trends in the field. For example, systems with goals and actions intended for health and wellbeing and applies some form of machine learning techniques. We highlight the importance and challenges associated with this emerging trend in lifestyle promotion. Finally, we discuss the case of CoachMe [11] a bot and web application for lifestyle promotion to a subject by comprising data corresponding to objective behaviour. We discuss the integration of an interactive machine learning algorithm into the system to classify users based on their activity performance.
With the increasing burden of sedentary lifestyle and overweight on our health, promoting healthier lifestyle becomes a necessity to prevent people from escalating into chronic diseases, such as obesity and diabetes. Developing systems to promote health and provide valuable information about user’s habit becomes increasingly effective. Machine learning is a fast-growing trend in the healthcare domain since it has the potential to be a powerful tool for human empowerment, touching everything from how we eat to how we diagnose diseases. Moreover, this can help health experts to identify trends that can lead to improved diagnoses and treatment, such as patient’s health history and behavioural information data. Therefore, machine learning can identify aspects about user activities, such as behavioural pattern or the efficacy of the application. Understanding user preferences and observing their behaviour, we can interact with users at the right time, through the right channel, with the right tone, and the most relevant content. Machine learning can assist in developing more effective diagnoses and treatment, preventing prescription errors. This paper focuses on using a supervised machine learning algorithm to classify users and help caregiver to personalise their intervention feedback. Combined human interaction into machine learning algorithms, the interactive machine learning highlights how to optimise human effort in machine learning model training [28]. Moreover, its methods are useful to analyse human behaviour and deduce health and wellbeing information. Recognising specific human behaviour analysis, such as eating pattern, daily physical activity are extremely import for healthcare providers to understand how to support their patients [3].
This content is AI-processed based on open access ArXiv data.
