Moving the mountain: Non-conventional human body monitoring to enable energy harvester powered systems

Wearable sensors are highly miniaturised electronic devices that are easily placed on the human body and can conveniently monitor a range of body parameters, facilitate diagnosis of diseases, and even automatically raise alarms to summon help in critical situations. They are quickly emerging as next generation devices for the prolonged monitoring of the human body and have substantial impacts for managing our ageing population: both in terms of automatic monitoring and alarm generation in older subjects; and in promoting proactive and preventative healthcare and exercise in younger subjects. However, delivering new sensors, enhancing their functionality, and maintaining battery life as we incorporate more power hungry complex electronics is a major engineering challenge.

To resolve this it is crucial that future devices include energy harvesting: here the batteries present are supplemented by using the intrinsic energy available in the environment to power the sensor. As a result, energy harvesting is the only method for creating sensors which are power autonomous and can go beyond the limited lifetimes provided by batteries. However, state-of-the-art miniature energy harvesters which are suitable for wearing on the human body can only provide very small amounts of output power. To realise the full potential benefits from these it is necessary to devise new techniques for closing the current gap between the power required by wearable sensors and the power provided by miniature energy harvesters.

This project will bridge this gap by performing human body monitoring in non-conventional places on the body. Current approaches place the energy harvester where the wanted physiological signal is strongest; this project will devise techniques to move the physiological monitoring to where the most power can be harvested. Measuring physiological parameters in non-conventional locations on the body, such as the arms and legs, means the signals are weaker and more prone to interference from artefacts due to motion. It also means that substantially more energy is available for harvesting and to power both the sensor and the signal processing required to correct for the presence of artefacts.

The research will investigate these new trade-offs that are available as wearable sensors begin to include energy harvesters. In particular it will establish the trade-off between: physiological signal strength, motion artefact corruption of the collected signal, and energy harvesting potential. This will involve collecting physiological data from non-conventional places on the body and creating digital signal processing approaches for analysing the signals and dealing with the new motion artefacts compared to conventional monitoring. The successful outcomes will contribute to our knowledge on digital signal processing and enhance our ability to create self-powered sensors that can perform complex signal processing. This will provide major benefits to future wearable sensors, to personalised and preventative healthcare, and to our ability to tackle the healthcare, societal, and personal implications of the UK's ageing population.

Society and economy
The ageing population is one of the major challenges currently facing the UK. The proposed research aims to directly tackle this through the creation of new wearable devices for preventative healthcare, helping to keep people healthy for longer. As a result the successful outcomes of this work will have a direct impact on quality of life, and the PI already has experience of such quality of life impacts through his work on EEG for improving epilepsy diagnosis.

To reach the general population through translation and commercialisation it is highly likely that patentable IP will be produced during this work. (In the areas of non-conventional human body monitoring, reduced reliance on batteries/improved power autonomy, and advanced signal processing.) This IP will be protected, managed and exploited in collaboration with UMIP (University of Manchester Intellectual Property) in a manner consistent with other research programmes.

Further, while the field of wearable medical devices is very timely and relevant, it is still at an early stage of development and there are many open questions with regards to the requirements, regulations, and policy surrounding the devices. For example, if a sensor node relies on non-constant power sources and energy harvesters, does this affect its use in safety critical situations? The PI will use the research carried out here as a platform to engage with policy processes, informing the debate on wearable devices in personalised and preventative healthcare through the release of a position paper.


Knowledge
While the current proposed research focuses on contributing to personalised and preventative healthcare, the successful outcomes also have clear potential to impact the Big Data community: novel low power sensor nodes are the principal driver behind Big Data as they are enabling new streams of information to be collected. The proposed research will further the state-of-the-art in sensor nodes and data collection techniques and the PI will engage with the Big Data community to share these.

The PI is well positioned to deliver this impact, being a member of the Big Data research community at the University of Manchester and through his conference organisation roles. In 2012 he organised a mini-symposium at the IEEE Engineering in Medicine and Biology conference on "Dealing with lots of data: Databases, performance metrics and competitions as tools for research".

To directly impact the Big Data community the physiological data collected in this study will be made publicly available, providing a rich new data source for use by others in data mining experiments.


People
The work will contribute to building a critical mass of Engineering skills in the UK to tackle the ageing challenge. The PI will continue his public outreach activities to enable this, incorporating the results from this research. This will also allow skills and knowledge dissemination beyond his immediate research team.

For tractability within the first grant scheme the research is targeted for preventative healthcare in younger populations and heart monitoring, but huge potential is present for mapping the research to new applications and other types of physiological monitoring. As a result this research will have a large number of impacts, and will provide a foundation for providing further impact in the future.