Forecasts suggest that there will be more than 27.1 billion Internet-connected devices (things) worldwide by the end of 2025. Application domains of the Internet of Things (IoT) span from domestic to industry, including but not limited to intelligent healthcare systems. These IoT systems are mainly composed of low-cost embedded systems with varying complexities and the capability to sense and act. In healthcare IoT systems, things are in the form of wearable devices (e.g., body sensors, smartwatches, etc.), personal devices (e.g., smartphones), environmental sensors (e.g., temperature, humidity, etc.), cameras, IoT assistants, and many others. Their primary role is to gather information to monitor and eventually support clinicians in diagnosing and treating health conditions. Things collect and manage data from the environment when both explicit (active) and implicit (passive) human-to-thing interactions occur.
This research project focuses on care for older people, one of the main aspects of IoT healthcare. The project aims to create an innovative method to acquire and fuse data collected from things in an IoT healthcare setting and explore how older individuals perceive the use of IoT technology so that trustworthy interactions can be established. Trust can be defined as “the belief that another party will behave according to a set of well-established rules and will thus meet one’s expectations”. Evaluating and computing trust by determining older individuals’ behaviours can allow the IoT system to recognise abnormal behaviours. This would help carers and/or clinicians to provide the required help and support whenever specific situations occur, e.g., an injury has occurred, a particular disease is getting worse, threatening behaviours from malevolent individuals, etc.
- To develop a simple and quick relevant task based on physical activities that older individuals can perform in a lab setting so that their behaviours can be assessed.
- To collect data during this task using commodity IoT devices.
- To assess how older individuals perceive the IoT system to ensure that IoT systems will adapt to their behaviours, not vice versa.
- To use statistical methods and Artificial Intelligence to extract the required information to determine individuals’ behaviours and therefore recognise abnormal behaviours.
- To validate the results obtained through data analysis techniques. This can involve motion capture and sensor-based data collected in a lab setting to validate, for example, the jerkiness of motion, hand grip strength, etc.
- To assess the use of the developed system.
- The final step is the implementation of the algorithm into an IoT system.
Several types of data will be collected synchronously, such as personal data (e.g., smartphone), vital signs acquisition (e.g., smartwatch), tone of voice (e.g., microphone), movements and facial expressions (e.g., cameras), environmental changes (e.g., luminosity), etc. The main challenges are data fusion and information extraction to determine the behaviours of humans performing physical activities for which we will use Artificial Intelligence. The jerkiness of motion, hand grip strength, balance tests, etc., will allow validation using data collected from motion and sensors. Finally, the created algorithm will be tested in a cohort of older individuals to determine its applicability to IoT scenarios.
Training will be provided throughout the study in several ways. Project-specific hands-on training will be provided by the supervisory team and colleagues as needed and following regular Development Needs Analysis. This will include lab inductions, health and safety training, seminars, outreach opportunities and journal clubs. The student will be engaging intensively with the Doctoral Training Network which provides additional training. As a member of the Liverpool Doctoral College, a wide range of additional training resources will be available. The student will have regular (at least monthly) formal meetings with the supervisory team and yearly meetings with two assigned Academic Advisors.
The University is fully committed to promoting equality and diversity in all activities. In recruitment, we emphasise the supportive nature of the working environment and the flexible family support that the University provides. The Institute holds a silver Athena SWAN award in recognition of its ongoing commitment to ensuring that the Athena SWAN principles are embedded in its activities and strategic initiatives.
We are looking for a self-motivated candidate with a 2.1 or 1st class degree in engineering, computer science or with equivalent relevant expertise. The candidate should have analysis, programming and signal/image processing skills, ideally in Artificial Intelligence, and be keen to learn human activity recognition and basic biomechanical techniques, be willing to work with human volunteers needed for the study.
Please send your CV and covering letter to email@example.com, please include Technologies for Healthy Ageing in the subject.
Expected interviews March 2023.
Open to students worldwide
This PhD studentship will be granted on current UKRI levels of support with a UK home fee rate; a stipend, bench fees and full tuition fees will be provided. International students are welcome to apply but would need to make up the difference between the home and overseas fee rate.
Selis V and Marshall A (2019). A Classification-Based Algorithm to Detect Forged Embedded Machines in IoT Environments. IEEE Systems Journal, 13(1), 389-399. doi:10.1109/JSYST.2018.2827700
Cudejko T, Gardiner J, Akpan A and D’Août K (2020). Minimal shoes improve stability and mobility in persons with a history of falls. Scientific Reports, 10, 21755.
Luperto M, et al. (2022). Integrating Social Assistive Robots, IoT, Virtual Communities and Smart Objects to Assist at-Home Independently Living Elders: the MoveCare Project. International Journal of Social Robotics, 1-31.