Analysing Smart Home Data to Identify Declines in Health and Risk of Falls
Location: Melbourne, VIC
Duration: 5 months
Proposed start date: August 2019
Umps Health has developed a smart home platform to support older Australians live safely and independently at home. The system uses smart plugs to measure a persons’ interaction with their existing home appliances, like when they boil the kettle, open the refrigerator or cook using the microwave. Umps Health’s analytics platform learns what the person’s normal daily routine looks like over 30-60 days, then looks for subtle deviations in activity which indicate a decline in health or a heightened risk of an incident. If anything significant is detected, a family member or carer is alerted by text message and through a mobile app.
Umps Health’s Smart Home is used by older Australians throughout the country, and is already being provided by a number of aged care providers under Federal Government funding mechanisms. Umps Health now has one of the largest data sets of appliance use correlated to changes in health and wellbeing, and wants to utilise this data to improve the accuracy and predictive capabilities of its analytics platform.
Research to be Conducted
This research project involves the design and development of an analytics platform that will analyse a user’s patterns of behaviour in real-time to identify risk. The analytics platform will make use of Umps Heath’s extensive data set. While Umps Health has an existing analytics solution in place, we will encourage and support you to explore new approaches. Key elements of the project include
- Researching and developing techniques to analyse users’ patterns of behaviour in real time and identify risk,
- Developing new ways to visualise and interpret behavioural data,
- Working with our engineering team to integrate new data analysis techniques and visualisations into our solution,
- Collaborating with clinicians, researchers, and healthcare providers to incorporate clinical insights into our data analytics solution.
During this project, you will have the opportunity to test your research in the field with our test-user group, and demonstrate tangible benefits to older Australians and their families.
We are looking for a PhD student with the following:
- A tertiary qualification in mathematics or statistics OR an applied science where you have worked with large cross section and time-series datasets OR demonstrated mathematical and analytical skills, including knowledge of statistical and mathematical modelling techniques
- In-depth knowledge of appropriate data-science tools and techniques
- Proficiency in Python and SQL (preferred)
- Proficiency in ML / AI techniques, including supervised machine-learning, unsupervised machine learning, time series analysis, outlier detection, survival analysis etc. (preferred)
- Experience working with behavioural and health data (preferred)
We anticipate that the successful candidate will develop new tools, methods and processes that improve the accuracy (specificity and sensitivity) of the Umps Health Smart Home at detecting declines in health and wellbeing. These tools will be deployed by our in-house development team as part of the next release of the Umps Health Analytics Platform.
The intern will receive $3,000 per month of the internship, usually in the form of stipend payments.
It is expected that the intern will primarily undertake this research project during regular business hours, spending at least 80% of their time on-site with the industry partner. The intern will be expected to maintain contact with their academic mentor throughout the internship either through face-to-face or phone meetings as appropriate.
The intern and their academic mentor will have the opportunity to negotiate the project’s scope, milestones and timeline during the project planning stage.
31 July 2019
APR – 1065