Oxford University to Reduce Hospital Attendance, Healthcare Costs and Improve Tests’ Results thanks to ArangoDB
The Institute of Biomedical Engineering of the University of Oxford, in partnership with the Oxford University Hospitals NHS Trust funded by the National Institute for Health Research is developing a system for cardio-pulmonary patients.”
by Dario Salvi, University of Oxford
The six-minute walk test (6MWT) is a standard method for measuring exercise capacity in patients with cardiopulmonary disease such as pulmonary hypertension (PHT). The 6MWT measures how far a patient can walk in 6 minutes. By assessing patients’ ability to exercise, the 6MWT provides a global assessment of respiratory, cardiovascular, cognitive and neuromuscular function.
The test is usually performed in hospital, by walking along a hospital corridor, where many factors on the day of the test can affect patients’ performance, including tiredness, unfamiliarity with the environment or anxiety. The test typically requires 2 physiologists to monitor the distance walked by the patient, their oxygen saturations by pulse oximetry, and to record symptoms felt during the test.
The University of Oxford, in collaboration with the Oxford University NHS Trust, is developing a mobile-phone based system that would allow patients to perform their test at home.
The system includes a mobile phone “app” that guides patients through the test, both indoors or outdoors. In the outdoor scenario, the app makes use of GPS to track the user’s position. In the indoor scenario it uses the phone’s accelerometer to count the number of laps over a fixed-length walkway.
Undertaking a 6MWT test within the patient’s home environment, using mobile technology to transmit the results via a software application on a mobile phone is appealing as it is likely to be more acceptable to the patient, may provide more consistent and functionally relevant data, and may reduce hospital attendances and healthcare costs. By utilizing biometric collection devices such as a pulse oximeter monitor with Bluetooth capability, telehealth can be used to capture the oxygen saturation levels in the blood from patients and transmit them to a secure server, allowing them to be reviewed by cardiologists and other healthcare professionals.
The system comprises an app, compatible with both Android and iOS operating systems and a server-side solution. The server is in charge of storing the tests results and showing the data to physicians, through a web interface. The server is hosted inside the demilitarised network of the NHS trust.
The app was developed with the Apache Cordova framework and the server was with Node.JS.
ArangoDB was the database of choice. Several factors favoured this decision:
- it works well with Node.JS
- it’s distributed and scalable
- allows multiple data models (document, graphs, key-value)
- it’s secure
- it has a SQL-like querying language
The last feature was particularly appreciated because of the fact that the team includes researchers who need to explore the data efficiently for research purposes and without the need of developing complex APIs.
The server was developed with the Express.js framework. The team evaluated the possibility of using the Foxx application container offered by ArangoDB, but in this case decided to keep the database and the application separated.
“At the moment the database is used as a simple document storage with a thin REST API on top to add security. The querying language was particularly useful for creating views and summaries. The web interface of the DB was also very handy when data needed to be inserted or modified manually. In the future, as the research goes on, it is likely that we will explore some advanced features of the DB.”
The querying language was particularly useful for creating views and summaries.
— Dario Salvi, University of Oxford
At the moment, the team is just getting started with ArangoDB.
“We have appreciated the stability of the product and the fast response from the development team to solve issues or to help when new features were released.”
Big thanks to Dario Salvi for investing the time into writing and sharing this use case!