Dr Bryn Jeffries

GradCertEd (Higher Education), M.IT, D.Phil (Laser Physics), M.Phys
Research Associate, School of Information Technologies

J12 - The School of Information Technologies
The University of Sydney

Biographical details

Dr Bryn Jeffries is a Physicist and Computer Scientist, with experience in industry and academia. He works in the School of Information Technologies at the University of Sydney, researching database systems and teaching several database units. Bryn also provides database infrastructure and expertise to the CRC for Alertness, Safety and Productivity.

Research interests

Researchers often collect large amounts of data without yet knowing where the value within it lies. They often then need to review at least some of this data manually in order to reveal its value, which can be a slow and sometimes subjective process. Dr Bryn Jeffries' research looks at more efficient means of extracting new and useful information from research and other types of data.

"Data science - the field that I work in - offers so many opportunities to work with data to answer complicated problems in a wide range of domains.

"For example, I'm currently working with data from the CRC for Alertness, Safety and Productivity, which aims to improve alertness and therefore safety in the workplace and during commutes, in part by improving people's sleep. The data comprises recordings from people with sleep disorders such as insomnia and sleep apnoea.

"To diagnose a person's condition, that person typically has to make multiple overnight visits to a sleep clinic, where they are monitored so that researchers can understand their brain and breathing activity during sleep, among other things. This data then needs to be analysed in order to decide what treatment is required. As well as being very time consuming, this process can also be rather subjective.

"We wanted to be able to be able to diagnose people and determine the best form of treatment without their having to make so many trips to the clinic. So we're collecting data outside the clinic, such as with activity meters and heart-rate monitors, and using machine learning techniques to identify different types of abnormal breathing from recordings of people while they're sleeping.

"This automated approach can provide faster and more consistent assessment than the traditional approach.

"This is just one example of data science in action. I love it because I love problem solving, and in this research I get to help solve real problems that affect millions of people around the world.

"I've been working at the University of Sydney since 2010, and have been a researcher here since 2014. I've been very fortunate to benefit from the experience of many wonderful people here. There is a great community of researchers across the whole University who are very willing to explore new ideas."

Teaching and supervision

Units of Study

Selected past Capstone, SSP/TSP and Honours topics

  • Feature Extraction from Audio-Visual Fatigue Studies, Patricia Lor, Honours 2015
  • EEG Analysis Pipeline: Exploratory Cluster Analysis on Insomnia Disorder, Stephen McCloskey, Honours 2016
  • Classifying Sleep Apnea: Tools and Techniques for classification in time-series data, Rafael Mazzoldi, Honours 2016
  • Predicting Subjective Sleep Quality Using Wearable Device Data, Sonali Kalthur, Capstone 2016
  • Detecting sleep in Insomniacs from wearable data, QiChang Feng, Capstone 2016
  • Analysing Actigraphy and Cogstate Capture Data for the Alertness CRC Insomnia Study, Nicholas Bosch, SSP 2016
  • Automated Processing of Light and Circadian Phase Data, Xenia Boynton, Capstone 2017
  • Predicting real-time Melatonin Amplitude from Ambulatory Monitoring via Machine Learning, Mohamad Elagha, Capstone 2017
  • Decomposition Of Recurrent Neural Networks For Prediction Of Circadian Phase, Siyi Chen, Honours 2018

Current research students

Project title Research student
Data mining sleep data for sleep disorders classification and sleep/wake detection Rim HAIDAR
Sleep Disorder Analysis Using Machine Learning Techniques Stephen MCCLOSKEY

Current projects

  • Deep learning techniques for astrometry, Michael Rizzuto, Honours 2018
  • Prediction of PSG metrics from smart shirt data, Nahian-al Hasan, SSP 2018

Associations

  • Institute of Physics

Selected grants

2017

  • Practical Databases: Managing and Analysing Data with SQL (Open Learning Environment - Undergraduate); Roehm U, Jeffries B; DVC Education/Small Educational Innovation Grant.

2014

  • Enhancing feedback in online assessment of computer programming; Jeffries B, Koprinska I, Viglas A, Gramoli V; DVC Education/Large Educational Innovation Grant.

Selected publications

Download citations: PDF RTF Endnote

Conferences

  • Haidar, R., Koprinska, I., Jeffries, B. (2017). Sleep Apnea Event Detection from Nasal Airflow Using Convolutional Neural Networks. The 24th International Conference On Neural Information Processing (ICONIP 2017) (proceedings part V), Cham: Springer. [More Information]
  • McBroom, J., Jeffries, B., Koprinska, I., Yacef, K. (2016). Exploring and following students' strategies when completing their weekly tasks. The 9th International Conference on Educational Data Mining (EDM 2016), Raleigh: International Educational Data Mining Society.
  • Gramoli, V., Charleston, M., Jeffries, B., Koprinska, I., McGrane, M., Radu, A., Viglas, A., Yacef, K. (2016). Mining autograding data in Computer Science Education. Eighteenth Australasian Computing Education Conference (ACE 2016), New York: Association for Computing Machinery (ACM). [More Information]
  • McBroom, J., Jeffries, B., Koprinska, I., Yacef, K. (2016). Mining behaviors of students in autograding submission system logs. The 9th International Conference on Educational Data Mining (EDM 2016), Raleigh: International Educational Data Mining Society.

2017

  • Haidar, R., Koprinska, I., Jeffries, B. (2017). Sleep Apnea Event Detection from Nasal Airflow Using Convolutional Neural Networks. The 24th International Conference On Neural Information Processing (ICONIP 2017) (proceedings part V), Cham: Springer. [More Information]

2016

  • McBroom, J., Jeffries, B., Koprinska, I., Yacef, K. (2016). Exploring and following students' strategies when completing their weekly tasks. The 9th International Conference on Educational Data Mining (EDM 2016), Raleigh: International Educational Data Mining Society.
  • Gramoli, V., Charleston, M., Jeffries, B., Koprinska, I., McGrane, M., Radu, A., Viglas, A., Yacef, K. (2016). Mining autograding data in Computer Science Education. Eighteenth Australasian Computing Education Conference (ACE 2016), New York: Association for Computing Machinery (ACM). [More Information]
  • McBroom, J., Jeffries, B., Koprinska, I., Yacef, K. (2016). Mining behaviors of students in autograding submission system logs. The 9th International Conference on Educational Data Mining (EDM 2016), Raleigh: International Educational Data Mining Society.

For support on your academic profile contact .