Dr Rana, a Senior Research Fellow at USQ’s Centre for Health, Informatics and Economic Research
, is developing predictive algorithms for early detection of mood changes, which are typical of mental health issues and relapses.
The USQ researcher has created a tool to automatically determine mood, simply from day-to-day phone conversations on a smartphone, and a system for early diagnosis of relapse by tracking mood in real-time.
“I am looking at developing a system of predicting moods based on acoustic properties like pitch and intensity in conversations,” Dr Rana said.
“Our algorithms can accurately detect affective events like laughter, crying and anger, which is key to determine mood from conversation.
“We can detect these events with an accuracy of 98 per cent where the current state of the art accuracy is around 72 per cent.
Dr Rana said the research goal was to use the change-in-mood data as a predictor of relapses in mental illness.
“Early diagnosis of relapse allows clinicians to prevent hospitalisation enabling early intervention, reducing the burden on the hospital system and providing patients with the best chance of maintaining education or employment,” Dr Rana said.
As part of the YSEP joint initiative between the Australian and Chinese Governments, Dr Rana will work with leading Chinese research partners to share their knowledge in developing high-precision predictive algorithms.
Dr Rana said this exchange offered a great opportunity to visit Chinese institutes leading the way in his field of interest.
“It is a very exciting opportunity which will help advance this research for assessment of mental illness,” Dr Rana said.
In the two-week exchange, Dr Rana will work with researchers from Tsinghua University, Baidu Research, AISpeech and Institute of Scientific and Technical Information of China.