AI to help reduce hospital readmissions – and improve patient outcomes
Through AI technology, we're investigating the causes of hospital readmissions to deliver better patient outcomes – and reduce the burden on our healthcare system.
Partnering with the Faculty of Medicine, Nursing and Health Sciences, the Faculty of IT is using AI to better understand the causes of hospital readmissions. The expected outcomes? Better patient outcomes and less financial burden on the healthcare system.
There’s a growing need among medical professionals and patients to improve the quality of care and lower hospital costs.
As part of this project, which is supported by the Monash Institute of Medical Engineering, our data science specialist Professor Wray Buntine led the biggest study of its kind in Australia.
In the research, AI technology was used to examine 10 years’ worth of patient medical information. This included 14,000 records with details on more than 327,000 hospital readmissions. Now, Professor Buntine and his team have developed a prediction model that achieves state-of-the-art prediction accuracy for patients of chronic liver disease and heart failure.
‘This study used a rich source of clinical patient data to infer medical risk predictions and improve the quality of healthcare. By examining these complex data sets, the machine learning algorithms we’ve developed can make predictions on medical risks, such as identifying if and when a patient will readmit – and whether this can be avoided.’ Professor Buntine said.
Dr Suong Thi Thanh Le, Gastroenterologist at Monash Health and Senior Lecturer from the Faculty of Medicine, Nursing and Health Sciences asserts that high readmission rates are issues across the country.
‘Chronic disease accounts for 70% of Australian disease burden and high healthcare utilisation. To accurately identify patients at risk of emergent readmission for chronic liver disease or heart failure means we could deliver more timely interventions which prevent hospitalisation.’ she said.
Findings of the project will be published at the 24th European Conference on Artificial Intelligence in August. Insights from this research will be further validated in real hospital settings to help predict patient readmissions.