Machine learning for ear infection diagnosis

Posted 11 Feb 2022

Dr Jacqueline Stephens has received a Flinders Foundation Health Seed Grant to investigate the use of machine learning in diagnosing ear disease in children.

Most children will experience otitis media (ear infection) at least once before the age of five. Typically, childhood ear infection resolves on its own but can persist, requiring antibiotics and even surgery.

Persistent, untreated ear disease can have negative impacts on a child’s development. Aboriginal and Torres Strait Islander children experience ear disease from an earlier age, more often and for longer durations than other children.

Early detection is crucial for ensuring timely and appropriate treatment, but current methods to diagnose ear disease and associated hearing loss can be complicated, expensive, and hard to access in rural or remote locations because of a lack of equipment or specially trained professionals.

Machine learning is the study of computer algorithms that can improve automatically through experience and the use of data.

Dr Stephens said this technology had the potential to simplify the detection process, and enable community-based workers and less confident technicians to more easily identify children with ear disease.

“The benefit of improved and easier diagnostic testing, particularly in community settings, has the potential to simplify diagnosis and, thus, ensure children receive timely treatment. This will minimise the impact of ear disease on health and wellbeing,” Dr Stephens said.

“This multifaceted project will bring us closer to our long-term goal to develop a portable diagnostic device that provides immediate feedback and allows a single operator to perform accurate and reliable ear health screening in community settings.

“The prompt treatment of children with otitis media will help to ensure the long-term impacts of this condition are avoided.”

Research category: Public Health

Project title: A multifaceted study of the application of machine learning for the diagnosis of childhood otitis media  

Lead researcher: Dr Jacqueline Stephens


Keep up to date

Subscribe to our newsletter to receive information on our latest news and events