While Artificial Intelligence (AI) technologies promise to augment clinical decision-making, translatable benefits to care quality and patient safety remain largely unmeasured.
The use of AI to assist clinicians in their safe and effective interpretation of chest X-rays in emergency departments is the focus of a new Digital Health CRC project, led by Macquarie University, Sydney Local Health District (LHD) and Australian MedTech partner, annalise.ai.
This new research project will evaluate the use of the Annalise Enterprise CXR solution, which is an AI-enabled software as a medical device (SaMD) that provides clinicians with a notification of suspected clinical findings on chest X-rays.
Unlike traditional computer aided detection models that detect a small number of findings, the annalise.ai solution has the ability to detect 124 findings and highlights the localisation of suspected findings on chest X-ray images. This project will, for the first time, provide an academic evaluation of this AI-assisted chest X-ray interpretation solution on clinical decision-making and patient management at the point of care.
Associate Professor Michael Dinh, Founder and Director at Royal Prince Alfred Hospital’s Green Light Institute for Emergency Care said the AI technology had potential to assist doctors at the point of care.
“This is an exciting opportunity that will help our front line staff in emergency departments make more informed decisions,” Professor Dinh said. “The ability for clinical staff to have immediate access to more detailed findings will ultimately lead to better outcomes for our patients.”
Professor Farah Magrabi from the Australian Institute of Health Innovation at Macquarie University explained the study will facilitate the evaluation of the annalise.ai CXR solution in a controlled online environment with simulated cases about patients presenting to the emergency department.
“Currently, chest X-rays are ordered for about half the patients who present to emergency departments, but the window for a radiologist to review them and write a report is very small, and there can be delays in times of high demand,” Professor Magrabi said.
“The findings detected by AI can provide the doctor with more timely clues on what to look for until they have advice from a formal report.”
The research team will study the impact of the annalise.ai solution on diagnostic accuracy, patient management, and cognitive load in comparison to current practice. Ultimately, the study aims to inform future research and implementation of AI into clinical practice.
Dr Mark Phillips, Head of Clinical Research & Medical Affairs for annalise.ai, said that annalise.ai is very excited by this research collaboration through the DHCRC.
“Chest X-ray remains a core medical imaging diagnostic tool, and in Emergency Medicine, the need for accurate, time-critical clinical decisions is paramount,” he said.
“This research project seeks to demonstrate the clinical workflow and decision-making benefits of the addition of the comprehensive Annalise Enterprise CXR solution in the clinical review of chest X-rays in an Emergency Department setting, in particular, the benefits in diagnostic accuracy, and improvements in patient management decisions, for emergency doctors.”
Annette Schmiede, Digital Health CRC CEO said the clinical evaluation of new technologies was an important element of the DHCRC’s R&D portfolio and welcomed the opportunity to work with annalise.ai as a new technology partner.
“Supported by the Commonwealth Government through their Cooperative Research Program, the Digital Health CRC invests in collaborative research programs to solve industry identified problems and to grow and improve the competitiveness, productivity and sustainability of Australia’s digital health sector,” Ms Schmiede said.
“annalise.ai is a rapidly growing medical device company and an Australian success story. We’re very pleased to enable the implementation and evaluation of their technology, which aims to improve clinical decision making and patient management in Australian hospitals.”
For more information on this project, you can contact one of the lead partner organisations at: