Associate Professor Leon Worth, Medical Director Infection Prevention, the Peter MacCallum Cancer Centre
People who are being treated for cancer face a number of challenges in addition to the cancer. Their treatment necessarily results in their immune system being compromised which means that patients are at a greater risk of infection. It’s harder to treat those infections when you don’t have a functioning immune system to assist.
Infection-related mortality is the leading cause of early death for many adult Australians with blood cancers undergoing leukemia treatment or allogeneic haematopoietic stem cell transplantation. Until now, systematic surveillance of these infections in clinical settings took too long and required too many resources to be feasible.
Researchers at the Peter MacCallum Cancer Centre are tackling this issue using a next-generation digital platform called EINSTEIN (Enhancing Infection Surveillance to Transform Excellence In National cancer care), a tool for providing clinicians with a platform for understanding and flagging infections. EINSTEIN draws on a digital learning health framework and data science to detect serious and emerging infections in the immune compromised cancer population.
The poor outcomes for patients who contract an infection while being treated highlight the critical need for an integrated clinical data platform to systematically and quickly identify both the infection and the at-risk patient.
Harnessing artificial intelligence to facilitate multi-dimensional and enhanced monitoring systems is the only viable and sustainable approach for identification and reporting of invasive fungal infections and similar complex infection syndromes in high-risk patient populations.
“The earlier we can detect and treat an infection the greater our chance of stopping it from becoming life threatening. With this platform we aim to develop a national scalable system standardising digital algorithms,” says Associate Professor Leon Worth, Medical Director Infection Prevention at the Peter MacCallum Cancer Centre and Principal Investigator for the project.
“The data will be incorporated into a clinical portal to accommodate everyday workflows in an electronic health record environment to assist clinical decision making around infection surveillance in the real world,” he said.
This project is led by the Peter MacCallum Cancer Centre in collaboration with the University of Queensland, Royal Melbourne Institute of Technology, BioGrid Australia and The Royal Melbourne Hospital.
BioGrid has already been working with Associate Professor Worth on the precursor to the EINSTEIN project - the Invasive Fungal Infections Surveillance (IFIS) Project – providing infrastructure and data curation services.
IFIS was a pilot project targeted specifically towards invasive fungal infection as a pilot, demonstrating that the approach works and is scalable. EINSTEIN will now extend the model to other types of infection, applying natural language processing to many different data sources including emergency presentations, pathology, and radiology to identify the conditions under which an infection might occur and which patients are experiencing those conditions.
“It’s another way of collating information: identify the key information points and then feed those back to clinicians to determine whether the patient is at risk of an infection, or in fact already has a specific type of infection,” says Associate Professor Ashley Ng, Senior Clinical Haematologist and Clinical Informatics Lead in the Department of Health Services Research at the Peter MacCallum Cancer Centre.
Once rolled out, the project will assess the success of getting the right information to the right clinicians so that the right decisions are made for the patients, more quickly.
BioGrid will again provide the data sharing and linkage infrastructure and services underpinning the initiative, as well as establishing the data agreements and streamlining data governance between participating institutions with the BioGrid collaboration agreement.
BioGrid will also securely code patient data, ensuring that no identifying information is supplied to researchers using the data.
Maureen Turner, Chief Executive Officer of BioGrid.