Efficient patient trial recruitment is one of the major barriers to medical research, both delaying trials and forcing others to terminate entirely. NIH has estimated that 80% of clinical trials fail to meet their patient recruitment timeline and many fail to recruit the minimum number of patients to power the study as originally anticipated.

A solution to this problem is to utilize the vast amounts of patient data that is already available in the form of the electronic health records (EHR) collected at hospitals, such as those stored in Queensland Health’s (QH) iEMR.

Current systems that use this data, however, are only able to exploit structured data captured into the iEMR’s database, through strict SQL queries. UQ has lead the creation of novel AI methods to identify suitable patients that exploit the largest, but underutilised portion of the EHR: unstructured data like that contained in clinical letters and notes.

Our AI methods have been shown effective in research-based empirical studies, and are sufficiently mature to be translated into a software tool that can be deployed along with EHRs at Health Service Providers. This project will create a software platform that will interface with and extract data from the Queensland Health (QH) iEMR (including both structured and unstructured data), and that will implement our AI algorithm to determine the likelihood that a patient matches the clinical trial inclusion/exclusion criteria based on the data contained in the iEMR about the patient.

Additional funding will be sought in the future to seek clinical validation.

Project members

Professor Guido Zuccon

Professorial Research Fellow
School of Electrical Engineering and Computer Science

Professor Clair Sullivan

Conjoint Professor
Centre for Health Services Research
Casual Academic (General)
School of Business

Mr Anton Van Der Vegt

Advanced QLD Industry Research Fellow
Centre for Health Services Research

Professor Jason Pole

Deputy Director, QDHeC
Centre for Health Services Research