Flinders University Develops AI-Powered Evaluation Framework for Clinical AI Tools
A research team at Flinders University in Australia has recently unveiled a groundbreaking evaluation framework powered by artificial intelligence (AI) to assess the practical application of clinical AI tools. Known as PROLIFERATE_AI, this innovative tool builds upon the foundation of PROLIFERATE, a framework introduced by the Flinders University Caring Futures Institute in 2021, which aims to ensure that healthcare innovations are designed to address user needs and enhance health outcomes.
How PROLIFERATE_AI Works
The AI-powered evaluation framework focuses on evaluating the adoption, usability, and impact of AI tools within networks of people, technologies, and processes. By integrating user feedback and predictive modeling, PROLIFERATE_AI aims to optimize AI technologies to meet user needs, improve health outcomes, and enable sustainable practices.
The research team recently demonstrated the effectiveness of this AI-powered framework in assessing an AI tool used in 12 emergency departments across South Australia. This tool assists doctors in diagnosing cardiac conditions quickly and accurately. Findings from the study, published in the International Journal of Medical Informatics, revealed that less experienced clinicians faced usability challenges in using the technology compared to their more experienced counterparts. This underscores the importance of role-specific training, workflow integration, and interface enhancements to ensure the tool’s accessibility and effectiveness across diverse clinical roles.
Since its initial application, PROLIFERATE_AI has been instrumental in refining human-machine interactions, addressing ethical considerations related to healthcare AI, and ensuring privacy and ethical responsibility in deploying AI in high-stakes environments such as emergency departments.
Recently, a demonstration of the framework with CSIRO, Australia’s scientific research agency, showcased its ability to model and predict user interaction with up to 95% accuracy. This capability enables organizations to quickly adapt to user needs and enhance outcomes.
Currently, PROLIFERATE_AI is being applied to a major project focused on implementing ICU non-pharmacological agitation management guidelines.
Significance of PROLIFERATE_AI
Dr. Maria Alejandra Pinero de Plaza, the lead researcher, emphasized the importance of evaluating the usability, adoption, and impact of AI systems in healthcare. She highlighted the need for AI systems to be easy to use, adaptable, and genuinely helpful for healthcare providers and patients when it matters most.
The Larger Trend in AI Development
Over the past few years, various frameworks, guidelines, recommendations, and policies have been released to guide the development, adoption, and application of AI in critical industries, including healthcare. These initiatives include the World Health Organization’s Ethics and Governance of Artificial Intelligence for Health, the EU Artificial Intelligence Act, the OECD AI Principles, Australia’s 8 AI Ethics Principles, and Singapore’s Model AI Governance Framework.
The US Food and Drug Administration is currently finalizing its marketing submission recommendations for developers of AI-powered medical devices. Additionally, the National Institute of Standards and Technology in the US released an open-source platform last year for assessing data risks of AI and machine learning models, including those in healthcare. HITRUST also introduced an assessment framework last year for mitigating risks in AI deployment.
