AI is now testing other AI health care tools

AI is testing other AI tools: ©Nanthiwan - stock.adobe.com

An artificial intelligence evaluation platform developed by Flinders University researchers in Australia has assessed the effectiveness of a cardiac AI tool trialed in South Australian hospitals, examining its ability to assist emergency department doctors and nurses in diagnosing heart conditions quickly and accurately.

The study, published in the International Journal of Medical Informatics, used the PROLIFERATE_AI evaluation tool to analyze the RAPIDx AI system, which was designed to help emergency physicians diagnose cardiac conditions by rapidly processing clinical and biochemical data. Given that chest pain is one of the most common reasons for emergency department visits, the AI tool is seen as a potential game-changer in expediting patient care.

“AI is becoming more common in health care, but it doesn’t always fit in smoothly with the vital work of our doctors and nurses,” said Maria Alejandra Pinero de Plaza, the lead researcher from Flinders University. “We need to confirm these systems are trustworthy and work consistently for everyone, ensuring they are able to support medical teams rather than slowing them down.”

The research involved assessing how well medical professionals engaged with the RAPIDx AI tool. The findings revealed that experienced clinicians, such as emergency department consultants and registrars, demonstrated a high level of comprehension and engagement with the tool. However, less experienced users, including residents and interns, encountered usability challenges. Additionally, registered nurses reported strong emotional engagement with the tool, acknowledging its potential to improve patient safety by reducing diagnostic uncertainty.

Pinero de Plaza emphasized that PROLIFERATE_AI goes beyond evaluating the technical accuracy of AI systems, focusing instead on usability and clinician trust. “Rather than focusing solely on technical performance, we evaluate AI tools based on real-world usability and clinician trust, ensuring that these technologies are not just innovative but also practical and accessible,” she said.

The study also highlighted the importance of targeted training and workflow-friendly interfaces to improve adoption rates among newer users. Clinicians expressed a desire for further automation of data integration into the tool’s user interface to enhance usability.

The researchers concluded that AI in health care should be developed with the needs of medical professionals in mind to ensure seamless integration into existing hospital workflows. “Our goal is to create AI solutions that empower doctors and nurses, not replace them,” Pinero de Plaza said. “Technology alone cannot solve the complexities of emergency care. We need AI systems that work seamlessly with clinicians, support decision-making under pressure, and integrate smoothly into existing workflows.”