A study from the Icahn School of Medicine at Mount Sinai, published in npj Health Systems, explores the effectiveness of AI in healthcare under high-demand conditions. Researchers highlight that AI systems function better when utilizing a coordinated network of specialized agents rather than a single, general-purpose system. This multi-agent approach not only maintains accuracy across complex clinical tasks—such as information retrieval and medication dosing—but also significantly reduces computing costs. Senior author Dr. Girish N. Nadkarni emphasizes that distributing specific tasks to specialized AI agents enhances efficiency and allows healthcare providers to focus more on patient care. The study revealed that while a single agent’s accuracy declined to only 16% under heavy demand, a specialized system maintained high performance. Future research aims to test these AI systems in actual clinical settings to further optimize healthcare operations. This innovative architecture could revolutionize how healthcare organizations scale AI, ensuring better quality and safety in patient care.
Source link
