Smart AI assistant reduces workload in intensive care units
Healthcare professionals in intensive care units spend significant time on administrative tasks, such as triage processing, preparation of ward rounds, and composing discharge letters.
This detracts from direct patient care. The Delphyr Co-Pilot project introduces a smart AI assistant to the ICU at Erasmus MC that automatically supports these tasks, freeing up more time for bedside care.
The administrative burden in healthcare is structurally increasing, contributing to stress, burnout, and staffing shortages. Erasmus MC handles approximately 3,000 ICU admissions annually, accounting for 20,000 patient days. With an administrative load of four hours per patient per day, a 30% time saving equates to around 24,000 hours annually, representing a structural saving of 14 FTEs. Delphyr Co-Pilot thus enhances staff availability and helps make ICU capacity more resilient.
The AI assistant integrates seamlessly with the electronic health record (EHR), complies with European regulations for medical devices, and has previously been successfully tested in a simulated setting. This project will validate it in real clinical practice at Erasmus MC. The project comprises four components: integration with the EHR, clinical usability testing, impact evaluation, and dissemination of results.
Deliverables include a validated AI solution, user data, evaluation reports, and an implementation model for scale-up in other hospitals.
