Radiology technology shifts focus toward workflow and AI-driven insights

The field of radiology is undergoing a shift in innovation, moving from a primary focus on image sharpness to optimizing clinical workflows. As imaging volumes rise and healthcare systems face staffing shortages and clinician burnout, new technologies such as photon-counting CT, robotic positioning, and generative artificial intelligence are being developed to increase productivity and reduce administrative burdens.

Complementing these workflow improvements, new research from the University of Pennsylvania highlights the potential for artificial intelligence to extract broader health insights from routine CT scans. A study published in npj Digital Medicine introduced Percival, a vision-language model trained on over 400,000 CT image volumes and accompanying reports. The model can identify patterns in scans that may point to undiagnosed conditions or future health risks, showing significant associations with various laboratory measurements, including markers related to kidney function.