Department Chair Scott Reeder, MD, PhD and Director of Informatics John Garrett, PhD explain how radiologists at the University of Wisconsin School of Medicine and Public Health are adopting and developing tools powered by artificial intelligence (AI) in a recent interview with HealthTech.
Within the department, AI image reconstruction algorithms enable improved image quality and faster captures for MRI, CT, and PET scans; cutting the scan time down by 30-50%. For patients, this means both shorter appointment times and reduced exposure to radiation. As appointment times are shortened, the wait time to schedule appointments decreases as an increased number of patients can be seen in a day.
AI isn’t only helping improve the scan duration and throughput; it’s being used to help analyze images and diagnose patients. Dr. Reeder shared how an AI tool can triage images from the emergency department.
“We see cases where an X-ray is acquired, and a minute or two later, it’s flagged and pops up on the workstation. I read that case next, and sure enough, if there is an abnormality, I call the emergency physician and say, ‘This patient has this urgent finding,’ and they can treat the patient immediately.”
Dr. Garrett explained that different AI models have different system requirements. While some can run on standard computers, others run on an on-premises server. Some scans that run dozens of triage tools even go through GPUs in the cloud, performing real-time data processing.
Read the full article on HealthTech to learn more about how AI is being used in clinical radiology settings.