WellSpan York Hospital Implements AI Imaging Solution

Pennsylvania-based WellSpan York Hospital has implemented Aidoc's artificial intelligence (AI) solution, which helps radiologists find imaging abnormalities that cannote be seen with the naked eye.

WellSpan York Hospital performs 580,000 scans per year, and the AI solution will be able to report findings from scans within two to six minutes, according to the news release.  

The time taken to report any errors is critical because every minute matters when treating certain conditions, such as a patient with or at risk for a stroke.

The system will scan radiology results across an array of neuroscience and heart and vascular conditions, including brain hemorrhages and pulmonary embolisms. It will both flag imaging abnormalities and list findings in order of urgency for radiologists to review.

“Our radiologists at WellSpan Health are all board certified and fellowship train[ed], and with the addition of the innovative Aidoc solution reviewing cases, our team is already flagging results that are making a huge difference for our patients,” said Edward Steiner, MD, chairman of WellSpan Imaging and Radiation Oncology, in the press release. “Our goal is to offer our patients the very best care, and Aidoc is making that possible, speeding up processes and often leading to faster discharges home.”

The care delays caused by the COVID-19 pandemic also serve as another incentive for faster diagnoses.

“It’s critical that we work smart through the backlog of cases caused by the winter surge of COVID-19 hospitalizations. Aidoc makes sure that nothing goes unnoticed,” said Steiner. “Countless procedures and routine care visits have been postponed or skipped through this pandemic, but Aidoc gives us an innovative tool to optimize our mission.”

WellSpan hopes to expand the AI system to multiple locations, including Ephrata Community Hospital and Good Samaritan Hospital.

AI is becoming increasingly common in medical imaging.

Last year, Tulane University researchers found that AI could accurately detect and diagnose colorectal cancer as well or better than pathologists.

In addition, the National Institutes of Health (NIH) released instructions for AI practices in medical imaging. The goal of this project was to enhance the use of clinical AI and improve the diagnosis process.  

The NIH and the Radiological Society of North America, and the Academy for Radiology and Biomedical Imaging Research also released ideas to improving AI within medical imaging. The four main recommendations included establishing standardized use cases, solidifying data sharing practices, defining and recognizing clinical and infrastructure needs, and confirming the technology's accuracy.