Alexander Boehmcker, former CEO of a large teleradiology practice and VP Europe at Aidoc, shares how AI can be a safety net for flagging incidental pulmonary embolisms, and why this is even more important during the summer months.
Summer is the time to disconnect and reconnect with your family and friends, and for radiologists, a time to recover from periods of heavy workloads during the first half of the year. Whether home or away, people continue to seek healthcare during the summer months, leading to more non-urgent outpatient cases which can pile up in hospitals for days or sometimes weeks. Personally, I have seen significant backlog of cases building up during summer months at hospitals, radiology practices and teleradiology companies with the potential risk for patient safety. All it takes is one incidental pulmonary embolism (iPE) case going undetected during this period of the year with low radiologist capacity.
Untreated PEs have a mortality rate as high as 30%, and in times of SARS-Covid-19, this risk is not diminishing. We have recently hosted panel discussions with radiology experts, all with vastly different AI use cases. What amazes me the most is the common thread: AI described as a safety net that radiologists can trust, to support them by flagging which scans to look at first. It’s great to be able to provide that support, and as summer approaches, I can’t help but think about the impact this safety net can have on patient care, especially when flagging a potential acute PE.
“Some [cancer] patients had pulmonary embolisms, but they were not detected in a timely manner. So, we decided to implement this tool to prioritize the follow-up scans where the emboli are present,” Dr. Erik Ranschaert, Tilburg, The Netherlands, shared on the live panel webinar, Is AI ready to become the standard of care? This is just one example.
Since AI doesn’t need a vacation, and is running continuously in the background, it can flag potential iPE cases just a few minutes after the scan has taken place. These cases can be flagged, prioritized, and brought to the attention of the on-duty radiologist quickly, enabling them to report this case with high urgency. The AI solution therefore has shown the ability to decrease the turn-around-time of these critical cases and has a positive impact on the reduction of length of stay of the patients in the hospital.1
How might AI contribute to decreased turnaround time and provide a safety net for your institution, especially during the summer months?
Feel free to contact me at [email protected], and let’s discuss.
1A clinical study in Cedars Sinai Medical Center has shown a decrease of length of stay of patients diagnosed with pulmonary embolism of -31.9% (from 8.77 days down to 5.97) after the implementation of the Aidoc PE solution (p-value < 0.05). “Artificial Intelligence Software for Flagging Pulmonary Embolism on CTPA Associated with Reduced Length of Stay.” Cedars Sinai Medical Center, abstract submitted to RSNA 2020.