Revolutionizing Medical Imaging through Artificial Intelligence
What do you find at the intersection of artificial intelligence (AI), the cloud and medical imaging? The pioneering San Francisco-based start-up, Arterys. As WIRED put it, it developed an AI algorithm using MR images to draw up the contours of the heart’s four chambers, measuring precisely how much blood they move with each contraction.
Cardiologists usually need 30 to 60 minutes to calculate the volume of blood transported with each pump, but Arterys’ AI comes up with the answer within seconds. I asked Carla Leibowitz, Arterys’ Head of Strategy and Marketing about their innovative solution.
If I were a cardiologist, what would it look like to use your product in real life?
Carla: “If you were to sit down to use our CardioDL product, you would send a traditional cardiac steady-state free precession (SSFP) MRI scan to our cloud, and once you log in and open the file, it would take 15 seconds for all the ventricular contours to be created. In case of our 4D Flow product, the scan is uploaded directly from the MRI machine to our cloud, so you would log in, and the volumetric scan would be ready for you to analyze it.
We also automate time-consuming tasks such as measuring for cardiac function, flow or distances, so the output is structured and clinicians spend less time dictating.”
Where could medical professionals find AI-supported medical imaging already?
Carla: “Our software has been used as a research tool at around 40 hospitals, including Mayo Clinic or Boston Children’s Hospital. Recently, we launched our first clinical program with GE Healthcare at Fairfax Radiology Consultants, a US private radiology practice, where clinicians have scanned to date 74 pediatric and adult patients with our 4D Flow branded asViosWorks. “
What are the first experiences?
Carla: “Physicians tend to embrace our FDA-cleared and CE-marked innovation, as we automate the very tedious things that they do not like to do, but need to get right. Melany B. Atkins, MD, Director of Cardiac Imaging at Fairfax said, “the results have been impressive. I have also found it to be particularly useful in assessing difficult conditions such as congenital heart disease, valvular heart disease and left ventricular hypertrophy”.”
What factors make people reluctant to use cloud-based medical imaging?
Carla: “The main concerns are around taking patient data to the cloud. However, we created a software that strips patient data from the images we send, and only users with the right credentials can reunite them. Our cloud never sees or stores any confidential patient information. “
Do you still have questions about Arterys or medical imaging? Reach out to me on Facebook or Twitter.