by
Lauren Dubinsky, Senior Reporter | July 23, 2018
From the July 2018 issue of HealthCare Business News magazine
Artificial intelligence is taking the healthcare industry by storm. Algorithms can now automatically detect suspicious lesions and identify and prioritize the most urgent cases.
For obstetrics and gynecology (OB/GYN), these tools hold particular promise for automating standard measurements taken with ultrasound, thereby shortening scan times and making them more consistent and repeatable.
“There has been a lot of focus on how to speed up those exams and make sure we get the right information every time, given the pressure on patient throughput and the increasing volumes,” Jeff Cohen, leader of the OB/GYN business unit at Philips, told HealthCare Business News.
Last year, the company launched a tool called fetal biometry that is powered by AI. It automatically recognizes that the user is looking at a fetal head or bone and takes the measurements.
According to Cohen, this saves a significant amount of time because most pregnancies are normal. The clinician just has to confirm fetal development and this new tool allows them to do that more quickly.
Canon Medical Systems USA Inc. offers a similar tool on its Aplio i800 ultrasound system, and other major ultrasound vendors, including GE Healthcare, are interested in the technology, but don’t yet have an offering.
“When it comes to interpretation, if you look at some of these clinical challenges they can be quite rare — happening in one in 1,000 patients,” said Roland Rott, general manager of women’s health ultrasound at GE. “It can be demanding for the broad field of ultrasound technicians and doctors to always identify that with 100 percent confidence.”
Not everyone believes AI is ready to fulfill all of its potential. Dr. Kathleen Bradley, a maternal-fetal medicine specialist who runs her own practice in Los Angeles called Maternal Fetal Medicine Consultations, believes that AI technology is promising, but that a lot of work needs to be done to understand and implement it in a practice.
“I don’t want to lose the personal touch [with the patient], which I think is so very important especially in my field of medicine,” she said. “I’m not a radiologist so I don’t read images behind a computer. I do everything in real time. I’m interested in seeing how AI will couple with the personalization of medicine.”
3D/4D and 5D
“Ten years ago, it was quite normal to have simple 2D images, which were slices of the body,” said Rott. “Today, 3D and 4D ultrasound is the norm.”
He added that with this technology, clinicians can acquire the full volume of information instead of a single slice. That can help them come to certain conclusions faster.
