To bring MRIs into pregnancy monitoring, MIT uses algorithms to chart the placenta

MIT's algorithms take a model of the curved, live placenta and flatten it, while maintaining its 3D volume, into a shape similar to how physicians see it after delivery. (Image courtesy of MIT CSAIL)

The health of the placenta is vitally important during a pregnancy, but assessing it can be a challenge.

While ultrasounds are one easy-to-use option, they often cannot provide enough fine detail to clearly show the small structures that help exchange nutrients between the mother and fetus. Being able to do so earlier in development could help physicians treat potential disorders.

MRI scans can provide the high resolution necessary—but the complex, 3D structure of the placenta, as it wraps around the inside of the womb, brings other challenges by making the scans difficult to interpret.

Virtual Roundtable

ASCO Explained: Expert predictions and takeaways from the world's biggest cancer meeting

Join FiercePharma for our ASCO pre- and post-show webinar series. We'll bring together a panel of experts to preview what to watch for at ASCO. Cancer experts will highlight closely watched data sets to be unveiled at the virtual meeting--and discuss how they could change prescribing patterns. Following the meeting, we’ll do a post-show wrap up to break down the biggest data that came out over the weekend, as well as the implications they could have for prescribers, patients and drugmakers.

To make that task easier, researchers at the Massachusetts Institute of Technology’s (MIT's) Computer Science and Artificial Intelligence Laboratory are working on algorithms that can flatten out an MRI’s 3D image into a format that doctors can easily recognize.

An animation of the 3D image flattening process
(Courtesy of MIT CSAIL)

“The idea is to unfold the image of the placenta while it’s in the body, so that it looks similar to how doctors are used to seeing it after delivery,” said Ph.D. student and lead author Mazdak Abulnaga, working with MIT professors Justin Solomon and Polina Golland.

Additionally, the algorithm could be used to check for biomarkers of poor health, according to Golland. This could accelerate turnover by locating potential problem areas without having radiologists examine multiple slices of the image.

“While this is just a first step, we think an approach like this has the potential to become a standard imaging method for radiologists,” Abulnaga told MIT News.

MIT CSAIL placenta mesh
The resulting image (MIT CSAIL)

The algorithm takes a model of the placenta’s shape and breaks it down into a mesh of thousands of tiny pyramids. This allows the computer to manipulate the curved image into something flatter, while still maintaining its 3D volume.

The researchers’ paper (PDF) will be presented this month in Shenzhen, China, at the International Conference on Medical Image Computing and Computer-Assisted Intervention. Going forward, the team hopes to compare in utero scans with images of the same placentas following birth.

Suggested Articles

The FDA named more than two dozen coronavirus antibody tests that should be taken off the market weeks after the agency clamped down on tests.

Inovio CEO J. Joseph Kim is undeterred by short sellers and other detractors who doubt his company can shuttle a COVID-19 DNA vaccine to market.

The machine-learning programs scroll through data to detect hard-to-spot patterns. Yet few have been tested against standard procedures.