AI algorithm estimates gestational age of fetus on par with skilled sonographer


Share post:

New analysis out of NEJM discovered that a man-made intelligence algorithm was in a position to estimate the gestational age of a fetus on par with a skilled sonographer performing a fetal biometry. 

The research, which was performed in North Carolina and Zambia, collected ultrasound sweeps from 4,695 pregnant volunteers. Researchers collected blind sweep movies on each a industrial ultrasound machine and on a transportable lower-cost machine, the Butterfly iQ. 

“We skilled a neural community to estimate gestational age from the sweeps and, in three check information units, assessed the efficiency of the unreal intelligence (AI) mannequin and biometry in opposition to beforehand established gestational age.”

Researchers report that the “deep studying mannequin outperformed biometry, with an total [mean absolute error] of three.9±0.12 days”, in the meantime the biometry’s MAE was 4.7±0.15 days. 

“Our information present that AI can estimate gestational age from a collection of blindly obtained ultrasound sweeps with accuracy just like that of a skilled sonographer conducting customary biometry,” researchers of the research wrote.

“This efficiency seems to increase to sweeps collected by untrained suppliers in Zambia utilizing low-cost ultrasound units. Whether or not this know-how might be efficiently disseminated into extant healthcare techniques in low-resource settings would require additional research.”


Maternal and toddler mortality remains to be a serious difficulty worldwide. In line with the World Well being Group, in 2017 about 810 ladies died each day from preventable causes associated to being pregnant and childbirth. 

Low revenue international locations are most impacted by maternal deaths. WHO experiences that in 2017 the maternal mortality fee in low-income international locations was 462 per 100,000 stay births, in comparison with 11 per 100,000 stay births in high-income international locations. 

The research’s researchers pitch this know-how as a instrument to probably assist maternity care in each high- and low-income international locations. 


Butterfly Community landed FDA clearance for its ultrasound scanner that makes use of easy-to-manufacture semiconductor chips. Just lately the corporate rolled out a brand new ultrasound platform that is ready to combine right into a well being system’s medical workflows. 

Different firms working within the maternal care house embody Royal Philips, which launched smartphone-connected ultrasound Lumify in 2015, and sensible ultrasound Exo, which nabbed $220 million in Collection C funding


Supply hyperlink


Please enter your comment!
Please enter your name here


Related articles