Researchers analyzed time-lapse images from 10,148 human embryos, which were obtained from the Center for Reproductive Medicine at Weill Cornell Medicine, to train the DNN, called Stork. Embryos were separated into three groups, based off quality: good quality (1,345 embryos), fair quality (4,062 embryos), and poor quality (4,741 embryos). Time-lapse images were taken from each embryo at numerous time points and focal depths, and images with readability issues were removed from the data set.
Using 12,001 embryo images total, researchers trained Stork to detect good and poor-quality embryos from the data set. Results showed that Stork was able to identify good-quality and poor-quality embryo images with 96.9 percent accuracy.
Study authors concluded that while the deep learning method can provide accurate, quality assessments for embryo grading, it still poses limitations. Because researchers trained Stork to assess embryo images that were considered only as good-quality and poor-quality, the algorithm cannot identify the likelihood of live birth.
To access the full report, click here.
More articles on artificial intelligence:
How AI can impact 7 areas of healthcare
AI may predict patients’ risk of premature death, study finds
Google to spearhead AI, tech global advisory council
Becker’s 3rd Annual Fall Chief Pharmacy Officer Summit is the top tier event that the people leading the future of pharmacy convene. This 3-day event is built specifically for Chief Pharmacy Officers and senior pharmacy leaders. The sessions examine the workforce, financial and clinical priorities shaping pharmacy today. Becker’s is where candid discussions flourish and new ideas thrive. Becker's 3rd Annual Fall Chief Pharmacy Officer Summit
