RIYADH: King Abdullah University of Science and Technology has developed a new tool to study models of human embryos grown under artificial laboratory conditions.
Scientists at the university have explained how the tool — deepBlastoid — is capable of analyzing pictures of the models with results of equal quality to human experts — but 1,000 times faster.
“Little is known about the very early stages of embryo development. With deepBlastoid we can scale up blastoid research to study embryo development and the effects of chemicals on the embryo and pregnancy,” said Mo Li, an associate professor at KAUST and an expert in stem cell biology.
The laboratory at which Li works pioneered the embryo models known as human blastoids.
The understanding of the human embryo in its early stages is significantly important for scientists in order to study fertility and detect pregnancy complications as well as developmental disorders.
Due to ethical considerations, this type of research on human embryos has been limited.
In this study, researchers at KAUST developed and trained their new tool using more than 2,000 microscopic images of blastoids.
They also used the tool to analyze the impact of various chemicals on blastoid development, examining an additional 10,000 images.
The insights gained from the KAUST research are of significance for women who are taking prescription medication or other drugs while planning for pregnancy, as they shed light on how such substances might interfere with early embryonic development.
Peter Wonka, professor and member of the Center of Excellence for Generative AI at KAUST, said that “deepBlastoid not only matches human performance in accuracy, it delivers an unparalleled increase in throughput. This efficiency allows scientists to analyze vast amounts of data in a short time, enabling experiments that were previously unfeasible.”
