Ever since its advent, in vitro fertilization has relied on trained specialists with a keen eye to select which embryos look more viable. This is a great responsibility since roughly only half of IVF procedures are successful. Any improvement of this process would be of great benefit to the expectant parents. Researchers at Weill Cornell Medicine have now developed an computer vision system, powered by artificial intelligence algorithms, that is able to identify with impressive accuracy whether a given human embryo will probably make it in the womb.

The system was trained using 12,000 photos of embryos, all photographed exactly 110 hours after fertilization. Each of the photos was marked with a grade by a trained embryologist, and these grades went into the system’s learning mechanism to identify the good embryos from the bad. Moreover, the actual outcome resulting from these embryos was known at the time of the research, so that went into the system’s assessment as well. The resulting algorithm has shown a 97% consistency with what humans would decide would be a viable embryo.

“By introducing new technology into the field of IVF we can automate and standardize a process that was very dependent on subjective human judgement. This pioneering work gives us a window into how this field might look in the future,” said Dr. Zev Rosenwaks, one of the researchers on the study and director and physician-in-chief of the Ronald O. Perelman and Claudia Cohen Center for Reproductive Medicine at NewYork-Presbyterian/Weill Cornell Medical Center.

More work can certainly improve the system if the researchers are able to train it with ever-larger data sets, which can certainly be gathered (and anonymized) from institutions around the world. We may one day see this process become nearly completely automated, with human embryos being selected by computers, a reality straight out of a science fiction book.

Image: Three examples of human embryos at the blastocyst stage photographed at multiple focal depths (four of seven focal planes shown here, from left to right). The embryos represent good (top), fair (middle) and poor quality (bottom) as designated by the embryologists’ grading system and additional statistical analysis.

Study in journal Digital Medicine: Deep learning enables robust assessment and selection of human blastocysts after in vitro fertilization…

Via: Weill Cornell Medicine…



Source link