Israel Develops AI to Detect Embryo Genetic Disorders Using Blood Test
An Israeli startup is developing a non-invasive, AI-powered early detection method for identifying genetic disorders in human embryos.
Using a simple blood test taken from a pregnant woman during the first trimester, IdentifAI Genetics can read the embryo's entire DNA and perform in-depth analysis to identify genetic disorders.
Professor Noam Shomron is chief scientific officer and co-founder of IdentifAI, and heads the functional genomics group at the Faculty of Medicine of Tel Aviv University.
“A pregnant woman carries tiny fragments of DNA that originate from the embryo, from the placenta. If we look at a developing embryo, we see tiny pieces of placental or embryonic DNA, and if we can separate and isolate them well enough, we can read the embryo's entire DNA”
AI can be used to identify any genetic disease caused by minor changes in DNA. According to Shomron, using this new tool offers several advantages over existing methods.
First, the procedure is non-invasive and therefore carries no risk to the pregnancy, unlike amniocentesis, in which a small amount of amniotic fluid is taken from the amniotic sac of the developing fetus. Second, while amniocentesis is usually performed between weeks 15 and 20 of pregnancy, IdentifAI relies on a blood sample taken already in the first trimester.
Often, tests known as preimplantation genetic testing (PGT) can harm the embryo, and their results are often unreliable, leading to the possibility of reporting that an abnormal embryo is normal, or that a healthy embryo is actually abnormal. All of this leads to frequent dismissals in laboratories.
“You do this at a very early stage of embryo development, at week 10 of pregnancy. It is comprehensive: it reads the DNA from start to finish”
IdentifAI relies on a new AI-based software tool known as Hoobari, which was first developed at Tel Aviv University. Beyond detecting genetic disorders in embryos, this technology could find important applications in early cancer detection.
“When it comes to cancer, that will probably happen in a few years, but when it comes to embryonic development, we have already identified it in real samples from genetic institutes. This means we are now developing a product from it that should reach the market in one to two years”
Shomron and researchers from Tel Aviv University also recently used modern technologies to study the electronic medical records of about 8,000 patients with bloodstream infections at the Sourasky Medical Center (Ichilov) in Tel Aviv. In this study, scientists were able to determine with 82% accuracy which patients were at greater risk of serious complications from bloodstream infections.
The groundbreaking study was conducted by Yazid Zoabi and Dan Lahav, students in Shomron's lab, in collaboration with Dr. Ahuva Weiss Meilik, head of the I-Medata artificial intelligence center at Ichilov Hospital, Professor Amos Adler, and Dr. Orly Kehat.
During the study, AI was able to identify patterns in medical files to determine which patients were at greater risk of serious illness or death from blood infection. The system is currently being integrated into Ichilov Hospital and will be used to help doctors assess and rank patients by risk.
AI and big data are playing an increasingly important role in healthcare. As technologies can speed up the process by which medical workers process and analyze large volumes of data, they can help with everything from medical records and images to diagnostics and the development of better drugs.
Professor Meir Feder, head of the recently established Center for Artificial Intelligence and Data Science (TAD) at Tel Aviv University, said that AI has already surpassed trained radiologists when it comes to medical imaging.
“When you have a better understanding of images, you can detect a disease and make the right treatment decision. Overall, personalized medicine can only be achieved with AI. Artificial intelligence systems can do this far more accurately and reliably”
Although many technologies are still relatively new, Feder believes that in the near future they will make reality what once seemed medically impossible.
“Wherever you look, AI can help in healthcare and biomedicine. It is already helping, and in the coming years its impact will be tenfold”
