In-Vitro-Fertilization is the only option for many couples who are unable to conceive and are desirous of having kids. Though useful, it does have limited successful outcomes. Many couples especially females have found it physically gruelling. It also causes an emotional drain and patients sometimes have to undergo numerous rounds of this therapy before they get success in it. All this adds up to the cost and overall it is an expensive procedure. Now researchers are trying to employ Artificial Intelligence in the whole process and see whether it can improve outcomes.
AI and IVF
IVF is a painstaking task for the couples as well as the doctors. It takes many years and repeated rounds to improve on the success of the procedure. Now, research teams are trying to see whether Artificial Intelligence can boost up positive results. They are employing it to pick up the most viable embryo to implant and thus enhance the success rates and cut down on the pregnancy loss and hence cost.
 |
AI in IVF [Source: The Week]
|
Hadi Shafiee, a researcher at Brigham and Women’s Hospital in Boston tries to use AI in clinical practice. She said:
“There is room for AI to push the science and go beyond what is already known,”
But scientists are still skeptical of the whole application. They feel that ethical issues need to be addressed and have to deal with and correct the likelihood of a model's prediction going wrong. Besides, data is still inadequate and hence the whole thing cannot be put into practice yet.
More details of the process
A standard IVF cycle lasts 3 weeks. In the USA, the cost of this can vary from $ 12000 to $ 25000. The process entails the stimulation of the ovaries to produce more eggs by means of hormonal therapy, and after these eggs are visualized by means of USG to retrieve these eggs by laparoscopy. After the eggs are retrieved, the eggs are fertilized in the lab under suitable conditions with the sperms. The resulting embryos are then transferred into the female's womb or uterus.
 |
Objective embryo assessment by AI [Source: New York-Presbyterian Health Matters]
|
In order to know which is the best embryo to be implanted into the uterus, various features are taken into account in order to grade them. But this evaluation and grading might vary from person to person. Hadi states:
“It’s a very subjective process,”
“It really depends on the experience of the embryologist and even things like if the person is tired.”
How AI can help in the long run?
Experts are of the opinion that soon AI will be able to pick up these good-quality embryos and their pickup would be equal to that of the most experienced embryologist. Thus things might get more objective and standardized. Hadi trained an AI called Convolutional Neural Network on images of embryos. Each of these embryos had a score given to it by experts. 97 different embryo cohorts were used in the study and it included a total of 742 embryos. The accuracy of the best embryo pick up was 90%. 73% of the time, the AI system could also accurately assess an embryo's implantation potential.  |
AI in IVF [Source: Wired]
|
Hadi said:
“AI can see other things and bring in other factors that can improve the procedure,”
“That’s the beauty of this approach.”
The research is still in the early stage. Though the results are heartening, more research is needed in that direction. Matthew Vermilyea, the vice president of scientific advancement at Ovation Fertility said:
“I don’t think at the moment that we’re to the point where AI will completely override our morphological grading scheme,”
“The plan is to use this as another tool in our toolbox to help give us more confidence that we’re selecting the right embryo.”
Comments
Post a Comment