Problems of fertility are becoming common these days. In a developing country, one in four couples is said to be impacted by infertility. About 48.5 million couples experience infertility across the world. Today, infertility is an issue that has become an epidemic. 

In Vitro Fertilization or IVF is a technique that aids people facing fertility issues to have a baby. Even though IVF treatment is one of the most common treatments in such scenarios, its outcomes are unpredictable. Moreover, abysmal access to IVF care makes matters work. Even in developed countries like the US, just 2% of people with infertility issues have tapped into IVF. 

Today, we are seeing exciting applications of data science in fertility that could improve embryologists' capacity cycle by 50% and increase the chances of live birth by 4%. IVF is the procedure by which an egg is removed from a person's ovaries and fertilized with sperm in a laboratory. The successfully fertilized egg-an embryo- is then implanted into the uterus to grow. There are two critical challenges to IVF: 

1. Poor access to care: When only 2% of the total population are provided access to care when the issue is much broader, it is evident that the problem of accessibility is a significant issue. 
2. Uncertain outcomes: Even if the process is successful, the chances of conceiving from a single IVF cycle is only 30%. Expenditure on an IVF treatment is also high, beyond the affordability of several patients. The uncertainty of outcome is yet another serious issue of IVF treatment. 

AI and fertility treatments 

The novel technique, developed in a collaborative endeavor between researchers at IVIRMA Valencia and AIVF, Israel, eliminates the requirement of invasive cell biopsy IVF treatment by utilizing computer vision and AI references of cell activity observed through time-lapse imaging. In addition, the innovative method can efficiently distinguish between euploid and aneuploid embryos, determining which ones will be optimal for IVF treatment.  

Over the last ten years, embryo growth has been visualized by time-lapse technology, which produces an image for each stage of the embryo's development until it becomes a blastocyst and is ready to be transferred to the uterus. However, this time-lapse imaging cannot analyze an embryo's chromosomal status, with Computer Vision with AI possibly providing the perfect solution. 

Chromosomally normal embryos develop as blastocysts at the initial point rather than aneuploid embryos, which can accurately be recognized through Computer Vision. The team combined their findings with computer vision-based measurements of cell edges in time-lapse videos of 111 euploid and 120 aneuploid embryos, discovering that aneuploid embryos reached their blastocyst stage earlier. 

According to the study director, their results show that an AI-based system can precisely measure microscopic cell edges in the dividing embryo. This allowed them to distinguish between euploid and aneuploid embryos. However, researchers reiterated that they would need further investigations to authenticate algorithms in larger datasets. Nevertheless, this computer vision approach is efficient and economical compared to other non-evasive procedures. It can also be built at home and may even become a most efficient way of examining the aneuploidy of embryos.  

While understanding the potential of AI, it is significant to realize the challenges it must overcome in the field. Mentioned following are some of the key challenges. 

• Tackling data availability issue 
• Integration into the workflow 
• Fostering trust and adoption 

Growing with AI 

Research scientists expect IVF to grow in prevalence with better success and fewer cost barriers to care. In addition, the development of early detection tools can warn patients who might experience issues in fertility, as opposed to today, where patients end up with unretrievable fertility potential. With enhanced visibility about their fertility, patients will be able to take early actions as well. 

Sources:

Forbes