Urszula Sankowska
MIM Fertility, Poland
Biography
AI-powered applications have the potential to vastly improve medical care. The path to this revolution is still rather long and still has some obstacles. However, one of the areas of medicine that sees more and more progress with respect to application of AI is IVF. This follows as the IVF process offers unique, controllable and reputable conditions. These opportunities are further strengthened by the fact that over 10% of the human population will encounter reproduction problems, thus In Vitro Fertilization is becoming one of the fastest growing areas of healthcare. In this talk, I will present deep learning tools that support doctors in IVF treatment.
First, I will talk about how to automate ovarian ultrasound monitoring with deep learning computer vision algorithms. The calculating and measuring of follicles is of major importance for IVF, e.g., it helps estimate the ovarian reserve as well as the timing of the whole process.
Second, I will talk about an analytical and predictive system aiming for fully automated identification of the highest quality embryos based on a large collection of time-lapse images of human embryos. Both AI-driven tools are designed for fertility clinics, diagnostic centers and hospitals and will respond to their well-identified needs. In this talk, I will describe the process of building these solutions and show the most important issues of applying deep learning in a medical context. Finally, I will talk about deployment of these tools in fertility clinics in Poland and worldwide.
Abstract
Abstract : Optimizing IVF treatment with AI