Indeed AI can improve embryo selection. Moreover, computer vision tools achieve superhuman performance in many tasks starting from recognizing addresses on letters, through telling different species of iris flowers, ending with complex, various quality medical images. This, however, raises new challenges as decisions taken by deep neural networks are often hard to understand and explain. We are currently working on new explanation tools, e.g., ones that will highlight features of embryos important to make decisions. We hope to find new indicators for embryo selection this way.
The new tool EMBRYOAID has been developed to provide embryo viability assessment, which supports implantation prediction as well as provides the ability to rank embryos in order of priority. Traditionally, embryo evaluation and selection has been a manual process, limiting patient access to treatment while also opening the door for human error. One of the key advantages that EMBRYOAID, the AI-driven embryo quality assessment tool, was designed to offer is accuracy and consistency in assessment. EMBRYOAID’s proprietary AI-based algorithms become more accurate the more data it gathers, leading to uniform and accurate embryo assessment.
