Infertility has a considerable impact on individuals' quality of life, affecting them socially and psychologically, with projections indicating a rise in the upcoming years. In vitro fertilization (IVF) emerges as one of the primary techniques within economically developed nations, employed to address the rising problem of low fertility. Expert embryologists conventionally grade embryos by reviewing blastocyst images to select the most optimal for transfer, yet this process is time-consuming and lacks efficiency. Blastocyst images provide a valuable resource for assessing embryo viability. In this study, we introduce an explainable artificial intelligence (XAI) framework for classifying embryos, employing a fusion of convolutional neural network (CNN) and long short-term memory (LSTM) architecture, referred to as CNN-LSTM. Utilizing deep learning, our model achieves high accuracy in embryo classification while maintaining interpretability through XAI.

Despite continuing controversies over its value in improving birth rates in IVF, testing embryos for their chromosomal content has become routine in many fertility clinics. Embryos with a normal complement of chromosomes (known as "euploid") are known to have a good chance of implanting in the uterus to become a pregnancy, while abnormal embryos (aneuploid) have no chance. Testing embryos for aneuploidy (known as PGT-A) has so far required a sample single cell or several cells taken from the embryo by biopsy, and this too has raised fears over safety such that a search for non-invasive methods has arisen in recent years. Now, a new study suggests that euploid embryos can be visually distinguished from aneuploid according to artificial intelligence references of cell activity as seen by time-lapse imaging--and thus without the need for cell biopsy. The results of the study will be presented today at the online annual meeting of ESHRE by Ms Lorena Bori from IVIRMA in Valencia, Spain, on behalf a joint research team from IVIRMA Valencia and AIVF, Israel, co-directed by Dr. Marcos Meseguer from Valencia and Dr. Daniella Gilboa from Tel-Aviv.

The automatic algorithms for perform at least as good as the average embryologist for blastocyst grading and indirectly for predicting fetal heart beat as described above. Developing the algorithm based on time-lapse sequences lead to an improved accuracy compared to using only still images. Training of deep learning algorithms is only based on raw image sequences and requires no prior knowledge of embryology. Thus, the algorithm learns by itself to extract the temporal and the morphological features that are most important for prediction of blastocyst grading. It is important to note that in order to design and train a deep neural network, a substantial amount of data (in this case image sequences) is required.

When she started in vitro fertilization, Katie Shepard, a medical device consultant from outside St. Paul, Minnesota, knew it could take more than one round to get pregnant. So, after the grueling regimen of hormone injections, ultrasound exams, egg retrieval and transfer of embryos back into her womb, she stayed optimistic -- until her second cycle. Of the 25 eggs harvested over the course of those two IVF treatments, only three developed into embryos. "It felt like someone took me out at the knees with a baseball bat," Shepard says. Worse, the embryos didn't take, nor did any from her third cycle.