AI and computer vision eliminate the need for cell biopsy in embryo testing


Credit: Pixabay / CC0 Public Domain

Despite ongoing controversy over its value in improving IVF births, testing embryos for their chromosomal content has become routine in many fertility clinics. Embryos with a normal complement of chromosomes (known as “euploids”) are known to have a good chance of implanting in the uterus to become pregnant, while abnormal (aneuploid) embryos do not. cap. Embryo aneuploidy testing (known as PGT-A) has so far required a sample of a single cell or several cells taken from the embryo by biopsy, and this has also raised safety concerns. , so that in recent times a search for non-invasive methods has emerged. years.

Now, a new study suggests that euploid embryos can be visually distinguished from aneuploid ones by artificial intelligence references of cellular activity, as seen by the time-lapse image and therefore without need for cell biopsy. The results of the study will be presented today at ESHRE’s annual online meeting by Ms. Lorena Bori of IVIRMA in Valencia, on behalf of a joint research team of IVIRMA Valencia and AIVF, Israel, co-directed by Dr. Marcos Meseguer from Valencia and Dr. Daniella Gilboa from Tel Aviv.

The visualization of embryo growth has been revolutionized in the last decade by time-lapse technology, which provides an image of every moment of an embryo’s development until, as a blastocyst, it is ready for transfer to the embryo. ‘uterus. However, so far the information obtained by time-lapse images has not been able to provide an accurate assessment of the chromosomal state of an embryo. Now, however, with AI can provide an objective and reliable prediction.

Behind the study it was found that chromosomally normal embryos begin their development as blastocysts at a time slightly earlier than aneuploid embryos, and this can be identified in computer vision by microscopic measurement of cell edges. · Cells. It is known to be an accurate method for quantifying the number of cells and the cell cycle of blastomeres (the forming the embryo). Applying this finding, the study retrospectively compared measurements based on computer vision of cell edges in time-lapse videos of 111 euploid and 120 aneuploid embryos.

The results showed that aneuploid embryos achieve their growth up to the blastocyst stage faster than euploid embryos, the authors said, due to their higher level of cellular activity.

“Our results demonstrate for the first time,” they added, “that an AI-based system can accurately measure the edges of microscopic cells in the dividing embryo, which allowed us to distinguish between euploid and aneuploid embryos.” .

“Our early results had shown that euploid and aneuploid embryos are visually different,” explained study director Marcos Meseguer, “significantly enough to merit computer vision research and to test whether a PGT-A test noninvasive could match the results of current invasive methods: without the cost or damage to the embryo that invasive methods could cause.We used the measurement of cell edges as a proxy for activity (which includes DNA replication and cell division) and we obtained 73% sensitivity and specificity in our results. “

Although Meseguer described the results and future research as “one of the milestones” of AI in reproductive medicine, he said additional studies are still needed to test and validate the algorithms in larger data sets. For example, the model so far classifies mosaic embryos (with a combination of euploid and aneuploid) as abnormal, although some studies have shown their viability during pregnancy.

However, Meseguer acknowledged that all the methods explored so far to test embryos without the need for biopsy have not proven to be as accurate as traditional biopsy methods. “Our current algorithm faces the same situation,” Meseguer said. “Our ability to predict is still limited, in which case our models could only be applied to those patients who do not require genetic testing according to a predefined medical indication. Therefore, our test so far could only be used to reduce the risk. to select a chromosomally abnormal embryo for transfer “.

However, the results of this time-lapse visualization approach show that it is fast and economical, especially when compared to the hitherto explored (which are based on the analysis of the culture media in which the embryo develops). “These [other non-invasive] the results, “Meseguer said,” take several days to occur due to genetic analysis, which forces patients to freeze all of their embryos and delay infertility treatment. ” artificial intelligence described in this study needs additional validation, is simple concept, can be built at home and yet can provide the most efficient means of proof for aneuploidy and its selection for transference.

New findings to increase IVF success rates

More information:
Presentation O-084, Monday, June 28, 2021: Computer vision can distinguish between euploid and aneuploid embryos. A new approach to artificial intelligence (AI) to measure chromosomal state-associated cell division activity.

Citation: AI and computer vision eliminate the need for cell biopsy in embryo testing (2021, June 28) retrieved June 28, 2021 at ai-vision-cell-biopsy-embryos.html

This document is subject to copyright. Apart from any fair treatment for the purposes of private study or research, no part may be reproduced without written permission. Content is provided for informational purposes only.

Source link