Deep learning for embryo evaluation using time-lapse: a systematic review of diagnostic test accuracy

Berman, Aya; Anteby, Roi; Efros, Orly; Klang, Eyal; Soffer, Shelly

doi:10.1016/j.ajog.2023.04.027

Cited by 2 publications

(1 citation statement)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some studies used deep learning algorithms on a single final image of embryos (Khosravi et al, 2019;Chavez-Badiola et al, 2020;Huang et al, 2022). Other studies applied deep learning algorithms to time-lapse images of embryos (Lee et al, 2021;Liao et al, 2021;Sawada et al, 2021;Berman et al, 2023;Sharma et al, 2024). While deep learning approaches are more accurate on TLIs than just single images, TLI facilities are unavailable in most IVF laboratories.…”

Section: Introductionmentioning

confidence: 99%

An artificial intelligence algorithm to select most viable embryos considering current process in IVF labs

Borna,

Sepehri,

Maleki

2024

Front. Artif. Intell.

View full text Add to dashboard Cite

BackgroundThe most common Assisted Reproductive Technology is In-Vitro Fertilization (IVF). During IVF, embryologists commonly perform a morphological assessment to evaluate embryo quality and choose the best embryo for transferring to the uterus. However, embryo selection through morphological assessment is subjective, so various embryologists obtain different conclusions. Furthermore, humans can consider only a limited number of visual parameters resulting in a poor IVF success rate. Artificial intelligence (AI) for embryo selection is objective and can include many parameters, leading to better IVF outcomes.ObjectivesThis study sought to use AI to (1) predict pregnancy results based on embryo images, (2) assess using more than one image of the embryo in the prediction of pregnancy but based on the current process in IVF labs, and (3) compare results of AI-Based methods and embryologist experts in predicting pregnancy.MethodsA data set including 252 Time-lapse Videos of embryos related to IVF performed between 2017 and 2020 was collected. Frames related to 19 ± 1, 43 ± 1, and 67 ± 1 h post-insemination were extracted. Well-Known CNN architectures with transfer learning have been applied to these images. The results have been compared with an algorithm that only uses the final image of embryos. Furthermore, the results have been compared with five experienced embryologists.ResultsTo predict the pregnancy outcome, we applied five well-known CNN architectures (AlexNet, ResNet18, ResNet34, Inception V3, and DenseNet121). DeepEmbryo, using three images, predicts pregnancy better than the algorithm that only uses one final image. It also can predict pregnancy better than all embryologists. Different well-known architectures can successfully predict pregnancy chances with up to 75.0% accuracy using Transfer Learning.ConclusionWe have developed DeepEmbryo, an AI-based tool that uses three static images to predict pregnancy. Additionally, DeepEmbryo uses images that can be obtained in the current IVF process in almost all IVF labs. AI-based tools have great potential for predicting pregnancy and can be used as a proper tool in the future.

show abstract

Section: Introductionmentioning

confidence: 99%

An artificial intelligence algorithm to select most viable embryos considering current process in IVF labs

Borna,

Sepehri,

Maleki

2024

Front. Artif. Intell.

View full text Add to dashboard Cite

show abstract

Embryo selection, AI and reproductive choice

Halsband

2024

AI Ethics

View full text Add to dashboard Cite

In reproductive medicine, current research into the use of artificial intelligence (AI) to improve embryo selection has been met with enthusiasm. Within ethics, previous assessments of AI-assisted embryo selection have focused, for example, on liability gaps or risks arising from opaque decision-making. I argue that this focus on the ethical issues raised by AI in embryo selection alone is incomplete because it neglects how AI’s convergence with other innovative reproductive technologies raises further ethical issues. I describe how AI is acting as a catalyst for a social disruption of human reproduction and for a profound change in reproductive morality. The social disruption is the result of the convergence of improved embryo culture, the optimization of embryo selection through AI and the possibility of selecting a screened embryo. This technological interplay creates a pull towards assisted reproduction, even for those prospective parents who can reproduce without medical assistance. In discussing a fictional case of prospective parents, I argue that this social disruption of reproduction is linked to a moral disruption. The moral disruption manifests itself in the deep uncertainty of prospective parents about legitimate ways of procreating. I explain how this uncertainty is rooted in a technology-induced disruption of the concept of reproductive choice. I then outline how the ethical debate about AI-assisted embryo selection should be reframed in light of this moral disruption.

show abstract

Deep learning for embryo evaluation using time-lapse: a systematic review of diagnostic test accuracy

Cited by 2 publications

References 49 publications

An artificial intelligence algorithm to select most viable embryos considering current process in IVF labs

An artificial intelligence algorithm to select most viable embryos considering current process in IVF labs

Embryo selection, AI and reproductive choice

Contact Info

Product

Resources

About