Artificial intelligence assessment of time-lapse images can predict with 77% accuracy whether a human embryo capable of achieving a pregnancy will miscarry

Hariharan, Rishabh; He, Peter; Meseguer, M.; Toschi, M.; Rocha, José Celso; Zaninović, Nikica; Malmsten, Jonas; Hickman, Cristina

doi:10.1016/j.fertnstert.2019.07.230

Cited by 7 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…AI has been applied to many different aspects of perinatal medicine. The included studies described AI applications in predicting general pregnancy risk characteristics, prenatal diagnoses, pregnancy hypertension disorder, fetal growth, stillbirth, gestational diabetes, preterm deliveries, the delivery route, and other conditions [ 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 , 69 , 69 , 70 , 71 , 72 , 73 , 74 , 75 , 76 , 77 , 78 , 79 ]. The studies were divided into n...…”

Section: Resultsmentioning

confidence: 99%

“…The next group of examined articles included four studies about stillbirth prediction ( Table 6 ) [ 48 , 49 , 50 , 51 ]. MLR [ 48 ], artificial intelligence analysis of time-lapse (TLM) embryo images [ 49 ], LR, ANNs, the gradient boosting DT [ 50 ], regularized logistic regression, the DT based on classification and regression trees, the RF, extreme gradient boosting (XGBoost), and the multilayer perceptron neural network were used in those studies [ 51 ].…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Application of Artificial Intelligence in Screening for Adverse Perinatal Outcomes—A Systematic Review

et al. 2022

View full text Add to dashboard Cite

(1) Background: AI-based solutions could become crucial for the prediction of pregnancy disorders and complications. This study investigated the evidence for applying artificial intelligence methods in obstetric pregnancy risk assessment and adverse pregnancy outcome prediction. (2) Methods: Authors screened the following databases: Pubmed/MEDLINE, Web of Science, Cochrane Library, EMBASE, and Google Scholar. This study included all the evaluative studies comparing artificial intelligence methods in predicting adverse pregnancy outcomes. The PROSPERO ID number is CRD42020178944, and the study protocol was published before this publication. (3) Results: AI application was found in nine groups: general pregnancy risk assessment, prenatal diagnosis, pregnancy hypertension disorders, fetal growth, stillbirth, gestational diabetes, preterm deliveries, delivery route, and others. According to this systematic review, the best artificial intelligence application for assessing medical conditions is ANN methods. The average accuracy of ANN methods was established to be around 80–90%. (4) Conclusions: The application of AI methods as a digital software can help medical practitioners in their everyday practice during pregnancy risk assessment. Based on published studies, models that used ANN methods could be applied in APO prediction. Nevertheless, further studies could identify new methods with an even better prediction potential.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Application of Artificial Intelligence in Screening for Adverse Perinatal Outcomes—A Systematic Review

et al. 2022

View full text Add to dashboard Cite

show abstract

“…CNN algorithms are a type of deep-learning model that attempts to simply replicate the human visual cortex with a simulated network of connected neuron layers (neural network) that, through iterative training, transforms input data into the desired output labels. There have been considerable studies on the utility of machine learning and AI-based image analysis on the selection of embryos for prediction of euploidy status, implantation potential and incidence of miscarriage (Barnes et al, 2023, Diakiw et al, 2022, Duval et al, 2023, Hariharan et al, 2019, Tran et al, 2018, VerMilyea et al, 2020). Studies have also proven the application of ML in the selection and assessment of sperm for use in ICSI by tacking sperm correlated with better quality blastocysts (Joshi et al, 2023, Mendizabal-Ruiz et al, 2022).…”

Section: Discussionmentioning

confidence: 99%

AI facilitated sperm detection in azoospermic samples for use in ICSI

Goss,

Vasilescu,

Vasilescu

et al. 2023

Preprint

View full text Add to dashboard Cite

Research question: Can artificial intelligence (AI) improve efficiency and efficacy of sperm searches in azoospermic samples? Design: This two-phase proof-of-concept study beginning with a training phase using 8 azoospermic patients (>10000 sperm images) to provide a variety of surgically collected samples for sperm morphology and debris variation to train a convolutional neural network to identify sperm. Secondly, side-by-side testing on 2 cohorts, an embryologist versus the AI identifying all sperm in still images (cohort 1, N=4, 2660 sperm) and then a side-by-side test with deployment of the AI model on an ICSI microscope and the embryologist performing a search with and without the aid of the AI (cohort 2, N=4, >1300 sperm). Time taken, accuracy and precision of sperm identification was measured. Results: In cohort 1, the AI model showed improvement in time-taken to identify all sperm per field of view (0.019+-0.30 x 10-5s versus 36.10+-1.18s, P<0.0001) and improved accuracy (91.95+-0.81% vs 86.52+-1.34%, P<0.001) compared to an embryologist. From a total of 688 sperm in all samples combined, 560 were found by an embryologist and 611 were found by the AI in <1000th of the time. In cohort 2, the AI-aided embryologist took significantly less time per droplet (98.90+-3.19s vs 168.7+-7.84s, P<0.0001) and found 1396 sperm, while 1274 were found without AI, although no significant difference was observed. Conclusions: AI-powered image analysis has the potential for seamless integration into laboratory workflows, and to reduce time to identify and isolate sperm from surgical sperm samples from hours to minutes, thus increasing success rates from these treatments.

show abstract

Super-Focus: Domain Adaptation for Embryo Imaging via Self-supervised Focal Plane Regression

Jacques²,

Chambost³

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Artificial intelligence assessment of time-lapse images can predict with 77% accuracy whether a human embryo capable of achieving a pregnancy will miscarry

Cited by 7 publications

References 0 publications

Application of Artificial Intelligence in Screening for Adverse Perinatal Outcomes—A Systematic Review

Application of Artificial Intelligence in Screening for Adverse Perinatal Outcomes—A Systematic Review

AI facilitated sperm detection in azoospermic samples for use in ICSI

Super-Focus: Domain Adaptation for Embryo Imaging via Self-supervised Focal Plane Regression

Contact Info

Product

Resources

About