A novel machine-learning framework based on early embryo morphokinetics identifies a feature signature associated with blastocyst development.
Artificial intelligence
Assisted reproduction
Blastocyst development
Embryo developmental patterns
Machine learning
Journal
Journal of ovarian research
ISSN: 1757-2215
Titre abrégé: J Ovarian Res
Pays: England
ID NLM: 101474849
Informations de publication
Date de publication:
15 Mar 2024
15 Mar 2024
Historique:
received:
25
08
2023
accepted:
16
02
2024
medline:
18
3
2024
pubmed:
16
3
2024
entrez:
16
3
2024
Statut:
epublish
Résumé
Artificial Intelligence entails the application of computer algorithms to the huge and heterogeneous amount of morphodynamic data produced by Time-Lapse Technology. In this context, Machine Learning (ML) methods were developed in order to assist embryologists with automatized and objective predictive models able to standardize human embryo assessment. In this study, we aimed at developing a novel ML-based strategy to identify relevant patterns associated with the prediction of blastocyst development stage on day 5. We retrospectively analysed the morphokinetics of 575 embryos obtained from 80 women who underwent IVF at our Unit. Embryo morphokinetics was registered using the Geri plus® time-lapse system. Overall, 30 clinical, morphological and morphokinetic variables related to women and embryos were recorded and combined. Some embryos reached the expanded blastocyst stage on day 5 (BL Group, n = 210), some others did not (nBL Group, n = 365). The novel EmbryoMLSelection framework was developed following four-steps: Feature Selection, Rules Extraction, Rules Selection and Rules Evaluation. Six rules composed by a combination of 8 variables were finally selected, and provided a predictive power described by an AUC of 0.84 and an accuracy of 81%. We provided herein a new feature-signature able to identify with an high performance embryos with the best developmental competence to reach the expanded blastocyst stage on day 5. Clear and clinically relevant cut-offs were identified for each considered variable, providing an objective tool for early embryo developmental assessment.
Sections du résumé
BACKGROUND
BACKGROUND
Artificial Intelligence entails the application of computer algorithms to the huge and heterogeneous amount of morphodynamic data produced by Time-Lapse Technology. In this context, Machine Learning (ML) methods were developed in order to assist embryologists with automatized and objective predictive models able to standardize human embryo assessment. In this study, we aimed at developing a novel ML-based strategy to identify relevant patterns associated with the prediction of blastocyst development stage on day 5.
METHODS
METHODS
We retrospectively analysed the morphokinetics of 575 embryos obtained from 80 women who underwent IVF at our Unit. Embryo morphokinetics was registered using the Geri plus® time-lapse system. Overall, 30 clinical, morphological and morphokinetic variables related to women and embryos were recorded and combined. Some embryos reached the expanded blastocyst stage on day 5 (BL Group, n = 210), some others did not (nBL Group, n = 365).
RESULTS
RESULTS
The novel EmbryoMLSelection framework was developed following four-steps: Feature Selection, Rules Extraction, Rules Selection and Rules Evaluation. Six rules composed by a combination of 8 variables were finally selected, and provided a predictive power described by an AUC of 0.84 and an accuracy of 81%.
CONCLUSIONS
CONCLUSIONS
We provided herein a new feature-signature able to identify with an high performance embryos with the best developmental competence to reach the expanded blastocyst stage on day 5. Clear and clinically relevant cut-offs were identified for each considered variable, providing an objective tool for early embryo developmental assessment.
Identifiants
pubmed: 38491534
doi: 10.1186/s13048-024-01376-6
pii: 10.1186/s13048-024-01376-6
pmc: PMC10941455
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
63Informations de copyright
© 2024. The Author(s).
Références
Mol Hum Reprod. 2017 May 1;23(5):292-303
pubmed: 28204536
Syst Biol Reprod Med. 2021 Feb;67(1):64-78
pubmed: 33719832
Reprod Biomed Online. 2017 Nov;35(5):511-520
pubmed: 28736152
F S Rep. 2022 Apr 15;3(2):116-123
pubmed: 35789724
Hum Reprod. 2015 Jan;30(1):28-36
pubmed: 25376459
Reprod Biomed Online. 2019 Jul;39(1):19-26
pubmed: 31109893
Reprod Biomed Online. 2022 Mar;44(3):435-448
pubmed: 35027326
Hum Reprod. 2013 Dec;28(12):3215-21
pubmed: 24070998
Fertil Steril. 2017 Nov;108(5):722-729
pubmed: 29101997
Hum Reprod. 2013 Oct;28(10):2643-51
pubmed: 23900207
J Assist Reprod Genet. 2015 Jul;32(7):1025-30
pubmed: 26126876
J Clin Med. 2023 Feb 23;12(5):
pubmed: 36902592
Reprod Biomed Online. 2012 Nov;25(5):474-80
pubmed: 22995750
J Assist Reprod Genet. 2023 Feb;40(2):223-234
pubmed: 36609943
Hum Reprod Open. 2018 Dec 20;2018(4):hoy022
pubmed: 30895262
J Clin Med. 2022 Feb 06;11(3):
pubmed: 35160310
PLoS One. 2017 Jun 1;12(6):e0178720
pubmed: 28570713
Fertil Steril. 2017 Mar;107(3):641-648.e2
pubmed: 28108009
Hum Reprod. 2020 May 1;35(5):1045-1053
pubmed: 32358601
Reprod Biomed Online. 2022 Jan;44(1):39-48
pubmed: 34819249
J Assist Reprod Genet. 2019 May;36(5):875-887
pubmed: 30710230
Nat Biotechnol. 2010 Oct;28(10):1115-21
pubmed: 20890283
J Clin Med. 2020 Jun 18;9(6):
pubmed: 32570935
Hum Reprod. 2023 Mar 1;38(3):387-399
pubmed: 36644923
Hum Reprod. 2017 Feb;32(2):307-314
pubmed: 28031323
Fertil Steril. 2001 Jun;75(6):1126-30
pubmed: 11384637
Hum Reprod. 2023 Oct 3;38(10):1901-1909
pubmed: 37649342
Hum Reprod. 2007 Feb;22(2):548-57
pubmed: 17095516
J Assist Reprod Genet. 2020 Jun;37(6):1409-1420
pubmed: 32436046
Fertil Steril. 2016 Feb;105(2):376-84.e9
pubmed: 26598211
Cochrane Database Syst Rev. 2015 Feb 27;(2):CD011320
pubmed: 25721906
Hum Reprod Open. 2020 Mar 19;2020(2):hoaa008
pubmed: 32206731
Hum Reprod. 2011 Oct;26(10):2658-71
pubmed: 21828117
Ultrasound Obstet Gynecol. 2014 Oct;44(4):394-401
pubmed: 24890582
Fertil Steril. 2013 Nov;100(5):1270-6
pubmed: 23931964
Fertil Steril. 2005 Jan;83(1):49-53
pubmed: 15652886
Acta Inform Med. 2019 Sep;27(3):205-211
pubmed: 31762579
Clin Exp Reprod Med. 2016 Jun;43(2):133-8
pubmed: 27358833
Reprod Biomed Online. 2013 Jan;26(1):42-9
pubmed: 23177416
Hum Reprod. 2011 Jun;26(6):1270-83
pubmed: 21502182
J Clin Med. 2020 Sep 12;9(9):
pubmed: 32932604
Hum Reprod. 2014 Dec;29(12):2650-60
pubmed: 25344070
Front Endocrinol (Lausanne). 2018 Jun 29;9:327
pubmed: 30008696
Fertil Steril. 2017 Mar;107(3):613-621
pubmed: 28069186
J Ovarian Res. 2019 Aug 9;12(1):73
pubmed: 31399038
PLoS One. 2022 Apr 25;17(4):e0267241
pubmed: 35468159
Reprod Biol Endocrinol. 2011 Aug 09;9:112
pubmed: 21824441
Hum Reprod Open. 2017 Aug 04;2017(2):hox011
pubmed: 31486806
Fertil Steril. 2020 Nov;114(5):914-920
pubmed: 33160513
Hum Reprod Update. 2023 Sep 5;29(5):570-633
pubmed: 37192834
Hum Reprod. 2013 Jun;28(6):1528-35
pubmed: 23482339
Front Endocrinol (Lausanne). 2022 Dec 09;13:1082597
pubmed: 36568064
Ultrasound Obstet Gynecol. 2016 Sep;48(3):289-95
pubmed: 26924732
Front Endocrinol (Lausanne). 2019 Feb 20;10:94
pubmed: 30842755
Hum Reprod. 2020 Apr 28;35(4):770-784
pubmed: 32240301
Reproduction. 2019 Oct;158(4):R139-R154
pubmed: 30970326
Hum Reprod. 2023 Sep 5;38(9):1690-1699
pubmed: 37381073
Fertil Steril. 2013 Aug;100(2):412-9.e5
pubmed: 23721712