Advertisement

Supporting first FSH dosage for ovarian stimulation with machine learning

  • Nuria Correa
    Affiliations
    Clínica Eugin-Eugin Group, Carrer de Balmes 236, Barcelona 08006, Spain

    Instituto de Investigación en Inteligencia Artificial, Consejo Superior de Investigaciones Científicas (IIIA-CSIC), Campus de la UAB, Carrer de Can Planas, Zona 2, Cerdanyola de Valles Barcelona 08193, Spain

    Universitat Autònoma de Barcelona (UAB), Plaça Cívica, Bellaterra Barcelona 08193, Spain
    Search for articles by this author
  • Jesus Cerquides
    Affiliations
    Instituto de Investigación en Inteligencia Artificial, Consejo Superior de Investigaciones Científicas (IIIA-CSIC), Campus de la UAB, Carrer de Can Planas, Zona 2, Cerdanyola de Valles Barcelona 08193, Spain
    Search for articles by this author
  • Josep Lluis Arcos
    Affiliations
    Instituto de Investigación en Inteligencia Artificial, Consejo Superior de Investigaciones Científicas (IIIA-CSIC), Campus de la UAB, Carrer de Can Planas, Zona 2, Cerdanyola de Valles Barcelona 08193, Spain
    Search for articles by this author
  • Rita Vassena
    Correspondence
    Corresponding author.
    Affiliations
    Clínica Eugin-Eugin Group, Carrer de Balmes 236, Barcelona 08006, Spain
    Search for articles by this author

      HIGHLIGHTS

      • We developed a ML model to recommend first FSH dosage for all types of patients.
      • The model performance surpassed the clinicians’ in both development and validation.
      • The model can serve as quality check, second opinion or learning tool for trainees.

      Abstract

      Research question

      Is it possible to identify accurately the optimal first dose of FSH in ovarian stimulation by means of a machine learning model?

      Design

      Observational study (2011–2021) including first IVF cycles with own oocytes. A total of 2713 patients from five private reproductive centres were included in the development phase (2011–2019) and 774 in the validation phase (2020–2021). Predictor variables included age, BMI, AMH, AFC and previous live births. Performance was measured with a proposed score based on the number of MII oocytes retrieved and dose received, recommended, or both.

      Results

      The included cycles were from women aged 37.7 ± 4.4 years (18–45 years), with a BMI of 23.5 ± 4.2 kg/m2, AMH of 2.4 ± 2.3 ng/ml, AFC of 11.3 ± 7.6, and an average number of MII obtained 6.9 ± 5.4. The model reached a mean performance score of 0.87 (95% CI 0.86 to 0.88) in the development phase, significantly better than for doses prescribed by clinicians for the same patients (0.83, 95% CI 0.82 to 0.84; P = 2.44 e-10). Mean performance score of the model recommendations was 0.89 (95% CI 0.88 to 0.90) in the validation phase, also significantly better than clinicians (0.84, 95% CI 0.82 to 0.86; P = 3.81 e-05). The model was shown to surpass the performance of standard practice.

      Conclusion

      This machine learning model could be used as a training and learning tool for new clinicians, and as quality control for experienced clinicians.

      KEYWORDS

      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'

      Subscribe:

      Subscribe to Reproductive BioMedicine Online
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect

      References

        • Allegra A.
        • Marino A.
        • Volpes A.
        • Coffaro F.
        • Scaglione P.
        • Gullo S.
        • La Marca A.
        A randomized controlled trial investigating the use of a predictive nomogram for the selection of the FSH starting dose in IVF/ICSI cycles.
        Reproductive BioMedicine Online. 2017; 34: 429-438https://doi.org/10.1016/j.rbmo.2017.01.012
        • Bastu E.
        • Celik C.
        • Keskin G.
        • Buyru F.
        Evaluation of embryo transfer time (day 2 vs day 3) after imposed single embryo transfer legislation: When to transfer?.
        Journal of Obstetrics and Gynaecology. 2013; 33: 387-390https://doi.org/10.3109/01443615.2012.761186
        • Chamayou S.
        • Sicali M.
        • Alecci C.
        • Ragolia C.
        • Liprino A.
        • Nibali D.
        • Guglielmino A.
        The accumulation of vitrified oocytes is a strategy to increase the number of euploid available blastocysts for transfer after preimplantation genetic testing.
        Journal of Assisted Reproduction and Genetics. 2017; 34: 479-486https://doi.org/10.1007/s10815-016-0868-0
        • Chen Y.
        • Wang Q.
        • Zhang Y.
        • Han X.
        • Li D.
        • Zhang C.
        Cumulative live birth and surplus embryo incidence after frozen-thaw cycles in PCOS: how many oocytes do we need?.
        Journal of Assisted Reproduction and Genetics. 2017; 34 (huinanhan): 1153-1159https://doi.org/10.1007/s10815-017-0959-6
        • De Geyter C.
        • Calhaz-Jorge C.
        • Kupka M.S.
        • Wyns C.
        • Mocanu E.
        • Motrenko T.
        • Baranowski R.
        ART in Europe, 2014: Results generated from European registries by ESHRE.
        Human Reproduction. 2018; 33: 1586-1601https://doi.org/10.1093/humrep/dey242
        • Drakopoulos P.
        • Blockeel C.
        • Stoop D.
        • Camus M.
        • De Vos M.
        • Tournaye H.
        • Polyzos N.P.
        Conventional ovarian stimulation and single embryo transfer for IVF/ICSI. How many oocytes do we need to maximize cumulative live birth rates after utilization of all fresh and frozen embryos?.
        Human Reproduction. 2016; 31: 370-376https://doi.org/10.1093/humrep/dev316
        • Ebid A.H.I.M.
        • Motaleb S.M.A.
        • Mostafa M.I.
        • Soliman M.M.A.
        Novel nomogram-based integrated gonadotropin therapy individualization in in vitro fertilization/intracytoplasmic sperm injection: A modeling approach.
        Clinical and Experimental Reproductive Medicine. 2021; 48: 163-173https://doi.org/10.5653/cerm.2020.03909
        • Esteves S.C.
        • Carvalho J.F.
        • Bento F.C.
        • Santos J.
        A novel predictive model to estimate the number of mature oocytes required for obtaining at least one euploid blastocyst for transfer in couples undergoing in vitro fertilization/intracytoplasmic sperm injection: The ART calculator.
        Frontiers in Endocrinology. 2019; 10: 1-14https://doi.org/10.3389/fendo.2019.00099
        • Fleming R.
        • Deshpande N.
        • Traynor I.
        • Yates R.W.S.
        Dynamics of FSH-induced follicular growth in subfertile women: Relationship with age, insulin resistance, oocyte yield and anti-Mullerian hormone.
        Human Reproduction. 2006; 21: 1436-1441https://doi.org/10.1093/humrep/dei499
        • Harrison R.F.
        • Jacob S.
        • Spillane H.
        • Mallon E.
        • Hennelly B.
        A prospective randomized clinical trial of differing starter doses of recombinant follicle-stimulating hormone (follitropin-β) for first time in vitro fertilization and intracytoplasmic sperm injection treatment cycles.
        Fertility and Sterility. 2001; 75: 23-31https://doi.org/10.1016/S0015-0282(00)01643-5
        • Howles C.M.
        • Saunders H.
        • Alam V.
        • Engrand P.
        Predictive factors and a corresponding treatment algorithm for controlled ovarian stimulation in patients treated with recombinant human follicle stimulating hormone (follitropin alfa) during assisted reproduction technology (ART) procedures. An analysis.
        Current Medical Research and Opinion. 2006; 22: 907-918https://doi.org/10.1185/030079906X104678
        • Ji J.
        • Liu Y.
        • Tong X.H.
        • Luo L.
        • Ma J.
        • Chen Z.
        The optimum number of oocytes in IVF treatment: An analysis of 2455 cycles in China.
        Human Reproduction. 2013; 28: 2728-2734https://doi.org/10.1093/humrep/det303
        • La Marca A.
        • Papaleo E.
        • Grisendi V.
        • Argento C.
        • Giulini S.
        • Volpe A.
        Development of a nomogram based on markers of ovarian reserve for the individualisation of the follicle-stimulating hormone starting dose in in vitro fertilisation cycles.
        BJOG: An International Journal of Obstetrics and Gynaecology. 2012; 119: 1171-1179https://doi.org/10.1111/j.1471-0528.2012.03412.x
        • Lledo B.
        • Ortiz J.A.
        • Llacer J.
        • Bernabeu R.
        Pharmacogenetics of ovarian response.
        Pharmacogenomics. 2014; 15: 885-893https://doi.org/10.2217/pgs.14.49
        • Maggiulli R.
        • Cimadomo D.
        • Fabozzi G.
        • Papini L.
        • Dovere L.
        • Ubaldi F.M.
        • Rienzi L.
        The effect of ICSI-related procedural timings and operators on the outcome.
        Human Reproduction. 2020; 35: 32-43https://doi.org/10.1093/humrep/dez234
        • Naether O.G.J.
        • Tandler-Schneider A.
        • Bilger W.
        Individualized recombinant human follicle-stimulating hormone dosing using the CONSORT calculator in assisted reproductive technology: A large, multicenter, observational study of routine clinical practice.
        Drug, Healthcare and Patient Safety. 2015; 7: 69-76https://doi.org/10.2147/DHPS.S77320
        • Nyboe Andersen A.
        • Nelson S.M.
        • Fauser B.C.J.M.
        • García-Velasco J.A.
        • Klein B.M.
        • Arce J.C.
        • Arce J.C.
        Individualized versus conventional ovarian stimulation for in vitro fertilization: a multicenter, randomized, controlled, assessor-blinded, phase 3 noninferiority trial.
        Fertility and Sterility. 2017; 107 (e4): 387-396https://doi.org/10.1016/j.fertnstert.2016.10.033
        • Olivennes F.
        • Trew G.
        • Borini A.
        • Broekmans F.
        • Arriagada P.
        • Warne D.W.
        • Howles C.M.
        Randomized, controlled, open-label, non-inferiority study of the CONSORT algorithm for individualized dosing of follitropin alfa.
        Reproductive BioMedicine Online. 2015; 30: 248-257https://doi.org/10.1016/j.rbmo.2014.11.013
        • Polyzos N.P.
        • Sunkara S.K.
        Sub-optimal responders following controlled ovarian stimulation: An overlooked group?.
        Human Reproduction. 2015; 30: 2005-2008https://doi.org/10.1093/humrep/dev149
        • Pouly J.L.
        • Olivennes F.
        • Massin N.
        • Celle M.
        • Caizergues N.
        • Contard F.
        Usability and utility of the CONSORT calculator for FSH starting doses: A prospective observational study.
        Reproductive BioMedicine Online. 2015; https://doi.org/10.1016/j.rbmo.2015.06.001
        • Steward R.G.
        • Lan L.
        • Shah A.A.
        • Yeh J.S.
        • Price T.M.
        • Goldfarb J.M.
        • Muasher S.J.
        Oocyte number as a predictor for ovarian hyperstimulation syndrome and live birth: An analysis of 256,381 in vitro fertilization cycles.
        Fertility and Sterility. 2014; 101: 967-973https://doi.org/10.1016/j.fertnstert.2013.12.026
        • Sunkara S.K.
        • Rittenberg V.
        • Raine-Fenning N.
        • Bhattacharya S.
        • Zamora J.
        • Coomarasamy A.
        Association between the number of eggs and live birth in IVF treatment: An analysis of 400 135 treatment cycles.
        Human Reproduction. 2011; 26: 1768-1774https://doi.org/10.1093/humrep/der106
        • Vaiarelli A.
        • Cimadomo D.
        • Conforti A.
        • Schimberni M.
        • Giuliani M.
        • D'Alessandro P.
        • Ubaldi F.M.
        Luteal phase after conventional stimulation in the same ovarian cycle might improve the management of poor responder patients fulfilling the Bologna criteria: a case series.
        Fertility and Sterility. 2020; 113: 121-130https://doi.org/10.1016/j.fertnstert.2019.09.012

      Biography

      Núria Correa is senior clinical embryologist and researcher at the R&D department of the Eugin Group. She is a PhD candidate at the Universitat Autònoma de Barcelona, working on a research project centred on the application of artificial intelligence in assisted reproduction.
      Key message
      A machine learning model was trained to recommend first FSH doses for ovarian stimulation. Compared with clinicians, the model achieved consistently better performance scores. The model could be used as a second opinion and as a learning tool for new clinicians to avoid as many non-optimal outcomes as possible.