An artificial intelligence platform to optimize workflow during ovarian stimulation and IVF: process improvement and outcome-based predictions

Published:October 19, 2021DOI:


      • An AI-based suite of tools for scheduling of retrievals and prediction of outcomes is described.
      • The database consisted of 2085 cycles with 1591 complete records over 4731 visits.
      • Algorithms were written using Python, NumPy logistic function, Panda and SKLearn.
      • The algorithm offers options to level-load the clinical and embryology workflow.
      • This suite of tools assigns a single day to monitor and a range of three retrieval days.


      Research question

      Can workflow during IVF be facilitated by artificial intelligence to limit monitoring during ovarian stimulation to a single day and enable level-loading of retrievals?


      The dataset consisted of 1591 autologous cycles in unique patients with complete data including age, FSH, oestradiol and anti-Müllerian concentrations, follicle counts and body mass index. Observations during ovarian stimulation included oestradiol concentrations and follicle diameters. An algorithm was designed to identify the single best day for monitoring and predict trigger day options and total number of oocytes retrieved.


      The mean error to predict the single best day for monitoring was 1.355 days. After identifying the single best day for evaluation, the algorithm identified the trigger date and range of three oocyte retrieval days specified by the earliest and the latest day on which the number of oocytes retrieved was minimally changed with a variance of 0–3 oocytes. Accuracy for prediction of total number of oocytes with baseline testing alone or in combination with data on the day of observation was 0.76 and 0.80, respectively. The sensitivities for estimating the total number and number of mature oocytes based solely on pre-IVF profiles in group I (0–10) were 0.76 and 0.78, and in group II (>10) 0.76 and 0.81, respectively.


      A first-iteration algorithm is described designed to improve workflow, minimize visits and level-load embryology work. This algorithm enables decisions at three interrelated nodal points for IVF workflow management to include monitoring on the single best day, assign trigger days to enable a range of 3 days for level-loading and estimate oocyte number.
      To read this article in full you will need to make a payment


        • Akkus Z.
        • Cai J.
        • Boonrod A.
        • Zeinoddini A.
        • Weston A.D.
        • Philbrick K.A.
        • Erickson B.J.
        A Survey of Deep-Learning Applications in Ultrasound: Artificial Intelligence-Powered Ultrasound for Improving Clinical Workflow.
        J. Am. Coll. Radiol. 2019; 16 (Sep): 1318-1328
        • Dent E.B.
        • Bozeman P.
        Discovering the foundational philosophies, practices, and influences of modern management theory.
        Journal of Management History. 2014; 10: 145-163
      1. Elliott, D.J., Young, R.S., Brice, J., Aguiar, R., Kolm, P. Effect of hospitalist workload on the quality and efficiency of care.

      2. JAMA Intern Med. 2014 May; 174(5):786-93.

        • Hariton E.
        • Chi E.A.
        • Chi G.
        • Jerrine R.
        • Morris J.R.
        • Braatz J.
        • Rajpurkar P.
        • Rosen M.A.
        Machine learning algorithm can optimize the day of trigger to improve in vitro fertilization outcomes.
        Fertil. Steril. e PUB. 2021;
        • Hauzman E.E.
        • Zapata A.
        • Bermejo A.
        • Bermejo A.
        • Iglesias C.
        • Pellicer A.
        • Garcia-Velasco J.A.
        Cycle scheduling for in vitro fertilization with oral contraceptive pills versus oral estradiol valerate: a randomized, controlled trial.
        Reprod. Biol. Endocrinol. 2013; 11: 96
        • Hribar M.R.
        • Read-Brown S.
        • Goldstein I.H.
        • Reznick L.G.
        • Lombardi L.
        • Parikh M.
        • Chamberlain W.
        • Chiang M.F.
        Secondary use of electronic health record data for clinical workflow analysis.
        J. Am. Med. Inform. Assoc. 2018; 25 (Jan 1): 40-46
      3. Kaiser, K., Marcos, M. Leveraging workflow control patterns in the domain of clinical practice guidelines.

      4. BMC Med. Inform. Decis. Mak. 2016; 16:20.

        • Kirkpatrick A.W.
        • McKee J.L.
        • Volpicelli
        • Ma I.W.
        The Potential for Remotely Mentored Patient-Performed Home Self-Monitoring for New Onset Alveolar-Interstitial Lung Disease.
        Telemed. J E Health. 2020; 26 (Oct): 1304-1307
      5. Letterie. G.S. Three ways of knowing: the integration of expertise, evidence based medicine and artificial intelligence in assisted reproductive technologies. J. Assist. Reprod. Genet (in press)

        • Letterie G.S.
        • Mac Donald A.W.
        Artificial Intelligence in IVF: A computer decision support system for day to day management of ovarian stimulation during in vitro fertilization.
        Fertil. Steril. 2020; 114: 1026-1031
        • Mecklai K.
        • Smith N.
        • Stern A.D.
        • Kramer D.B.
        Remote patient monitoring-overdue or overused.
        N. Engl. J. Med. 2021; 384: 1381-1383
        • Pereira I.
        • von Horn P.
        • Depenbusch M.
        • Schultze-Mosgau A.
        • Griesinger G.
        Self-operated endovaginal telemonitoring: a prospective, clinical validation study Fertil. Steril. 2016; 106 (Aug): 306-310
        • Raza Khan U.
        • Zia T.A.
        • Pearce C.
        • Perera K.
        The MyHealthRecord System Impacts on Patient Workflow in General Practices.
        Stud. Health Techno. Inform. 2019; 266 (Aug 8): 162-167
        • Thompson C.B.
        The literature of scientific management.
        Quarterly J. Economics. 1914; 28: 506-557
        • Vankipuram M.
        • Kahol K.
        • Cohen T.
        • Patel V.L.
        Toward automated workflow analysis and visualization in clinical environments.
        J. Biomed. Inform. 2011; 44 (Jun): 432-440
        • Veinot T.C.
        • Senteio C.R.
        • Hanauer D.
        • Lowery J.C.
        Comprehensive process model of clinical information interaction in primary care: results of a "best-fit" framework synthesis.
        J. Am. Med. Inform. Assoc. 2018; 25: 746-758
        • Viardot-Foucault V.
        • Nadarajah S.
        • Lye W.K.
        • Tan H.H.
        GnRH antagonist pre-treatment: one centre's experience for IVF-ICSI cycle scheduling.
        Reprod. Biomed. Online. 2015; 30 (Apr): 366-372
        • Weigl M.
        • Müller A.
        • Holland S.
        • Woloshynowych M.
        Work conditions, mental workload and patient care quality: a multisource study in the emergency department.
        BMJ Qual. Saf. 2016; 25 (Jul): 499-508
        • Zheng K.
        • Ratwani R.M.
        • Adler-Milstein J.
        Studying Workflow and Workarounds in Electronic Health Record-Supported Work to Improve Health System Performance.
        Ann. Intern. Med. 2020; 172 (Jun 2): S116-S122


      Gerard Letterie: Gerard Letterie is a board-certified Reproductive Endocrinologist, a partner in Seattle Reproductive Medicine, Seattle WA and co-founder of Quick Step Analytics LLC. He completed a Residency in Obstetrics and Gynecology at Walter Reed Hospital in Washington DC and a fellowship in REI at the National Institutes of Health, Bethesda, MD, USA.
      Key Message
      A first-iteration algorithm is described to improve workflow, minimize visits during ovarian stimulation for IVF and level-load embryology work. This artificial intelligence platform is data driven and offers an opportunity to optimize workflow during ovarian stimulation and IVF, and to reduce workload in clinic and laboratory settings without compromising care.