Interpretable, not black-box, artificial intelligence should be used for embryo selection
Open accessThe DOI matches its Crossref record; 12 of the 12 references it cites were found in a registry or an index.
Abstract
Artificial intelligence (AI) techniques are starting to be used in IVF, in particular for selecting which embryos to transfer to the woman. AI has the potential to process complex data sets, to be better at identifying subtle but important patterns, and to be more objective than humans when evaluating embryos. However, a current review of the literature shows much work is still needed before AI can be ethically implemented for this purpose. No randomized controlled trials (RCTs) have been published, and the efficacy studies which exist demonstrate that algorithms can broadly differentiate well between ‘good-’ and ‘poor-’ quality embryos but not necessarily between embryos of similar quality, which is the actual clinical need. Almost universally, the AI models were opaque (‘black-box’) in that at least some part of the process was uninterpretable. This gives rise to a number of epistemic and ethical concerns, including problems with trust, the possibility of using algorithms that generalize poorly to different populations, adverse economic implications for IVF clinics, potential misrepresentation of patient values, broader societal implications, a responsibility gap in the case of poor selection choices and introduction of a more paternalistic decision-making process. Use of interpretable models, which are constrained so that a human can easily understand and explain them, could overcome these concerns. The contribution of AI to IVF is potentially significant, but we recommend that AI models used in this field should be interpretable, and rigorously evaluated with RCTs before implementation. We also recommend long-term follow-up of children born after AI for embryo selection, regulatory oversight for implementation, and public availability of data and code to enable research teams to independently reproduce and validate existing models.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Rudin, C. (2019) · Nature Machine Intelligence · doi:10.1038/s42256-019-0048-x · Cited by 10,561 · cited 5×
“accuracy relative to their black-box counterparts (Chen et al., 2019; Rudin, 2019).”
BackgroundInfluentialSelf-citationCharles, C., Gafni, A., & Whelan, T. (1997) · Social Science & Medicine · doi:10.1016/s0277-9536(96)00221-3 · Cited by 4,137 · cited 1×
“Over the past few decades, clinical practice has shifted from a paternalistic model to a model of shared decision-making aimed at promoting patient autonomy (Charles et al., 1997).”
BackgroundAnanny, M., & Crawford, K. (2018) · New Media & Society · doi:10.1177/1461444816676645 · Cited by 1,836
Complacency and Bias in Human Use of Automation: An Attentional Integration
Parasuraman, R., & Manzey, D. H. (2010) · Human Factors: The Journal of the Human Factors and Ergonomics Society · doi:10.1177/0018720810376055 · Cited by 1,710 · cited 1×
“It is known that people tend to be complacent about use of automation (Parasuraman and Manzey, 2010) and accepting AI in a role when they are familiar with AI in that role or when they believe it performs well (Kramer et al., 2018).”
BackgroundNagendran, M., Chen, Y., Lovejoy, C. A., Gordon, A. C., Komorowski, M., Harvey, H., Topol, E. J., Ioannidis, J. P. A., Collins, G. S., & Maruthappu, M. (2020) · BMJ · doi:10.1136/bmj.m689 · Cited by 1,152 · cited 1×
“The lack of RCTs appears to be typical of much of AI in medicine (Nagendran et al., 2020).”
BackgroundCounterfactual explanations without opening the black box: automated decisions and the GDPR
Wachter, S., Mittelstadt, B., & Russell, C. (2017) · SSRN Electronic Journal · doi:10.2139/ssrn.3063289 · Cited by 1,061
Potential liability for physicians using artificial intelligence
Price, W. N., Gerke, S., & Cohen, I. G. (2019) · JAMA · doi:10.1001/jama.2019.15064 · Cited by 579
Meaningful information and the right to explanation
Selbst, A. D., & Powles, J. (2017) · International Data Privacy Law · doi:10.1093/idpl/ipx022 · Cited by 516
Rudin, C., & Radin, J. (2019) · Harvard Data Science Review · doi:10.1162/99608f92.5a8a3a3d · Cited by 432 · cited 1×
“Explaining a black-box model lends unwarranted authority to it (Rudin and Radin, 2019), which might deter development of an inherently interpretable model with the same accuracy.”
BackgroundSelf-citationArtificial Intelligence and Patient-Centered Decision-Making
Bjerring, J. C., & Busch, J. (2021) · Philosophy & Technology · doi:10.1007/s13347-019-00391-6 · Cited by 274 · cited 2×
“Opaque AI models compromise shared decision-making due to the inability of the clinician and patient to understand the model’s decision (Bjerring and Busch, 2020), for example, information as to why a particular embryo is selected or is not selected (such as, the number and symmetry of the cells, or if the cells are f…”
BackgroundComputer knows best? The need for value-flexibility in medical AI
McDougall, R. J. (2019) · Journal of Medical Ethics · doi:10.1136/medethics-2018-105118 · Cited by 262 · cited 1×
“There have been calls for systems to be ‘value flexible’ (McDougall, 2019).”
BackgroundArtificial intelligence in healthcare: a critical analysis of the legal and ethical implications
Schönberger, D. (2019) · International Journal of Law and Information Technology · doi:10.1093/ijlit/eaz004 · Cited by 230
Research works printed but not linked: 35
- Afnan, 2020, Generating translatable evidence to improve patient care: the contribution of human factors, Reprod Biomed Online, 41, 353, 10.1016/j.rbmo.2020.04.025 · DOI
- Agar, 2004, Liberal Eugenics: In Defense of Human Enhancement, 10.1002/9780470775004 · DOI
- Blackmer, 2018, EU general data protection regulation, OJEU, 2014, 45
- Bronet, 2015, Is there a relationship between time-lapse parameters and embryo sex?, Fertil Steril, 103, 396, 10.1016/j.fertnstert.2014.10.050 · DOI
- Buda, 2021, A data set and deep learning algorithm for the detection of masses and architectural distortions in digital breast tomosynthesis images, JAMA Netw Open, 4, e2119100, 10.1001/jamanetworkopen.2021.19100 · DOI
- Chavez-Badiola, 2020, Predicting pregnancy test results after embryo transfer by image feature extraction and analysis using machine learning, Sci Rep, 10, 4394, 10.1038/s41598-020-61357-9 · DOI
- Curchoe, 2020, Evaluating predictive models in reproductive medicine, Fertil Steril, 114, 921, 10.1016/j.fertnstert.2020.09.159 · DOI
- Elwyn, 2003, Shared decision making: developing the OPTION scale for measuring patient involvement, Qual Saf Health Care, 12, 93, 10.1136/qhc.12.2.93 · DOI · Paper page
- Fernandez, 2020, Artificial intelligence in the IVF laboratory: overview through the application of different types of algorithms for the classification of reproductive data, J Assist Reprod Genet, 37, 2359, 10.1007/s10815-020-01881-9 · DOI
- Gardner, 2000, Blastocyst score affects implantation and pregnancy outcome: towards a single blastocyst transfer, Fertil Steril, 73, 1155, 10.1016/S0015-0282(00)00518-5 · DOI · Paper page
- Gardner, 2015, Diagnosis of human preimplantation embryo viability, Hum Reprod Update, 21, 727, 10.1093/humupd/dmu064 · DOI · Paper page
- Garvelink, 2019, Decisional conflict scale use over 20 years: the anniversary review, Med Decis Making, 39, 301, 10.1177/0272989X19851345 · DOI · Paper page
- Huang, 2019, Is differences in embryo morphokinetic development significantly associated with human embryo sex?, Biol Reprod, 100, 618, 10.1093/biolre/ioy229 · doi:10.1093/biolre/ioy229
- Kan-Tor, 2020, Can deep learning automatically predict fetal heart pregnancy with almost perfect accuracy?, Hum Reprod, 35, 1473, 10.1093/humrep/deaa083 · doi:10.1093/humrep/deaa083
- Kemper, 2020, Preimplantation genetic testing for aneuploidy: are we examining the correct outcomes?, Hum Reprod, 35, 2408, 10.1093/humrep/deaa224 · DOI
- Khosravi, 2019, Deep learning enables robust assessment and selection of human blastocysts after in vitro fertilization, Npj Digit Med, 2, 1, 10.1038/s41746-019-0096-y · DOI · Paper page
- Leahy, 2020, Automated measurements of key morphological features of human embryos for IVF, Med Image Comput Comput Assist Interv, 12265, 25
- Légaré, 2010, Are you SURE? Assessing patient decisional conflict with a 4-item screening test, Can Fam Physician, 56, e308
- Liu, 2016, Time-lapse deselection model for human day 3 in vitro fertilization embryos: the combination of qualitative and quantitative measures of embryo growth, Fertil Steril, 105, 656, 10.1016/j.fertnstert.2015.11.003 · DOI
- Liu, 2019, Assessing efficacy of day 3 embryo time-lapse algorithms retrospectively: impacts of dataset type and confounding factors, Hum Fertil (Camb), 22, 182, 10.1080/14647273.2018.1425919 · doi:10.1080/14647273.2018.1425919
- Milewski, 2017, How much information about embryo implantation potential is included in morphokinetic data? A prediction model based on artificial neural networks and principal component analysis, Adv Med Sci, 62, 202, 10.1016/j.advms.2017.02.001 · doi:10.1016/j.advms.2017.02.001
- Morales, 2008, Bayesian classification for the selection of in vitro human embryos using morphological and clinical data, Comput Methods Programs Biomed, 90, 104, 10.1016/j.cmpb.2007.11.018 · DOI
- Parfit, 1984, Reasons and Persons
- Raef, 2020, Computational prediction of implantation outcome after embryo transfer, Health Inform J, 26, 1810, 10.1177/1460458219892138 · DOI
- Rosenwaks, 2020, Artificial intelligence in reproductive medicine: a fleeting concept or the wave of the future?, Fertil Steril, 114, 905, 10.1016/j.fertnstert.2020.10.002 · doi:10.1016/j.fertnstert.2020.10.002
- Savulescu, 2006, Behavioural genetics: why eugenic selection is preferable to enhancement, J Appl Philos, 23, 157, 10.1111/j.1468-5930.2006.00336.x · DOI
- Savulescu, 2009, The moral obligation to create children with the best chance of the best life, Bioethics, 23, 274, 10.1111/j.1467-8519.2008.00687.x · DOI · Paper page
- Savulescu, 2016, The Oxford Handbook of Reproductive Ethics
- Savulescu, 2001, Procreative beneficence: why we should select the best children, Bioethics, 15, 413, 10.1111/1467-8519.00251 · DOI · Paper page
- Tarín, 2014, Changes in sex ratio from fertilization to birth in assisted-reproductive-treatment cycles, Reprod Biol Endocrinol, 12, 56, 10.1186/1477-7827-12-56 · DOI
- Tran, 2019, Deep learning as a predictive tool for fetal heart pregnancy following time-lapse incubation and blastocyst transfer, Hum Reprod, 34, 1011, 10.1093/humrep/dez064 · DOI · Paper page
- Voter, 2021, Diagnostic accuracy and failure mode analysis of a deep learning algorithm for the detection of intracranial hemorrhage, J Am Coll Radiol, 10.1016/j.jacr.2021.03.005 · DOI
- Wang, 2018, Euploidy in relation to blastocyst sex and morphology, J Assist Reprod Genet, 35, 1565, 10.1007/s10815-018-1262-x · DOI
- Wilkinson, 2019, Do à la carte menus serve infertility patients? The ethics and regulation of in vitro fertility add-ons, Fertil Steril, 112, 973, 10.1016/j.fertnstert.2019.09.028 · DOI
- Zmuidinaite, 2021, Current advancements in noninvasive profiling of the embryo culture media secretome, Int J Mol Sci, 22, 1, 10.3390/ijms22052513 · DOI
Cited by
Ethics of artificial intelligence in medicine
Savulescu, J., Giubilini, A., Vandersluis, R., & Mishra, A. (2024) · Singapore Medical Journal · doi:10.4103/singaporemedj.smj-2023-279 · Cited by 68
“These include “broad consent”, which enables individuals to consent to a wide range of more or less specified future uses; (10) meta-consent, where individuals retain control over which type of consent they will want to give to future uses (e.g. blanket consent on certain types of uses, but…”
Self-citationThe ethical requirement of explainability for AI-DSS in healthcare: a systematic review of reasons
Freyer, N., Groß, D., & Lipprandt, M. (2024) · BMC Medical Ethics · doi:10.1186/s12910-024-01103-2 · Cited by 67
“Out of these, 3 explicitly advocate for ad-hoc explainable models [51, 57, 58].”
BackgroundMethodInfluentialSchmidt, E., Putora, P. M., & Fijten, R. (2025) · Philosophy & Technology · doi:10.1007/s13347-024-00834-9 · Cited by 14
Metadata
- Authors
- Afnan, M. A. M., Liu, Y., Conitzer, V., Rudin, C., Mishra, A., Savulescu, J., & Afnan, M.
- Venue
- Human Reproduction Open
- Publisher
- Oxford University Press (OUP)
- Published
- 2021-09-01
- Volume
- 2021
- Issue
- 4
- Pages
- hoab040-hoab040
- ISSN
- 2399-3529 (ISSN-L)
- OpenAlex
- W3209259671
- Cited by
- 112
- References
- 12
- FWCI
- 3.68
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.