Moving Beyond the Black Box: A Proposed Integrated Clinical AI Scoring (ICAIS) Matrix for Embryo Selection
Abstract Background Embryo selection remains a critical determinant of IVF success. Despite advances in morphology, time-lapse imaging, and preimplantation genetic testing for aneuploidy (PGT-A), implantation failure still occurs in morphologically high-quality and euploid embryos, highlighting limitations in current assessment methods. The current study is a theoretical framework development study based on literature synthesis (2000–2026) from major databases. The proposed Integrated Comprehensive AI-based Implantation Scoring system (ICAIS) comprises three domains: morphology (0–9 points), morphokinetics (0–3 points), and an AI-derived adjustment factor (− 2 to + 3 points), yielding a total score of 0–15.The aim is to develop an integrated, transparent scoring system combining morphology, morphokinetics, and artificial intelligence (AI) to improve prediction of embryo implantation potential. Results The ICAIS model stratifies embryos into low (0–7), intermediate (8–11), and high (12–15) implantation potential categories. The integrated approach demonstrates improved theoretical discrimination between embryos with similar morphology but different implantation potential, and better alignment with known euploidy and developmental competence patterns. Conclusion ICAIS provides a structured and interpretable framework that integrates conventional and AI-based embryo assessment. It has the potential to enhance embryo selection and address unexplained implantation failure, although prospective clinical validation is required.
Authors
- Nehal Adel (ORCID: https://orcid.org/0000-0002-2853-8334)
- Mohamed Elmahdy (ORCID: https://orcid.org/0000-0003-1570-9015)
- Hassan Maghraby
- Rogaya Aborawi
- Abdulaziz Belhaj
- Heba Maghraby
Institutions
- Alexandria University (EG)
Publication Details
- Journal
- Research Square
- Published
- 2026-11-16
- DOI
- https://doi.org/10.21203/rs.3.rs-11354964/v1
- Primary Topic
- Reproductive Biology and Fertility
- Type
- preprint