Related Experiment Video
Updated: Aug 27, 2025
06:17
Quantifying Levels of Dopaminergic Neuron Morphological Alteration and Degeneration in Caenorhabditis elegans
Published on: November 20, 2021
4.3K
Association between a deep learning-based scoring system with morphokinetics and morphological alterations in human
Kenji Ezoe1, Kiyoe Shimazaki1, Tetsuya Miki1
1Kato Ladies Clinic, 7-20-3 Nishishinjuku, Shinjuku-ku Tokyo 160-0023, Japan.
Reproductive Biomedicine Online
|September 26, 2022
Summary
The deep learning scoring system, iDAScore, is strongly linked to embryo development events. This AI tool can help explain blastocyst selection for patients undergoing fertility treatments.
Area of Science:
- Embryology
- Artificial Intelligence in Medicine
- Reproductive Biology
Background:
- Assessing pre-implantation embryo development is crucial for successful in vitro fertilization (IVF).
- Traditional methods rely on morphokinetics and morphology, but advanced AI tools are emerging.
- The iDAScore is a deep learning-based system designed to evaluate embryo quality.
Purpose of the Study:
- To investigate the association between the iDAScore and key biological events during human pre-implantation embryonic development.
- To determine if iDAScore can predict or reflect specific morphokinetic and morphological changes.
- To explore the potential of iDAScore in improving embryo selection for IVF.
Main Methods:
- A retrospective observational study included 925 patients undergoing minimal stimulation IVF cycles.
- Data from expanded blastocysts obtained between October 2019 and December 2020 were analyzed.
- The association between iDAScore and various developmental events (fertilization, cleavage, compaction, blastulation) was statistically evaluated.
Main Results:
- Low iDAScore was significantly associated with prolonged cytoplasmic halo duration and delayed pronuclear breakdown.
- Embryos with abnormal cleavage patterns received lower iDAScores.
- Delayed compaction, blastulation, and abnormal blastomere behavior (fragmentation, extrusion) were linked to lower iDAScores, with morphology also being significantly associated.
- Multiple linear regression confirmed strong associations between iDAScore and late-stage morphokinetic/morphological events.
Conclusions:
- The iDAScore demonstrates a significant correlation with morphokinetic and morphological alterations in pre-implantation embryos.
- The system is particularly relevant for evaluating events during the late pre-implantation period.
- Deep learning models like iDAScore show promise for enhancing embryo selection in IVF, potentially offering clearer explanations to patients.
