Misaghi Hooman, Cree Lynsey, Knowlton Nicholas
Department of Obstetrics, Gynaecology and Reproductive Sciences, University of Auckland, Auckland, New Zealand.
School of Mathematical and Computational Sciences, Massey University, Albany, New Zealand.
J Assist Reprod Genet. 2025 Aug 20. doi: 10.1007/s10815-025-03585-4.
The ability to detect, monitor, and precisely time the morphokinetic stages of human pre-implantation embryo development plays a critical role in assessing their viability and potential for successful implantation. Therefore, there is a need for accurate and accessible tools to analyse embryos. This work describes a highly accurate, machine learning model designed to predict 17 morphokinetic stages of pre-implantation human development, an improvement on existing models. This model provides a robust tool for researchers and clinicians, enabling the automation of morphokinetic stage prediction, standardising the process, and reducing subjectivity between clinics.
A computer vision model was built on a publicly available dataset for embryo Morphokinetic stage detection. The dataset contained 273,438 labelled images based on Embryoscope/ + © embryo images. The dataset was split 70/10/20 into training/validation/test sets. Two different deep learning architectures were trained and tested, one using EfficientNet-V2-Large and the other using EfficientNet-V2-Large with the addition of fertilisation time as input. A new postprocessing algorithm was developed to reduce noise in the predictions of the deep learning model and detect the exact time of each morphokinetic stage change.
The proposed model reached an overall test F1-score of 0.881 and accuracy of 87% across 17 morphokinetic stages on an independent test set.
The proposed model shows a 17% accuracy improvement, compared to the best models on the same dataset. Therefore, our model can accurately detect morphokinetic stages in static embryo images as well as detecting the exact timings of stage changes in a complete time-lapse video.
检测、监测并精确计时人类植入前胚胎发育的形态动力学阶段的能力，在评估其生存能力和成功植入的潜力方面起着关键作用。因此，需要准确且易于使用的工具来分析胚胎。这项工作描述了一种高度准确的机器学习模型，旨在预测人类植入前发育的17个形态动力学阶段，是对现有模型的改进。该模型为研究人员和临床医生提供了一个强大的工具，能够实现形态动力学阶段预测的自动化，使过程标准化，并减少不同诊所之间的主观性。
基于公开可用的胚胎形态动力学阶段检测数据集构建了一个计算机视觉模型。该数据集包含基于Embryoscope/ + ©胚胎图像的273,438张标注图像。数据集按70/10/20划分为训练/验证/测试集。对两种不同的深度学习架构进行了训练和测试，一种使用EfficientNet-V2-Large，另一种在EfficientNet-V2-Large的基础上增加受精时间作为输入。开发了一种新的后处理算法，以减少深度学习模型预测中的噪声，并检测每个形态动力学阶段变化的确切时间。
在独立测试集上，所提出的模型在17个形态动力学阶段的总体测试F1分数达到0.881，准确率达到87%。
与同一数据集上的最佳模型相比，所提出的模型准确率提高了17%。因此，我们的模型能够准确检测静态胚胎图像中的形态动力学阶段，以及在完整的延时视频中检测阶段变化的确切时间。
