Division of Reproductive Endocrinology and Infertility, Obstetrics and Gynecology, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street, Suite 10A, Boston, MA, 02114, USA.
Division of Engineering in Medicine, Brigham and Women's Hospital, Harvard Medical School, 65 Landsdowne Street, Cambridge, MA, 02139, USA.
J Assist Reprod Genet. 2022 Oct;39(10):2343-2348. doi: 10.1007/s10815-022-02585-y. Epub 2022 Aug 13.
To determine whether convolutional neural networks (CNN) can be used to accurately ascertain the patient identity (ID) of cleavage and blastocyst stage embryos based on image data alone.
A CNN model was trained and validated over three replicates on a retrospective cohort of 4889 time-lapse embryo images. The algorithm processed embryo images for each patient and produced a unique identification key that was associated with the patient ID at a timepoint on day 3 (~ 65 hours post-insemination (hpi)) and day 5 (~ 105 hpi) forming our data library. When the algorithm evaluated embryos at a later timepoint on day 3 (~ 70 hpi) and day 5 (~ 110 hpi), it generates another key that was matched with the patient's unique key available in the library. This approach was tested using 400 patient embryo cohorts on day 3 and day 5 and number of correct embryo identifications with the CNN algorithm was measured.
CNN technology matched the patient identification within random pools of 8 patient embryo cohorts on day 3 with 100% accuracy (n = 400 patients; 3 replicates). For day 5 embryo cohorts, the accuracy within random pools of 8 patients was 100% (n = 400 patients; 3 replicates).
This study describes an artificial intelligence-based approach for embryo identification. This technology offers a robust witnessing step based on unique morphological features of each embryo. This technology can be integrated with existing imaging systems and laboratory protocols to improve specimen tracking.
确定卷积神经网络（CNN）是否可以仅基于图像数据准确确定卵裂期和囊胚期胚胎的患者身份（ID）。
该 CNN 模型在一个回顾性队列的 4889 个延时胚胎图像上经过三个重复进行了训练和验证。该算法对每个患者的胚胎图像进行处理，并生成一个唯一的识别密钥，该密钥与第 3 天（65 小时受精后（hpi））和第 5 天（105 hpi）的患者 ID 相关联，形成我们的数据库。当该算法在第 3 天（70 hpi）和第 5 天（110 hpi）的稍后时间点评估胚胎时，它会生成另一个与库中患者唯一密钥匹配的密钥。使用第 3 天和第 5 天的 400 个患者胚胎队列测试了这种方法，并测量了 CNN 算法正确识别胚胎的数量。
CNN 技术在第 3 天的 8 个患者胚胎队列的随机池中以 100%的准确率匹配患者识别（n=400 个患者；3 个重复）。对于第 5 天的胚胎队列，在 8 个患者的随机池中准确率为 100%（n=400 个患者；3 个重复）。
本研究描述了一种基于人工智能的胚胎识别方法。该技术提供了基于每个胚胎独特形态特征的强大见证步骤。该技术可以与现有的成像系统和实验室协议集成，以提高标本跟踪。
