Abstract：
Only 30% of embryos from in vitro fertilized oocytes successfully implant and develop to term, leading to repeated transfer cycles. To reduce time-to-pregnancy and stress for patients, there is a need for a diagnostic tool to better select embryos and oocytes based on their physiology. The current standard employs brightfield imaging, which provides limited physiological information. Here, we introduce METAPHOR: Metabolic Evaluation through Phasor-based Hyperspectral Imaging and Organelle Recognition. This non-invasive, label-free imaging method combines two-photon illumination and AI to deliver the metabolic profile of embryos and oocytes based on intrinsic autofluorescence signals. We used it to classify i) mouse blastocysts cultured under standard conditions or with depletion of selected metabolites (glucose, pyruvate, lactate); and ii) oocytes from young and old mouse females, or in vitro-aged oocytes. The imaging process was safe for blastocysts and oocytes. The METAPHOR classification of control vs. metabolites-depleted embryos reached an area under the ROC curve (AUC) of 93.7%, compared to 51% achieved for human grading using brightfield imaging. The binary classification of young vs. old/in vitro-aged oocytes and their blastulation prediction using METAPHOR reached an AUC of 96.2% and 82.2%, respectively. Finally, organelle recognition and segmentation based on the flavin adenine dinucleotide signal revealed that quantification of mitochondria size and distribution can be used as a biomarker to classify oocytes and embryos. The performance and safety of the method highlight the accuracy of noninvasive metabolic imaging as a complementary approach to evaluate oocytes and embryos based on their physiology.
摘要：
只有30%来自体外受精卵母细胞的胚胎成功植入并发育至足月，导致重复的传输周期。为了减少患者的怀孕时间和压力，需要一种诊断工具，以更好地选择胚胎和卵母细胞的基础上，他们的生理。当前的标准采用明场成像，提供有限的生理信息。这里,我们介绍了METAPOR：通过基于相量的高光谱成像和细胞器识别进行代谢评估。这种非侵入性的，无标记成像方法结合了双光子照明和AI，以提供基于固有自发荧光信号的胚胎和卵母细胞的代谢谱。我们用它来分类i）在标准条件下培养的小鼠胚泡或消耗选定的代谢物（葡萄糖，丙酮酸,乳酸)；和ii)来自年轻和老年小鼠雌性的卵母细胞，或体外老化的卵母细胞。成像过程对胚泡和卵母细胞是安全的。对照与对照的METAPHOR分类代谢物耗尽的胚胎达到了93.7%的ROC曲线下面积(AUC),相比之下，使用明场成像进行人体分级的比例为51%。青年与青年的二元分类使用METAPHOR进行的老年/体外衰老卵母细胞及其囊胚形成的AUC分别为96.2％和82.2％，分别。最后,基于黄素腺嘌呤二核苷酸信号的细胞器识别和分割表明，线粒体大小和分布的定量可以作为生物标志物对卵母细胞和胚胎进行分类。该方法的性能和安全性突出了非侵入性代谢成像的准确性，作为根据其生理学评估卵母细胞和胚胎的补充方法。
