The Application Value Of Raman Spectroscopy In Assessmeng Of Embryo Quality And Developmental Potential

Part ? The Application Value of Raman Spectroscopy on of Embryo Quality AssessmentObjective The aim of this study was to investigate the application value of Raman spectroscopy in embryo quality assessment.Methods A total of 122 culture media samples of Day 3 embryos from 16 patients undergoing ICSI were individually collected from the Reproductive Medicine Center of the 901 Hospital of the PLA Joint Logistic Support Force(901th PLA)in this study.Conventional morphological scoring method was used to evaluate the Day 3 embryos,and the embryos were divided into the good-quality embryo group(66 samples)and the poor-quality embryo group(56 samples)according to the results of morphological evaluation,then the culture media of the two groups were analyzed through the method of Raman spectroscopy combined with the traditional chemometrics method(Principal Component Analysis(PCA)).The acquired Raman data were modeled and predicted by combining the advanced machine learning algorithm-Convolutional Neural Network(CNN).Results There was significantly difference in Raman spectroscopy of PC1 between goodquality embryo group and poor-quality embryo group(P < 0.05),while Raman spectroscopy of PC2 and PC3 showed no significant difference between the two groups.The CNN algorithm was used to build the model and forecast,it was showed that the specificity,sensitivity and accuracy of Raman spectroscopy to predict good-quality embryo were 71.21%,73.21% and 72.13% respectively.Conclusions Raman spectroscopy of good-quality embryos and poor-quality embryos is significantly different.Combined with PCA analysis and CNN algorithm,Raman spectroscopy could distinguish good-quality embryos and poor-quality embryos,and has certain application value in embryo quality assessment.Part ? The Application Value of Raman Spectroscopy in Predicting Embryo Developmental PotentialObjective The aim of this study was to investigate the application value of Raman spectroscopy in predicting embryo developmental potential.Methods A total of 111 embryo culture media samples of D3 good-quality embryos from 12 patients undergoing ICSI were individually collected from the Reproductive Medicine Center of the 901 Hospital of the PLA Joint Logistic Support Force(901th PLA)in this study.The corresponding D3 embryos were continuously cultured to D5/D6 for blastocyst,and were divided into blastocyst formation group(n=57)and the no-blastocyst group(n=54),then the embryo culture media of D3 embryos were collected.The Raman spectroscopy combined with the traditional chemometrics method [Principal Component Analysis(PCA)],was used to analyze embryo culture media of each D3 embryos in two groups,the acquired Raman data were modeled and predicted by combining the advanced machine learning algorithm-Convolutional Neural Network(CNN).Results There were significantly difference in Raman spectroscopy of PC1 and PC2 between blastocyst formation group and the no-blastocyst group(P < 0.05),while Raman spectroscopy in PC3 showed no significant difference between the two groups.The CNN algorithm was used to build the model and forecast,the specificity,sensitivity and accuracy of Raman spectroscopy to predict blastocyst formation potential of D3 goodquality embryos were 75.44%,74.07% and 74.77% respectively.Conclusions Raman spectroscopy of D3 good-quality embryo culture medium in blastocyst group and the non-blastocyst group were significantly different.Raman spectroscopy combined with PCA analysis and CNN algorithm could predict the development potential of D3 good-quality embryo.