P-285 An analysis of automated morphometric measurements finds that a combination of a large blastocyst size and a short tB-tSB time interval doubles the implantation rate

Abstract Study question Are automated blastocyst morphometric measurements combined with morphokinetic pattern associated implantation rate? Summary answer Automated demonstrated that a larger size and shorter time interval tB-tSB are higher potential. What is known already Optimization of embryo selection important for increasing Transfer high-quality based on conventional morphological parameters has been shown to improve IVF clinical outcome. Novel quality, including morphokinetics from time-lapse monitoring manual analysis parameters, have promising results regarding However, measurement time-consuming task subjected intra- inter-observer variations. The introduction would remove subjective further rates. design, size, duration A nested retrospective case control 608 day-5 transferred blastocysts was conducted included women who underwent treatment in three public units between 2014 2017. Participants/materials, setting, methods measured at the mean tEB-tPNf (85.82 h) by training pixel-wise segmentation model (MaskRCNN) videos. Morphometric following: (μm), inner cell mass (ICM) ICM ratio, shape. Annotation variables all intervals (hours) pronuclei fading (tPNf) expanded (tEB). Main role chance implanted embryos significantly compared non-implanted (152.10 ±19.22μm vs 144.25±18.52μm, respectively, p < 0.001), while times tSB-tPNf, tB-tPNf, tEB-tPNf, tB-tSB, tEB-tSB were (tSB-tPNf: 72.10±5.60h 73.30±5.80h, = 0.016; tB-tPNf: 80.08±5.96h 82.54±5.92h, 0.001; tEB-tPNf: 84.95±5.43h 86.58±4.93h, tB-tSB: 8.21±2.90h 9.49±3.62h, tEB-tSB: 13.50±3.00h 14.58±3.75h, 0.001). In multivariable logistic independent effect woman age potential, found be positively potential (OR 1.017, 95% CI 1.006-1.027, 0.002) negatively 0.918, 0.861-0.980, 0.010 OR 0.903, 0.874-0.932, 0.001, respectively). Blastocyst than had 2.028 greater did not meet these criteria 2.028, 1.420-2.894, adjusted age, among meeting aforementioned 1.7 (adjusted 1.714, 1.182-2.485, 0.005). AUC value prediction 0.69 (p 0.01). Limitations, reasons caution study's limitations include its nature absence some patient characteristics. Wider implications findings combination may double rate. inclusion algorithm reduce variations should incorporated into models prediction. Trial registration number 0006-20-HMO