Fig. 4: Performances of BELA and traditional machine learning models on the Florida dataset.
From: Automatic ploidy prediction and quality assessment of human blastocysts using time-lapse imaging
Average AUC with standard errors is shown for EUP versus ANU and EUP versus CxA prediction tasks on the Florida dataset for BELA and logistic regression models trained only on embryologist-derived BS and/or maternal age across four replicates (four-fold cross-validation) (n = 4). Statistical significance was performed using a two-sided t-test, and relevant lack of statistical significance are shown via ‘ns’. Source data are provided as a Source Data file.
