Skip to main content
. 2022 Feb 7;21(1):e12443. doi: 10.1002/rmb2.12443

FIGURE 1.

FIGURE 1

Layer algorithm from images and clinical data for a neural network for prediction of clinical pregnancy or live birth. In the image analysis, blastocyst images those resulted in positive serum human chorionic gonadotropin and fetal heartbeats were regarded as “viable’” and the other images were regarded as “non‐viable.” (A) Prediction algorithm from Gardner's grading scales evaluated by an embryologist (control model). (B) Prediction algorithm from blastocyst images (image‐only model; FITTE). (C) Prediction algorithm from the ensemble of blastocyst images and non‐image clinical data (ensemble model). Res Net; residual network. AMH; Anti‐mullerian hormone. ART; assisted reproductive technology