Usefulness of embryo evaluation via artificial intelligence–based image analysis

Objective: Life Whisperer Viability (LW) is a software that evaluates embryos from a single image of blastocyst using artificial intelligence and is expected to improve the objectivity of embryo evaluation. In the present study, we retrospectively analyzed the clinical utility of LW. Methods: This study included 198 cycles in 135 patients who underwent frozen single embryo transfer. LW scores of transferred embryos were obtained from images taken approximately 116 h after fertilization. Based on LW scores, patients were divided into four groups (poor: 0–2.4; fair: 2.5–7.4; good: 7.5–8.9; excellent: 9.0–10), and rates of positive fetal heartbeat were compared among the groups. In addition, logistic regression analysis was performed to evaluate the association between LW score and the presence or absence of fetal heartbeat. Results: The rate of positive fetal heartbeat was significantly higher in the excellent group (64.3%, 36/56) than in the poor group (27.6%, 8/29) (P<0.05). Logistic regression analysis showed that the odds ratio for LW score was 1.150 (95% confidence interval: 1.040–1.280; P<0.05), indicating a significant association between LW score and presence or absence of fetal heartbeat. Conclusion: LW enables selection of embryos with a high potential for clinical pregnancy, demonstrating its clinical utility.