Skip to main content
. 2026 Sep 21;17:1916182. doi: 10.3389/fendo.2026.1916182

Figure 4.

Panel A, B, and C display SHAP summary beeswarm plots illustrating the influence of different clinical features on model predictions, each ordered by feature importance and color-coded by feature value. Panel D shows a scatter plot of SHAP values versus EMT (trigger day), color-coded by female age. Panel E presents SHAP values for progesterone (P, trigger day) similarly color-coded. Panel F plots SHAP values against oocytes retrieved, and panel G plots SHAP values against high-quality blastocysts, both with coloring by female age.

SHAP analysis revealing feature importance and dependence in the ensemble model across three stages. SHAP summary plot (beeswarm) showing the relative importance of variables in predicting live birth (A) baseline stage, (B) stimulation stage, (C) embryo development stage). (D) SHAP dependence plot showing the effect of trigger-day endometrial thickness (EMT) on model-predicted live birth probability, stratified by female age. (E) SHAP dependence plot depicting the association between trigger-day progesterone (P) and live birth prediction, stratified by female age. (F) SHAP dependence plot of the number of oocytes retrieved, demonstrating its impact on live birth prediction across different female age groups. (G) SHAP dependence plot of high-quality (HQ) blastocyst count, visualizing its contribution to live birth prediction by female age. Female factor 4 indicated advanced age/DOR.