Figure 2 | Scientific Reports

Figure 2

From: Multimodal machine learning for modeling infant head circumference, mothers’ milk composition, and their shared environment

Figure 2

Feature interdependency network illustrating correlations among predictor variables and their associations with head circumference at 3 months. Each node corresponds to a feature from various modalities (encoded by color). The closer the features the more similar they can be considered with regard to their correlation structure. Node sizes represent the strength of association between head circumference at 3 months and the corresponding feature based on the p-value of a significance test (Kendall's Tau for continuous variables, Wilcoxon rank sum test for binary variables). If this association passes a significance threshold of p < 0.05, the corresponding feature is represented by a filled node (no multiple test correction for visualization purposes). Prominent clusters of human milk components (yellow), infant feeding (green), and maternal diet (grey) emerge. For a detailed description of features, see Supplemental Table S1.

Back to article page