Fig. 6From: A hierarchical procedure to select intrauterine and extrauterine factors for methodological validation of preterm birth risk estimationFuzzy C-means clustering for the risk factors selected by Akaike Information Criterion (AIC) and cluster centroids description. We present three clusters obtained by fuzzy C-means method on the 9 most informative variables selected by AIC at the explanatory phase. We split mothers based on the high/low preterm birth risk classification and we show them in ascending order with respect to cluster 1 membership score (i.e. 0 = no membership; 1 = total membership). We also show the informative preterm birth risk factors for Cluster 1, 2 and 3. The centroids matrix is also provided as a Supplementary materialBack to article page