Fig. 2: EM patient clustering based on the #Enzian classification system.

EM patient clustering according to #Enzian classification. a Schematic representation of the #Enzian classification coding system. Possible values and lesion types for different anatomical regions. Example annotations are provided to illustrate how the coding is applied from surgeon’s annotation, decomposed and translated into a severity score (SC) derived from the lesion types and their respective severity levels (0-3). The dimensionality reduction model (Kernel-PCA) is presented in 2D and 3D. EM patients (orange) and controls (blue) are plotted. Unsupervised clustering (K-means) was used for classifying patients. Clusters were named according to the mean Enzian Severity Index (mESI) of each group as control (Ctrl, 0.00), #I (0.02), #II (0.12), #III (0.31), #IV (0.48), and #V (0.89) respectively. b Principal Component Analysis (PCA) showing the distribution of each lesion classifier from #Enzian among the different clusters. c Parallel categories plot representing the #Enzian classification across different patient clusters. This plot visualizes the distribution of lesion types and severities. The categories show the relationship between the extent of disease and patient groups, providing a comprehensive overview of the classification. d PCA showing the #Enzian cluster, rASRM group, and ESI generated for each patient. Asterisks: patients showing dissociation between rASRM classification and ESI. A: retrovaginal space, B: sacrouterine ligaments, C: rectum, CTRL: control, EM: endometriosis, FA: adenomyosis, FI: intestinum, Fother: other, FU: ureter, O: ovarian, P: peritoneal, T: tubo-ovarian, Tpt: patency test.