Fig 1.
Flowchart of Methods, Testing and Results.
Fig 2.
Training progress and SOM plots.
Note: The “Training progress” graph indicates as the SOM training iterations distance from each node's weights to the samples represented by that node reduces and plateaus to indicate no more iterations were required. The “Counts plots” indicates reasonable samples were mapped to each node on the map. The “Neighbour distance plot” or U-Matrix indicates the distance between each node and its neighbours.
Fig 3.
Hierarchical Cluster Options for SOM.
Note: Clusters 3 to 12 solutions mapped onto the SOM grid. Colours indicate different clusters. The final 10 cluster solution selected for further analysis has been highlighted with a red border.
Table 1.
Frequency Distribution of Initial Depression Ordered SOM Cluster Solution.
Fig 4.
Mean depression scores and percent depression across final depression clusters.
Note: “Mean Depression Score” is the average total PHQ-9 score which ranged from 0 to 27. “Percent Depressed” based on a total PHQ-9 ≥ 10.
Table 2.
Demographic Profile Across SOM Clusters.
Fig 5.
Predicted probability of depression across age and family income poverty ratio for each cluster.
Table 3.
Binary Logistic Regression Model Odds Ratios with 95% Confidence Intervals.
Fig 6.
Importance of medical categories that make up the key significant clusters.
Note: Based on total boosted relative importance percentage across all clusters. Summed percentage from boosted regression across all five key significant clusters, thus total >100%.
Fig 7.
Total percentage importance of medical conditions for each key significant cluster.
Note: Clusters presented in order of percent depressed. Note: Percentage sum does not take account of direction of relationship.