Skip to main content

Table 2 List of the possible data configurations (topological feature sets, datasets) used to train classification models

From: A deep neural network approach to predicting clinical outcomes of neuroblastoma patients

DatasetsTopological featuresTotal size
Fischer-MCentralities12
 Modularities{30, 39}a
 Both{42, 51} a
Fischer-RCentralities12
 Modularities{36, 47}a
 Both{48, 59} a
FischerbCentralities24
 Modularities{75, 77}a
 Both{99, 101}a
  1. aThe number of modules for each graph, corresponding to one clinical outcomes of interest, is different
  2. bThis is the combined dataset in which the topological features of both ‘Fischer-M’ and ‘Fischer-R’ are concatenated