Skip to main content

Table 1 The table reports the m/z values selected in [3] for the classification of the same dataset with four different models

From: A scale space approach for unsupervised feature selection in mass spectra classification for ovarian cancer detection

m/z features in[3] stable feature in our model
818.480 Yes
1001.654 Yes
1144.796 Yes
1255.593 Yes
1276.861 Yes
2374.244 No
4260.403 Yes
4292.900 Yes
4377.853 Yes
6004.416 Yes
6548.771 Yes
7046.018 Yes
7060.121 Yes
7096.922 No
7202.716 Yes
8540.536 Yes
8605.678 Yes
8664.385 No
8706.065 Yes
8709.548 Yes
9367.113 No
9870.937 No
  1. The values in bold represent the features recurring between the various models reported by the authors. As it can be seen from the table, the proposed method selects as stable features most of the values and all the recurring m/z. Since our data are subject to a re-sampling step before feature selection, the matching is measured up to a small approximation error.