Table 3 The results obtained for the various classifier systems in discriminating between the cancer and non-cancer (respiratory) samples.
From: Potential for Raman spectroscopy to provide cancer screening using a peripheral blood sample
Classifier Method | Accuracy (%) | Sensitivity (%) | Specificity (%) |
|---|---|---|---|
LDA on the best 25 features from the t-test | 57.9 | 55.6 | 60.0 |
PCA-LDA (LDA on 25 principal components) | 65.0 | 64.7 | 65.3 |
100 best > PCA 25 > LDA | 65.8 | 64.9 | 66.7 |
Genetic Evolutionary Algorithm | 83 | 75 | 75 |