Table 4 Results of persistent homology application combined with supervised algorithms with different k-fold cross-validation value for discrimination of the Gleason score 2, 3, 4 and 5 of cancerous prostate gland images
From: Machine learning techniques on homological persistence features for prostate cancer diagnosis
k-fold/method | LDA | NBC | SVM | DTC | RF | |||||
|---|---|---|---|---|---|---|---|---|---|---|
Accuracy (%) | AUC | Accuracy (%) | AUC | Accuracy (%) | AUC | Accuracy (%) | AUC | Accuracy (%) | AUC | |
K = 2 fold (\({\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 2}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$2$}}\) training \({\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 2}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$2$}}\) testing) | 48.5 | 0.5671 | 50.5 | 0.5671 | 56.7 | 0.6162 | 95.8 | 0.9889 | 66.5 | 0.8242 |
K = 3 fold (\({\raise0.7ex\hbox{$2$} \!\mathord{\left/ {\vphantom {2 3}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$3$}}\) training \({\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 3}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$3$}}\) testing) | 50.2 | 0.5722 | 52.3 | 0.5822 | 55.9 | 0.6233 | 95.4 | 0.9896 | 67.8 | 0.8221 |
K = 4 fold (\({\raise0.7ex\hbox{$3$} \!\mathord{\left/ {\vphantom {3 4}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$4$}}\) training \({\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 4}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$4$}}\) testing) | 51.2 | 0.5971 | 51.7 | 0.5945 | 58.2 | 0.6541 | 96.1 | 0.9879 | 66.7 | 0.8202 |
K = 5 fold (\({\raise0.7ex\hbox{$4$} \!\mathord{\left/ {\vphantom {4 5}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$5$}}\) training \({\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 5}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$5$}}\) testing) | 53.6 | 0.6149 | 53.6 | 0.6156 | 59.6 | 0.6728 | 96.9 | 0.9929 | 68.0 | 0.8237 |