Skip to main content

Table 2 Classification performance of the individual classifier for region (a), (b), (d), and (e) and the integrated using the selected feature set

From: The development and evaluation of a computerized diagnosis scheme for pneumoconiosis on digital chest radiographs

Lung region Training dataset Test dataset
  AUC Spe Sen Acc AUC Spe Sen Acc
(a) 0.986 0.832 0.894 0.852 0.879 0.753 0.844 0.786
(b) 0.997 0.996 0.737 0.910 0.939 0.934 0.738 0.866
(d) 0.953 0.811 0.724 0.782 0.835 0.762 0.623 0.711
(e) 0.994 0.971 0.800 0.913 0.889 0.880 0.697 0.817
Integrated 0.997 0.995 0.913 0.969 0.961 0.933 0.849 0.905
  1. AUC = area under ROC curve; Spe = specificity; Sen = sensitivity; Acc = accuracy.
  2. Lung regions (a) and (b) correspond to upper and middle lung field of the right lung, and region (d) and (e) correspond to upper and middle lung field of the left lung, respectively.