Skip to main content

Table 3 Performance of classification tree model for classification of neoplasia, non-neoplasia, and individual histologic types of breast neoplasia in confocal fluorescence images

From: Micro-anatomical quantitative optical imaging: toward automated assessment of breast tissues

  Sensitivity Specificity
Classification tree model 81 % (65/80) 93 % (167/179)
Cross-validated model 75 % (60/80) 93 % (167/179)
  Correctly classified  
All sites 90 % (232/259)  
Ductal carcinoma in situ (DCIS) 96 % (25/26)  
Invasive ductal cancer (IDC) 92 % (34/37)  
Invasive lobular carcinoma (ILC) 35 % (6/17)