Skip to main content

Table 1 Classification Accuracy

From: MetaSel: a metaphase selection tool using a Gaussian-based classification technique

Class 1 2 3 4
1 5,103 (99.42%) 0 3 (0.30%) 36(0.48%)
2 0 1,157 (90.67%) 92(9.26%) 34(2.54%)
3 0 99(7.76%) 889(89.44%) 37(2.50%)
4 30(0.58%) 20(1.57%) 10(1.01%) 307(93.25%)
Total 5133 1276 994 414
  1. Two hundred metaphase spreads were used to determine the accuracy of the proposed rules. From these 192 metaphase images, 7817-segmented objects were obtained and classified into four classes. The proposed rules can efficiently and accurately classify the individual chromosome with > 90% accuracy.