Table 1 Confusion matrix of Resnet50 classification results
From: The value of deep neural networks in the pathological classification of thyroid tumors
Classification | ||||||
|---|---|---|---|---|---|---|
Pathological diagnosis | PTC | MTC | FTC | Adenoma | Adenomatous goiter | Normal thyroid |
PTC | 119(98.35%) | 2(1.65%) | ||||
MTC | 9(42.86%) | 7(33.33%) | 2(9.52%) | 3(14,29%) | ||
FTC | 2(13.33%) | 10(66.67%) | 1(6.67%) | 1(6.67%) | 1(6.67%) | |
Adenoma | 3(13.64%) | 1(4.55%) | 12(54.55%) | 3(13.64%) | 3(13.64%) | |
Adenomatous goiter | 1(8.33%) | 2(16.67%) | 6(50.00%) | 3(25.00%) | ||
Normal thyroid | 52(100.00%) |