Skip to main content

Table 3 Nuclear characteristics of different sebaceous lesions

From: Machine learning for classification of cutaneous sebaceous neoplasms: implementing decision tree model using cytological and architectural features

Variables

Sebaceous hyperplasia (n = 52)

Sebaceoma (n = 15)

Sebaceous adenoma (n = 13)

Extraocular sebaceous carcinoma (n = 20)

Periocular sebaceous carcinoma (n = 23)

Chromatin

     

 Fine

52 (100.0%)

8 (53.3%)

5 (38.5%)

0 (0.0%)

1 (4.3%)

 Coarse

0 (0.0%)

4 (26.7%)

3 (23.1%)

5 (25.0%)

3 (13.0%)

 Clumpy

0 (0.0%)

2 (13.3%)

4 (30.8%)

3 (15.0%)

1 (4.3%)

 Vesicular

0 (0.0%)

1 (6.7%)

1 (7.7%)

12 (60.0%)

12 (52.2%)

 Mixed

0 (0.0%)

0 (0.0%)

0 (0.0%)

0 (0.0%)

6 (26.1%)

Nucleoli

     

 Inconspicuous

48 (92.3%)

2 (13.3%)

2 (15.4%)

1 (5.0%)

3 (13.0%)

 One evident

4 (7.7%)

5 (33.3%)

4 (30.8%)

2 (10.0%)

6 (26.1%)

 One prominent

0 (0.0%)

5 (33.3%)

3 (23.1%)

7 (35.0%)

6 (26.1%)

 Multiple

0 (0.0%)

3 (20.0%)

4 (30.8%)

10 (50.0%)

8 (34.8%)

Nuclear contour

     

 Smooth

52 (100.0%)

13 (86.7%)

13 (100.0%)

4 (20.0%)

0 (0.0%)

 Irregular mild

0 (0.0%)

2 (13.3%)

0 (0.0%)

5 (25.0%)

4 (17.4%)

 Irregular moderate

0 (0.0%)

0 (0.0%)

0 (0.0%)

7 (35.0%)

12 (52.2%)

 Significant

0 (0.0%)

0 (0.0%)

0 (0.0%)

4 (20.0%)

7 (30.4%)

Nuclear size (relative to nuclear size adjacent keratinocytes)

  

 Less than 1 times

52 (100.0%)

10 (66.7%)

7 (53.8%)

1 (5.0%)

0 (0.0%)

 Between 1 and 2 times

0 (0.0%)

5 (33.7%)

6 (46.2%)

9 (45.0%)

4 (17.4%)

 More than 2 times

0 (0.0%)

0 (0.0%)

0 (0.0%)

10 (50.0%)

19 (82.6%)