Table 2 Comparison of histopathologic morphological characteristics between p16-negative and p16-positive OPSCC.

From: Extracting interpretable features for pathologists using weakly supervised learning to predict p16 expression in oropharyngeal cancer

  

p16-negative n = 95

p16-positive n = 45

P value

Mean number of nuclei, n (range)

 

68.85 (5.00–216.00)

185.13 (12.00–459.00)

 < 0.001

Mean circularity of the Nuclei (range)

 

0.76 (0.62–0.88)

0.76 (0.40–0.88)

0.973

Mean Maximum nucleus caliper, pixels (range)

 

20.64 (12.83–31.79)

15.46 (10.48–21.50)

 < 0.001

Mean minimum nucleus caliper, pixels (range)

 

12.82 (7.86–21.52)

9.69 (6.13–13.40)

 < 0.001

Perinuclear halo, n (%)

Negative

21 (22.1)

14 (31.1)

0.297

 

Positive

74 (77.9)

31 (68.9)

 

Distinct nucleoli, n (%)

Negative

56 (58.9)

30 (66.7)

0.458

 

Positive

39 (41.1)

15 (33.3)

 

Intercellular bridges, n (%)

Negative

58 (61.1)

45 (100.0)

 < 0.001

 

Positive

37 (38.9)

0 (0.0)

 

Keratin pearls, n (%)

Negative

87 (91.6)

43 (95.6)

0.501

 

Positive

8 (8.4)

2 (4.4)

 

Necrosis, n (%)

Negative

95 (100.0)

45 (100.0)

1.000

  1. P values for continuous variables are based on the t test. P values for categorical variables are based on Fisher’s exact test of association.
Back to article page