Table 2 Participant characteristics of the FTA and FTC groups.

From: Ultrasound-based classification of follicular thyroid Cancer using deep convolutional neural networks with transfer learning

Variable

Category/Group

FTA

FTC

P-value

Age

 

48.89 ± 12.96

44.77 ± 15.89

0.01

Sex

Female

150 (81.97%)

111 (77.08%)

0.34

 

Male

33 (18.03%)

33 (22.92%)

 

Diameter

< 2 cm

22 (12.02%)

17 (11.81%)

0.55

 

> 2 cm

34 (18.58%)

26 (18.06%)

 
 

> 3 cm

48 (26.23%)

29 (20.14%)

 
 

> 4 cm

79 (43.17%)

72 (50.00%)

 

Composition

Cystic

23 (12.57%)

2 (1.39%)

0.00

 

Solid

159 (86.89%)

141 (97.92%)

 
 

Predominantly solid

1 (0.55%)

1 (0.69%)

 

Margin

Regular

171 (93.44%)

61 (42.36%)

0.00

 

Irregular

12 (6.56%)

83 (57.64%)

 

Halo

Absent

174 (95.08%)

101 (70.14%)

0.00

 

Irregular halo

9 (4.92%)

43 (29.86%)

 

Echo

Anechoic

3 (1.64%)

0 (0.00%)

0.00

 

Isoechoic

169 (92.35%)

119 (82.64%)

 
 

Very hypoechoic

11 (6.01%)

25 (17.36%)

 

Texture

Even

105 (57.38%)

20 (13.89%)

0.00

 

Uneven

78 (42.62%)

124 (86.11%)

 

Internal septations

Absent

183 (100.00%)

134 (93.06%)

0.00

 

Present

0 (0.00%)

10 (6.94%)

 

Nodule within nodule

Absent

183 (100%)

141 (97.92%)

0.17

 

Present

0 (0.00%)

3 (2.08%)

 

Calcifications

Absent

172 (93.99%)

113 (78.47%)

0.00

 

Present

11 (6.01%)

31 (21.52%)

 

ACR-TIRADS

ACR-TR1

0 (0.00%)

0 (0.00%)

 
 

ACR-TR2

56 (30.60%)

7 (4.86%)

0.00

 

ACR-TR3

79 (43.17%)

28 (19.44%)

 
 

ACR-TR4

48 (26.23%)

109 (75.7%)

 
 

ACR-TR5

0 (0.00%)

O (0.00%)

 

C-TIRADS

C-TR2

0 (0.00%)

0 (0.00%)

 
 

C-TR3

182 (99.45%)

132 (91.67%)

0.00

 

C-TR4A

1 (0.55%)

12 (8.33%)

 
 

C-TR4B to 5

0 (0.00%)

0 (0.00%)

 
  1. TI-RADS, Thyroid Imaging Reporting and Data System; ACR, American College of.
  2. TI-RADS, Thyroid Imaging Reporting and Data System; ACR, American College of Radiology; C–TI-RADS, Chinese TI-RADS; SD, standard deviation; FTA, follicular thyroid adenoma; FTC, follicular thyroid cancer.