Table 1 Clinicopathological characteristics of the study cohorts.

From: A robust artificial intelligence system for predicting EBV status in gastric cancer biopsy and resection specimens

Type

Internal

Development

Cohort

Internal

Validation

Cohort

External Validation Cohort

Dataset

KSH-DEV

 

KSH-GAT

KSH-GGT

KSH-GCB

KSH-STO

TCGA-STAD

Acquisition period

2011–2014

2011–2014

2016–2020

2016–2020

2022–2023

2013–2023

-

Patients included in this study

24

708

285

285

181

854

347

EBV-Positive

24

98

18

18

15

20

26

EBV-Negative

0

610

267

267

166

834

321

Age, years

61.13

62.98

64.57

64.57

67.20

64.37

64.05

Sex

       

Male

22

504

185

185

124

542

225

Female

2

204

100

100

57

312

112

Unknown

0

0

0

0

0

0

10

Histologic type

       

Papillary adenocarcinoma

0

1

0

0

0

8

7

Tubular adenocarcinoma

24

629

258

258

161

554

66

Well differentiated

0

210

130

130

21

85

0

Moderately differentiated

8

191

91

91

55

261

0

Poorly differentiated

16

228

37

37

85

208

0

Unknown

0

0

0

0

0

0

66

Mucinous adenocarcinoma

0

4

0

0

1

10

18

Poorly cohesive carcinoma

0

21

2

2

10

198

54

Signet ring cell carcinoma

0

52

15

15

7

3

14

Adenosquamous carcinoma

0

0

0

0

0

2

0

Medullary carcinoma

0

1

9

9

0

3

0

Undifferentiated carcinoma

0

0

0

0

0

3

0

Mixed

0

0

0

0

2

71

0

Unknown

0

0

1

1

0

2

178

Digital Slide Scanner

Leica Aperio AT2

Leica Aperio AT2

Leica Aperio AT2

Leica Aperio GT450

Leica Aperio GT450

Leica Aperio GT450

Various

Magnification

40x

40x

40x

40x

40x

40x

20x – 40x

Resolution

0.2521µm2

0.2521µm2

0.2521µm2

0.2635µm2

0.2635µm2

0.2635µm2

0.2325µm2 ~ 0.5005µm2

Sample Type

WSI

TMA

Biopsy

Biopsy

Biopsy

WSI

WSI