Table 1 clinical characteristics of patients in three independent cohorts

From: Deep adaptive learning predicts and diagnoses CSVD-related cognitive decline using radiomics from T2-FLAIR: a multi-centre study

 

Training cohort

Hospital based external validation cohort (Zheer)

Community based external validation cohort (Xianlin)

F/X2

P

Participants

572

96

115

  

Without CI

346

48

53

  

With CI

226

48

62

  

Age (years, Mean ± SD)c

64.37±8.72

63.10±8.62

67.93±6.04a,b

10.738

<0.001*

Male, n (%)d

322 (56.3%)

47 (49.0%)

45 (39.1%)

11.993

0.002*

Education (years, Mean ± SD)c

10.71±4.64

4.94±4.12a

10.77±3.86b

70.235

<0.001*

MMSE (Mean ± SD)c

27.40±3.24

24.65±4.77a

27.49±2.43b

28.505

<0.001*

MoCA (Mean ± SD)c

23.31±4.69

19.70±5.97a

23.29±4.30b

23.602

<0.001*

WMH Fazekas 1, n (%)

268 (46.9%)

30 (31.3%)

69 (60.0%)

  

WMH Fazekas 2–3, n (%)d

304 (53.1%)

66 (68.8%)

46 (40.0%)

17.367

<0.001*

LI, n (%)d

251 (43.9%)

25 (26.0%)

20 (17.4%)

35.013

<0.001*

CMBs, n (%)d

172 (30.2%)

33 (34.4%)

28 (24.3%)

2.631

0.268

  1. Continuous variables are presented as mean ± SD, categorical variables are presented as counts(n) and percentages (%).
  2. CI cognitive impairment, CMBs cerebral microbleeds, LI lacunar infarction, SD standard deviation, MMSE Mini-Mental State Examination, MoCA Montreal Cognitive Assessment, WMH white matter hyperintensity.
  3. *Significance at level P < 0.05.
  4. ap < 0.05, significant difference from the training cohort.
  5. bp < 0.05, significant difference from the hospital based external validation cohort.
  6. cAnalyzed by analysis of variance.
  7. dAnalyzed by chi-square test.
Back to article page