Skip to main content

Table 1 Demographic of patients included in the study

From: Comparing a pre-defined versus deep learning approach for extracting brain atrophy patterns to predict cognitive decline due to Alzheimer’s disease in patients with mild cognitive symptoms

 

BioFINDER-1 cohort

ADNI cohort

Non-AD

Early AD

Non-AD

Early AD

No. of subjects

332

108

223 (67.2%)

109 (32.8%)

85 (78.7%)

23 (21.3%)

MCI

92 (41.2%)

78 (71.6%)

85 (100%)

23 (100%)

Female

106 (47.5%)

51 (46.8%)

27 (31.8%)

10 (43.5%)

Age (years)

70.2 ± 5.5

72.1 ± 4.7

70.9 ± 7.0

74.0 ± 4.9

Education (years)

12.0 ± 3.7

11.8 ± 3.5

16.1 ± 2.6

16.1 ± 2.9

Baseline MMSE score

28.2 ± 1.7

27.1 ± 1.7

28.3 ± 1.5

25.9 ± 2.2

APOE ε4

0.4 ± 0.6

1.0 ± 0.7

0.5 ± 0.7

0.9 ± 0.7

ADAS-cog delayed recall (no of errors)

4.2 ± 2.4

6.7 ± 2.1

4.1 ± 2.2

6.6 ± 2.4

Hippocampal volume (mm3)

3339 ± 480

2927 ± 415

3592 ± 447

3130 ± 483

Intracranial volume (cm3)

1136 ± 153

1114 ± 107

1502 ± 134

1440 ± 155

Aβ-positive

25%

100%

29%

82%

  1. Values are n (%) or mean ± standard deviation. Aβ-status defined by CSF Aβ42/Aβ40 in BioFINDER-1 (available in N=238) and by 18F-florbetapir in ADNI (available in N=85). The BioFINDER cohort was used for both training, validation and testing, in a double cross-validation setting. The ADNI cohort was used for external evaluation