Skip to main content

Table 2 Dataset characteristics

From: Automated freezing of gait assessment with marker-based motion capture and multi-stage spatial-temporal graph convolutional neural networks

ID

#Trials

#FOG trials

#FOG

%TF

Total duration

Avg duration (± SD)

S1

27

3

9

33.9

1.05

0.35 (± 0.28)

S2

22

11

13

12.3

4.01

0.36 (± 0.12)

S3

27

4

5

6.89

0.49

0.12 (± 0.02)

S4

21

9

18

36.7

2.45

0.27 (± 0.19)

S5

24

1

3

36.1

0.29

0.29

S6

7

5

7

14.4

0.85

0.17 (± 0.08)

S7

31

1

1

20.1

0.08

0.08

D2

68

68

134

28.4

24.4

0.36 (±0.18)

  1. Overview of the number of motion capture trials (#Trials), number of FOG trials (#FOG trials), number of FOG episodes (#FOG), percentage time spent frozen (%TF), total duration of the FOG trials (in minutes), and average duration of the FOG trials (± standard deviation (SD)) (in minutes). For dataset 1, the characteristics are given per subject. For dataset 2 (D2), which was only used to enrich the training dataset and not for model evaluation, a single summary is provided