Table 3 Overview of source \({\mathcal {D}}_S\) and target domain \({\mathcal {D}}_T\) datasets. Datasets were constructed from the original sensor signal using 2.56 [s] epoch sliding windows with a 50% overlap.

From: Interpretable deep learning for the remote characterisation of ambulation in multiple sclerosis using smartphones

(n)

\({\mathcal {D}}_S\)

\({\mathcal {D}}_T\)

UCI HAR\(^{1}\)

WISDM\(^{1}\)

FL

Subjects

30

51

97\(^{a}\)

Tests

61

252

970\(^{b}\)

Samples

10013

54781

82450

  1. \(^{1}\) See supplementary material for more information on the UCI HAR and WISDM datasets.
  2. \(^{a}\) HC, n=24; PwMSmild, n=52; PwMSmod, n=21; see demographics Table 1 for more details.
  3. \(^{b}\) Randomly sampling \(m=10\) tests per subject.
Back to article page