Automatic Functional Shoulder Task Identification and Sub-Task Segmentation Using Wearable Inertial Measurement Units for Frozen Shoulder Assessment

. 2020 Dec 26;21(1):106. doi: 10.3390/s21010106

Algorithm 1: Sliding Window.
Input:	the pre-processed sensing data ${\tilde{S}}_{s e q} = {{\tilde{s}}_{i} \| i = 1, 2, \dots, n_{s e q}}$ , window size $w s$ , the starting point of the segment $s p$ , ending point of the segment $e p$ , sliding samples $s s$
Output:	a set of segments $W = {w_{j} \| j = 1, 2, \dots, n_{s l}}$
1:	*Begin*
2:	initialize $s p \leftarrow 1$ , e $p \leftarrow w s$ and $j \leftarrow 1$
3:	while e $p \leq n_{s e q}$ do
4:	$w_{j} \leftarrow {{\tilde{s}}_{s p}^{j}, {\tilde{s}}_{s p + 1}^{j}, \dots, {\tilde{s}}_{e p - 1}^{j}, {\tilde{s}}_{e p}^{j}}$
5:	$W \leftarrow w_{j}$
6:	$j \leftarrow j + 1$
7:	$s p \leftarrow s p + s s$
8:	e $p \leftarrow s p + w s - 1$
9:	endwhile
10:	*End*