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 2: Sub-task Information Generator.
Input:	a sequence of modified class labels $D_{r} = {{\overset{´}{c}}_{j}^{r} \| j = 1, 2, \dots, n_{s}}$ , window size $w s$ , sliding samples $s s$
Output:	a sequence of sub-task labels $D_{g} = {{\overset{´}{c}}_{i}^{g} \| i = 1, 2, \dots, n_{g}}$
1:	*Begin*
2:	initialize i $\leftarrow 1$
3:	for $j = 1$ to $n_{s} - 1$ do //for the first $n_{s} - 1$ modified class labels
4:	while $i \leq (j \times s s)$
5:	${\overset{´}{c}}_{i}^{g} \leftarrow {\overset{´}{c}}_{j}^{r}$
6:	$D_{g} \leftarrow {\overset{´}{c}}_{i}^{g}$
7:	$i \leftarrow i + 1$
8:	endwhile
9:	endfor
10:	for $i$ to $i + w s$ do //for the last $n_{s}$ modified class labels
11:	${\overset{´}{c}}_{i}^{g} \leftarrow {\overset{´}{c}}_{n s}^{r}$
12:	$D_{g} \leftarrow {\overset{´}{c}}_{i}^{g}$
13:	endfor
14:	*End*