Biomedical Engineering Reference
In-Depth Information
60
50
Min = 18 minutes
Max = 166 minutes
Mean = 80 minutes
40
30
20
10
0
0
20
40
60
80
100
120
140
160
180
Breathing Recording Time (minutes)
Fig. 6.5 Frequency distribution of recording times for the breathing datasets. The breathing
recordings lasted anywhere from 18 min (min) to 166 min (max), with 80 min as the average time
6.5.2 Selection of the Estimated Feature Metrics (x)
The objective of this section is to find out the estimated feature metrics (x) from
the candidate feature combination vector (x) using discriminant criteria based on
clustered degree. Figure 6.6 a shows all the results of the objective function
(J) with respect to the feature metrics number. That means each column in Fig. 6.6
represents the number of feature extraction metrics in Table 6.1 . For example, let
us define the number of feature extraction metrics as three (z = 3). Here, the
feature combination vector can be x ¼ BRF, PCA, STD
ð
Þ with three out of 10
feature
metrics,
having
the
number
of
feature
combination
vector
ð
C 10 ; 3
ð
Þ¼ 120
Þ: The red spot shows the objective function J ðÞ for each feature
(a)
(b)
3.5 x 10 9
20
Individual
Average
Standard deviation
ADT+VEL+BRF+PCA+MLR+STD
3
15
2.5
2
10
ADT+BRF+PCA+MLR+STD
1.5
< Extended Range >
1
BRF+PCA+MLR,
BRF+PCA+STD,
BRF+MLR+STD,
PCA+MLR+STD
5
0.5
BRF+PCA+MLR+STD
0
0
2
3
4
5
6
7
8
9
3
3.5
4
4.5
5
5.5
6
Number of Feature Selection Metrics(z)
Number of Feature Selection Metrics(z)
Fig. 6.6 Objective functions for selection of feature metrics. This figure shows objective
functions with respect to the feature metrics number to select the estimated feature metrics (x),
a the whole range, and b extended range
Search WWH ::




Custom Search