Irregular Breathing Classification from Multiple Patient Datasets - Prediction and Classification of Respiratory Motion - page 125

(a) (b)

1 1

0.9 0.9

0.8 0.8

0.7 0.7

0.6 0.6

0.5 0.5

0.4 0.4

0.3 0.3

Ψ

th = 0.8
Ψ th = 0.85
Ψ th = 0.9

Ψ

th = 0.8
Ψ th = 0.85
Ψ th = 0.9

0.2 0.2

0.1 0.1

0 0

0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1

FP Rate FP Rate

(c)

1

0.9

0.8

0.7

0.6

0.5

0.4

0.3

Ψ th = 0.8
Ψ th = 0.85
Ψ

0.2

0.1

th = 0.9

0

0 0.2 0.4 0.6 0.8 1

FP Rate

Fig. 6.15 ROC graph of irregular detection with different regular thresholds and observation
period, a observation period T i = 100 s, b observation period T i = 300 s, and c observation
period T i = 900 s

Table 6.2 shows the classification performances of the irregular detection. We
notice that irregular breathing patterns can be detected with the performance of
97.5 % TPR using the spirometry data and the ANN-based method [ 16 ]. Irregular
breathing detection with sleep disordered breathing data [ 17 ] shows a better
performance of 98 % TP/(TP ? FP). However, sleep-disorder data can not take
the place of the breathing motion for lung cancer treatment [ 17 ]. Our proposed
classification shows results of the classifier performance of 97.83 % TPR with 448
samples breathing motion data. That means the proposed classifier can achieve
acceptable results comparable to the classifier studies using the spirometry data.