Biomedical Engineering Reference
In-Depth Information
Fig. 6.16 Area under the
ROC curve. The maximum
AUCs for the observation
period T
i
of 100, 300, and
900 s are 0.77, 0.92, and
0.93, respectively
1
T
i
= 100
T
i
= 300
T
i
= 900
0.95
0.9
0.85
0.8
0.75
0.7
0.75
0.8
0.85
0.9
Regular threshold (
Ψ
th
)
Table 6.2
Classifier studies of irregular breathing detection
Studies
Performance
(%)
Measurement datasets
Methods
Regular/irregular classification
[
15
]
TPR: 92.6
Spirometry data of 250
ANN-based
Regular/irregular classification
[
16
]
TPR: 97.5
Spirometry data of 205
ANN-based
Regular/irregular classification
[
17
]
TP/(TP ? FP):
98
74 sleep disordered breathing
data
ANN-based
Proposed classification
TPR: 97.8
Breathing motion data of 448
EM/ANN-
based
TPR true positive rate, ANN artificial neural network
6.6 Summary
In this Chapter we have presented an irregular breathing classifier that is based on
the regular ratio
ðÞ
detected in multiple patients-datasets. Our new method has
two main contributions to classify irregular breathing patterns. The first contri-
bution is to propose a new approach to detect abnormal breathing patterns with
multiple patients' breathing data that better reflect tumor motion in a way needed
for radiotherapy than the spirometry. The second contribution is that the proposed
new method achieves the best irregular classification performance by adopting EM
based on the Gaussian Mixture model with the usable feature combination from
the given feature extraction metrics.
Search WWH ::
Custom Search