The voice information is the most direct and effective reflection of the physiology and psychology health of neonates. Nowadays, the traditional nursing model in the nursing center is the enumeration, so that the nursing efficiency is low and it is lack of scientific decision guidance. Therefore, a method for neonates nursing decision based on voice identification is proposed in this paper. First of all, on the basement of the physiology structure and emotion characteristics of neonates, the voice information under various emotion conditions are analyzed, and the corresponding nursing activity are provided. Furthermore, the multi-fractal de-trended fluctuation analysis algorithm for feature extraction is presented to acquire voice message parameters with various emotion. Finally, the nursing decision model is established. By the identification analysis of different voice signals, the decision for targeted nursing activity is realized, which is verified by the real sampled voice data.
[1]
H. Stanley,et al.
Multifractal Detrended Fluctuation Analysis of Nonstationary Time Series
,
2002,
physics/0202070.
[2]
Damián Gulich,et al.
A criterion for the determination of optimal scaling ranges in DFA and MF-DFA
,
2014
.
[3]
J. Lin,et al.
Fault diagnosis of rolling bearings using multifractal detrended fluctuation analysis and Mahalanobis distance criterion
,
2012,
18th International Conference on Automation and Computing (ICAC).
[4]
Jian Sun,et al.
A Degradation Feature Extraction Method for Hydraulic Pumps Based Upon MUWDF and MF-DFA
,
2016,
Journal of Failure Analysis and Prevention.