Automatic identification of the signal for early hypertension treatment with laser and electrical stimulations(PDF)
《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]
- Issue:
- 2018年第10期
- Page:
- 1203-1209
- Research Field:
- 医学信号处理与医学仪器
- Publishing date:
Info
- Title:
- Automatic identification of the signal for early hypertension treatment with laser and electrical stimulations
- Author(s):
- FU Meihong; LI Chumin; CHEN Zhongben
- School of Biomedical Engineering, Xinhua College of Sun Yat-sen University, Guangzhou 510520, China
- Keywords:
- ; Key words: early hypertension; signal source of laser and electrical stimulations; audio signal; template library match; automatic identification
- PACS:
- R318.5；TN912.3
- DOI:
- DOI:10.3969/j.issn.1005-202X.2018.10.018
- Abstract:
- Abstract: Objective Based on the situation that simulation signals of different models are needed to solve the problems of individual differences and adaptability in early hypertension treatment with laser and electrical simulations, to propose a method of automatically identifying the simulation signals for providing a convenient and efficient means to choose effective treatment signals in clinic. Methods The digital audio signals which have been clinically proven to be effective were taken as the original signals, and the Mel frequency cepstral coefficients of this kind of signal were extracted to establish the HMM model library. After grouping the experimental signals and identifying the effective signals, the criterion of automatic identification was determined. Then the characteristic parameters of the test signals were extracted, and the matching probability of the test signals and model was calculated by Viterbi algorithm, and the test signals were automatically matched with the signals in the template library. Finally, according to the matching results, the test signals were classified into different categories in template library. Results and Conclusion HMM model could be used to classified test signals effectively. The signals with small relative value of the output probability logarithm between test signals and original signals could be sorted out accurately and effectively from a large number of test signals. After the clinical validation, the selected signals can be used as the signal source for early hypertension treatment with laser and electrical simulations.
Last Update: 2018-10-23