Nonlinear predictability analysis of diabetic autonomic neuropathy
- VernacularTitle:糖尿病自主神经病变的非线性预测分析
- Author:
Qing WANG
;
Yongqin LI
;
Qinkai DENG
- Publication Type:Journal Article
- From:
Chinese Journal of Tissue Engineering Research
2005;9(7):190-192
- CountryChina
- Language:Chinese
-
Abstract:
BACKGROUND: Researches on the diagnosis and detection methods of diabetic autonomic neuropathy(DAN) is very weak at present.OBJECTIVE: To get the data about the degree of autonomic neuropathy by using the nonlinear heart rate variability(HRV) so as to provide clinical evidence for the early diagnosis of DAN.DESIGN: A single sample study.SETTINGS: Computer department in a university; Staff room of medical physics of department of biomedical engineering in a university.PARTICIPANTS: The experiment was completed in the Staff Room of Medical Physics, Department of Biomedical Engineering in the First Military Medical University of Chinese PLA from March to May 2002. Thirty-four diabetic patients, including 22 with DAN complications, were selected from the inpatients of Department of Endocrinology in the Affiliated Nanfang Hospital of the First Military Medical University of PLA, and other 40 normal subjects, who were from the faculty in the First Military Medical University of Chinese PLA and students in the Department of Physical Education, Chinese PLA Institute of Physical Education, were taken as the controls. The R wave to R wave intervals(RRI) time series is sampled and corrected to 512 points for analysis.INTERVENTIONS: Based on the virtual instrumental workbench-LabVIEW, 74 standard adjacent RRI signals were selected from 34 diabetic patients and 40 normal cases, and then analyzed with the method of nonlinear predictability.MAIN OUTCOME MEASURES: Collective and analytical results of the 74 standard adjacent RRI signals.RESULTS: There was no significant difference between the diabetic patients without DAN and the controls by the analysis with normalized mean square error(NMSE) NMSE ( P = 0. 075 ), but the significant difference between the DAN patients and the controls( P = 0. 001) . While significant analysis between the diabetic patients without DAN complications and the controls was significant by the analysis of mean absolute error(MAE) (P = 0. 007 and P = 0. 000), there was also significant difference between the diabetic patients with and without DAN ( P = 0. 001 ).CONCLUSION: Different impaired degree of autonomic nerve system in diabetic patients can result in nonlinear reduction in HRV analysis, but nonlinear predictability analysis can prove the nonlinear model of HRV,. and can also provide important state information of sympathetic and parasympathetic systems so that it can provide reliable evidence for the earlier diagnosis and the prognosis evaluation of DAN.
