缺血缺氧脑损伤心率变异的1f分形特性短时检测.pdfVIP

ASPHYXIC-ISCHEMIC BRAIN INJURY DETECTION BY SHORT-TIME ANALYSIS OF 1/f FRACTAL CHARACTER IN HEART RATE VARIABILITY ABSTRACT According to the statistics from World Health Organization (WHO), the number of people suffered from ischemia has risen up rapidly, which becomes a top killer among other deadly diseases. In developed countries, stroke is the third most common cause of death. About 80% of all acute strokes are caused by ischemia. In China, 1.2 1.5 million people are down with all strokes and 0.8 1 million are dead of it every year. Meanwhile, 75% of these survivors are disabled and the recurrent rate is up to 41%. The longer time of ischemia persists, the severer the symptom is and the higher rate of sequelae. Due to this important clinical significance, ischemia has attracted more attention in the past years. The objective of this paper is to find one or more parameters from Hear Rate Variability (HRV) analysis under ischemia, hoping to give a primary alarm during asphyxia. Thus, we may be able to prevent long-term blood deficiency in the brain and create more time for first-aid rescue. Based on the traditional point progress model, we discussed 1/f fractal character in HRV from three different views : scale character, self-similar progress and instantaneous energy mutation. Correspondingly, we introduced three distinct methods: Allan Factor (AF), Wavelet-Transform Standard Deviation (WTSD) and Nonlinear Energy Operator (NEO). During data analysis, we not only made a comparison between the above three methods and also gave an explanation related with auto-nerve modulation mechanism, which showed that short-time analysis of 1/f fractal character in