Nonlinear features of heart rate variability in paranoid schizophrenic

摘要: Cardiovascular mortality is significantly increased in patients suffering from schizophrenia. However, psychotic symptoms are quantified by means of the scale for assessment positive and negative symptoms, but many investigations try to introduce new etiology psychiatric disorders based on combination biological, psychological social causes. Classification between healthy paranoid cases has been achieved time, frequency, Hilbert---Huang (HH) a those features as hybrid features. Those extracted transform each intrinsic mode function (IMF) detrended time series case case. Short-term ECG recordings 20 unmedicated acute schizophrenia obtained matched peers have utilized this investigation. Frequency features: very low frequency (VLF), (LF), high (HF) HF/LF (ratio) produced promising success rate equal 97.82 % training 97.77 validation IMF1 ninefolds. Time---frequency [LF, HF ratio, mean, maximum (max), minimum (min) standard deviation (SD)] provided 100 both trials ninefolds IMF2. By utilizing ninefolds, Hilbert---Hang HF, mean value envelope ( $$\bar{a}$$ ¯ )] 96.87 95.5 validation, respectively. analyzing first IMF using [mean, max, min, SD, median, quartile (Q1), third (Q3), kurtosis, skewness, Shannon entropy, approximate entropy energy, ), level variation ([ $$\dot{a}$$ ? (t)]^2), central $$(\bar{W})$$ W ) number zero signal crossing $$(\left| {\bar{W}} \right|)$$ ] 90 validation. Time, HH [energy, VLF, LF, ratio 97.5 95.24 sixfolds. classification rate, emerged highest than domain separately.