Multiparametric analysis of non-invasive blood pressure variability including methods of nonlinear dynamics

• Published in: Proceedings of the 20th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Vol.20 Biomedical Engineering Towards the Year 2000 and Beyond (Cat. No.98CH36286) Vol. 6; pp. 3068 - 3071 vol.6
• Main Authors: Voss, A., Wessel, N., Hasart, A., Malberg, H., Schirdewan, A., Osterziel, K.J.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 1998
• Subjects: Automatic control; Biomedical monitoring; Blood pressure; Blood pressure variability; Cardiology; Cardiovascular diseases; Frequency domain analysis; Pathology; Patient monitoring; Pressure control
• ISBN: 0780351649; 9780780351646
• ISSN: 1094-687X; 1558-4615
• DOI: 10.1109/IEMBS.1998.746139

The objective of this study was to determine the clinical feasibility of continuous, non-invasive blood pressure recordings for characterizing pathological changes in cardiovascular diseases. Tachograms (interbeat intervals and systolic pressure over the time) of two groups, one consisting of 26 cardiac patients with dilated cardiomyopathy (DCM) and the second group consisting of 24 healthy persons as a control group, were investigated. The tachograms were obtained from a Portapres noninvasive blood pressure monitor device. Artifacts and misclassifications were automatically detected and removed. From these corrected tachograms the authors extracted 41 parameters from different domains. Two (of 10) time domain, 4 (10) frequency domain and 4 (10) NLD parameters did separate significantly (p<0.01) the controls from the DCM patients. In this multiparametric and multidimensional approach the combination of 6 parameters from heart rate and mainly blood pressure variability improves the classification result of the discriminant function techniques to 90%. Furthermore, the authors confirmed the effectiveness of the measures from symbolic dynamics as a useful approach to classify the dynamics and intermittent variability change in blood pressure time series.