Stationarity in Neuromonitoring Data
Purpose: Signals reflecting the metabolic and circulatory status of an injured central nervous system are normally corrupted systematically. The patient is part of a therapeutic control-loop and the signals acquired are rather determined by the quality of
- PDF / 296,253 Bytes
- 3 Pages / 595.28 x 790.87 pts Page_size
- 8 Downloads / 152 Views
Abstract Purpose: Signals reflecting the metabolic and circulatory status of an injured central nervous system are normally corrupted systematically. The patient is part of a therapeutic control-loop and the signals acquired are rather determined by the quality of control (stationarity of signals) than by the underlying pathological process. Methods: To verify the control-loop hypothesis, neuromonitoring data from 12 randomly selected severely head injured patients (initial GCS £ 8, 7 men, 5 women) were analysed for circulatory (blood pressure, intracranial pressure [ICP], cerebral perfusion pressure [CPP]) and metabolic (arterial blood gases, jugular bulb oxygenation [SjvO2], brain tissue oxygen partial pressure [ptiO2]) variables (n = 10). A total of 120 time series of generally not equidistant sample intervals were assessed for stationarity by Wallis & Moore’s runs test. Results: Non-stationarity could only be proven in 23 time series, i.e. the control-loop hypothesis was violated. Trends were mainly found in CPP (n = 5) and ICP (n = 4). The remaining cases spread out on all but one (temperature) signal. Nine patients showed at least one time series with a trend. One patient had clear trends in five out of ten variables that focused on SjvO2, ptiO2, ICP and CPP. Conclusions: Absence of stationarity in about 20% of time series is credited to an effective therapeutic controlloop. For analytical purposes, however, the benefit seems to be overestimated. Consequently, neuromonitoring should be considered the analysis of short-term disturbances that are intentionally compensated for by a short response time. Information content is thus reduced even if the number of sensor devices increases.
H.E. Heissler () and J.K. Krauss Klinik für Neurochirurgie 7249, Medizinische Hochschule Hannover, 30623 Hannover, Germany e-mail: [email protected] K. König and E. Rickels Klinik für Unfallchirurgie, Orthopädie und Neurotraumatologie, AKH, Celle, Germany
Keywords Severe head injury • Neuromonitoring • Therapeutic control loop • Stationarity • Entropy
Introduction Multimodal neuromonitoring is a concept of measuring signals in a critically ill patient to gain data about the patient’s current status. Since the degree of miniaturisation and integration of sensors has been evolved to reasonable calibres, a multitude of physiological parameters (circulatory, metabolic, etc.) that produce streams of data have been feasible. Our fundamental idea was to establish an early warning system based upon artificial neural networks (ANN) to predict intracranial pressure (ICP) elevations, hypothesising that an increase in ICP is a multivariate phenomenon, i.e. elevations could be traced back to some changes in the signals involved in monitoring. Real-world neuromonitoring implies that data are conditioned, i.e. signals (time series from biosignals) are kept within reasonable physiological limits and devolutions depict controlled variables from a therapeutic control-loop rather than a freely running signal (Fig. 1). Thus, we may hav
Data Loading...
