Fuzzy pattern classification of hemodynamic data can be used to determine noninvasive intracranial pressure
The authors previously introduced a method in which intracranial pressure (ICP) was estimated using parameters (TCD characteristics) derived from cerebral blood flow velocity (FV) and arterial blood pressure (ABP). Some results suggested that this model might be influenced by the patient’s state of cerebral autoregulation and other clinical parameters. Hence, it was the aim of the present study to improve the method by modifying the previously used global procedure in certain subgroups of patients.
In 103 traumatic brain injured patients (3–76 years, mean: 31 ±16 years) signal data of FV, ABP and ICP were used to generate samples of TCD characteristics together with time corresponding ICP. Fuzzy Pattern Classification was used to identify cluster subsets (classes) of the sample space. On each class a local estimator of ICP was defined. This approach provides a non-invasive assessment of ICP (nICP) as follows: Using FV and ABP the TCD characteristics were computed and related to the matching classes. nICP was calculated as a weighted sum of local ICP estimations.
ICP A and B waves and long-term trends could be visibly assessed. The median absolute difference between ICP and nICP was 5.7 mmHg.
The class structure of the model facilitates nICP assessment in heterogeneous patient groups and supports a stepwise extension of the target patient group without affecting the former validity.
KeywordsIntracranial pressure cerebral autoregulation cerebral blood flow transcranial Doppler ultrasonography fuzzy pattern classification
Unable to display preview. Download preview PDF.
- 1.Aaslid R, Lundar T, Lindegaard KF, Nornes H (1993) Estimation of cerebral perfusion pressure from arterial blood pressure and transcranial Doppler recordings. In: Miller JD, Teasdale GM, Rowan JO, Galbraith SL, Mendelow AD (eds) Intracranial pressure VI. Springer, Berlin, pp 226–229Google Scholar
- 2.Baur M, Bocklisch SF (2001) Similarity based local model approach for nonlinear modelling. In: Proc 6th European Control Conference. ECC01, Porto, Portugal, pp 3905–3910Google Scholar
- 3.Bocklisch SF, Päßler M (2000) Fuzzy time series analysis. In: Hampel et al (eds) Fuzzy control — theory and practice. Physica-Verlag, HeidelbergGoogle Scholar
- 4.Chan KH, Miller JD, Dearden NM, Andrews PJ, Midgley S (1992) The effect of changes in cerebral perfusion pressure upon middle cerebral artery blood flow velocity and jugular bulb venous oxygen saturation after severe brain injury. J Neurosurg 77: 117–130Google Scholar
- 7.Klingelhöfer J, Conrad B, Benecke R, Sander D (1987) Relationship between intracranial pressure and intracranial flow patterns in patients suffering from cerebral diseases. J Cardiovasc Ultrasonogr 6: 249–254Google Scholar
- 9.Lundberg N (1960) Continuous recording and control of ventricular fluid pressure in neurosurgical practice. Acta Psych Neurol Scand [Suppl] 149: 1–193Google Scholar
- 16.Zabolotny W, Czosnyka M, Smielewski P (1994) Portable software for intracranial pressure recording and waveform analysis. In: Nagai H, Kamiya K, Ishii S (eds) Intracranial Pressure IX. Springer, Berlin, pp 439–440Google Scholar