Article
Analysis of intracranial pressure time-series using wavelets (HAAR basis functions)
Search Medline for
Authors
-
Hans E. Heissler
- Klinik für Neurochirurgie, Medizinische Hochschule Hannover, Germany
-
Kathrin König
- Allgemeines Krankenhaus Celle, Klinik für Unfallchirurgie, Orthopädie und Neurotraumatologie, Celle, Germany
-
Joachim K. Krauss
- Klinik für Neurochirurgie, Medizinische Hochschule Hannover, Germany
-
Eckhard Rickels
- Allgemeines Krankenhaus Celle, Klinik für Unfallchirurgie, Orthopädie und Neurotraumatologie, Celle, Germany
| Published: | September 16, 2010 |
|---|
Outline
Text
Objective: Cerebrospinal fluid dynamics still remain only incompletely understood. In particular, due to the lack of standards interpretations of intracranial pressure (ICP) remains subjective. Transforming ICP into frequency domain commenced in the early eighties, arriving at the conclusion that cerebrospinal parameters mirror in ICP spectral composition. The classical mathematical tools applied were not suitable in handling intrinsic signal nonstationarity thus affecting analyses greatly. To overcome these obstacles already eminent at that time, we have focussed on a novel approach based upon orthogonal basis functions.
Methods: During routine diagnostic volume-pressure testing epidural ICP was acquired in 118 patients with suspected cerebrospinal fluid circulatory disorders. Pressure was digitized by 40 samples/s and conditioned to separate infra frequent signal components. ICP fluctuations were computed by subtraction of original and infra frequent ICP constituents. Subsequently multiresolution analysis was performed on fluctuations by discrete HAAR wavelet transform and coefficients displayed in dyadic fashion (scalogram).
Results: As expected the decomposition of ICP fluctuations led to typical patterns in the scalogram. Episodes of pathologically classified wave activity and artificial ICP changes were topographically detectable in the time-frequency-plane. By selective deletion of wavelet coefficients unique signal characteristics could be made visible by inverse transform. Wavelet coefficients establish an objective basis for intracranial pressure assessment independent from the rater's experience with this matter.
Conclusions: The wavelet approach is a sophisticated signal processing method to estimate the spectral development of intracranial pressure in time in one procedural step. It is therefore superior to classical FOURIER methods that are limited in analysing real-world data. HAAR wavelets are fast and robust. Their disadvantages do not counterbalance the advantages in this biomedical application.
