gms | German Medical Science

62nd Annual Meeting of the German Society of Neurosurgery (DGNC)
Joint Meeting with the Polish Society of Neurosurgeons (PNCH)

German Society of Neurosurgery (DGNC)

7 - 11 May 2011, Hamburg

Ultrasound-based diagnosis of intracerebral hemorrhage with multifractal analysis

Meeting Abstract

Search Medline for

  • A. Becker - Klinik für Neurochirurgie, Universitätsklinikum Marburg, Marburg
  • D. Kuhnt - Klinik für Neurochirurgie, Universitätsklinikum Marburg, Marburg
  • L. Benes - Klinik für Neurochirurgie, Universitätsklinikum Marburg, Marburg
  • C. Nimsky - Klinik für Neurochirurgie, Universitätsklinikum Marburg, Marburg

Deutsche Gesellschaft für Neurochirurgie. Polnische Gesellschaft für Neurochirurgen. 62. Jahrestagung der Deutschen Gesellschaft für Neurochirurgie (DGNC), Joint Meeting mit der Polnischen Gesellschaft für Neurochirurgen (PNCH). Hamburg, 07.-11.05.2011. Düsseldorf: German Medical Science GMS Publishing House; 2011. DocP 036

doi: 10.3205/11dgnc257, urn:nbn:de:0183-11dgnc2578

Published: April 28, 2011

© 2011 Becker et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.



Objective: In pediatrics, ultrasound allows for sufficient visualization of intracerebral hemorrhage (ICH). Due to the cranium, in adulthood ultrasound images contain multiple artefacts so that the diagnosis of ICH is hindered. This is based on scattering and attentuation of the ultrasound. However, considering these artefacts, intracerebral structures can be detected as long as there is a sufficient acoustic window. We evaluated the cranium, the echogenic properties of subcortical brain parenchyma and a previously diagnosed ICH in a series of 5 patients with multifractal analysis. Based on the concept by Mandelbrot, multifractals describe images with non-linear structures containing density turnovers.

Methods: Five patients with ICH were evaluated with tissue harmonic imaging of a Siemens SonoLine Elegra, equipped with a 2.5 MHz multiarray probe. The mechanical index was 1.2 and the total gain was adapted until adequate visualization of the intracerebral structures was achieved. After depicting the landmarks mesencephalon, thalamus, third ventrical, opposite lateral ventricle and the contralateral cranium, the ICH, which was previously detected in cranial computed tomography, was also displayed and saved as DICOM. The digital RGB images were subsequently converted into 8-bit grayscale images and normalized according to the histogram. A 30x30 pixel region of interest (ROI) was placed in each of the following: brain parenchyma, ICH and contralateral cranium. The f(α)-spectrum was displayed with multifractal analysis, whereby α(q) with highly positive q-values corresponds to regions with high densitiy and α(q) with highly negative q-values corresponds to regions with low density. To determine the density, ∆α = αmax - αmin was calculated. The t-test was used for comparison of the mean-values.

Results: Mean ∆α was 1.25 (SD 0.15) for ROIs in ICH, 1.67 (0.12) for parenchyma and 1.41 (0.18) for the contralateral cranium. This was statistically significant for ICH/parenchyma (p < 0.001) and cranium/ICH (p < 0.001).

Conclusions: Multifractal analysis of ultrasound data contributes to the differentiation between brain parenchyma and ICH. This can be used as a bedside examination. The hypothesis needs to be strengthened with a larger patient collective to prove that ultrasound is a reliable tool in the primary diagnostics and follow-up for patients with ICH.