Examenspresentation av Elein Khaled

Supervisor: Seyed Moein Pishnamaz

Examiner: Mikael Persson, Professor

Abstract:
Hydrocephalus is a disease that affects 1-2 newborns per 1,000, characterized by an
abnormal accumulation of cerebrospinal fluid (CSF) in the brain ventricles. The
primary treatment is the insertion of a shunt in the head to drain the CSF to another
part of the body. However, there is a risk of shunt malfunction. Therefore, regular
monitoring is necessary, especially in newborns. CT Scans and MRI are the most
commonly used diagnostic tools for hydrocephalus. However, the potential risks
associated with the frequent use of ionizing radiation in infants and the sensitivity of
MRI to motion complicates its use in children.
This thesis aims to find alternative monitoring methods by further developing a
medical device called StrokeFinder MD100, designed for microwave diagnosis of
strokes in adults, to make it applicable also for children to detect changes in ventricles’
volume. For this purpose, a matching gel was prepared to be placed between the
child’s head and the device’s antennas to fill the air gap created by the small size of
the child’s head compared to that of an adult. First, the optimal dielectric properties
for the matching layer were determined using CST and Matlab simulations. Then,
the two matching mediums with the determined dielectric properties were made and
attached to the device’s antennas.
Two phantom models of the brain were made to assess the functionality of the
matching mediums, and another head phantom with an adjustable ventricle was used
to mimic the hydrocephalic state. A series of measurements were performed using a
vector network analyzer to determine the effects of the matching medium on the wave
penetration in the phantom. Finally, an algorithm was developed to analyze the data
collected by the device and detect changes in ventricle size for the head phantom.
Based on this approach, the most appropriate matching medium that provides stable
measurements suitable for the child’s head was prepared. Moreover, the algorithm
successfully detects changes in ventricle size. The results are promising and represent
a step forward in the development stages of the device.

Welcome!

Elein and Mikael