Open Access Repository

Modelling the electric field from implantable defibrillators


Downloads per month over past year

Gale, TJ 1995 , 'Modelling the electric field from implantable defibrillators', PhD thesis, University of Tasmania.

PDF (Whole thesis)
whole_GaleTimot...pdf | Download (8MB)
Available under University of Tasmania Standard License.

| Preview


This thesis presents a mathematical model of the electric field from implantable
defibrillators, together with the numerical implementation, validation
and examples of application of the model.
The model was based on Laplace's equation for potential and was implemented
using the boundary element method with constant quadrilateral
elements and realistic torso structures. An efficient out-of-core solver was developed,
allowing any size problem to be solved, subject only to computer
speed and time available. A method was also developed that allowed matrices
calculated in one problem to be used in other, similar problems, often reducing
calculation times by an order of magnitude. Model validation included comparison of myocardial potentials from the
model to those from a finite element model (r.e..2.8%) and from measurements
in a sheep (c.c.=0.464, r.e..23.6%). Validation was also done against
resistance and voltage at defibrillation threshold from 29 patients implanted
with a transvenous system and 8 patients with the transvenous system and an
additional subcutaneous patch. Without the patch, the relative error between
the average of the clinical results and the model result was 9.4% (voltage)
and 0.8% (resistance). The average of the relative errors between each clinical
result and the model result was 23.4% (voltage) and 11.6% (resistance). With
the patch, the equivalent relative errors were 33.9%, 19.4%, 44.0% and 22.5%.
Transvenous, epicardial and subcutaneous electrode configurations were
modelled in a series of investigations. The best transvenous configuration was
with a right ventricular cathode and an anode in the inferior vena cava, where
defibrillation voltage and energy were reduced by 35% and 55%, respectively,
compared to a standard configuration with the anode in the superior vena
cava. Configurations with a right ventricular cathode and large epicardial
patch performed best, though, and reduced voltage and energy by up to 59%
and 79%, respectively. The optimal length of the right /ventricular transvenous
electrode was approximately 60mm. An infarcted heart was also modelled.
For future work, anisotropy may be added to the heart and skeletal muscle
of the model. Anisotropic regions may be represented by many small boundary
element regions or by finite elements. Automated construction of the torso
mesh and an algorithm for automatically optimising electrode position may
be developed. Individual patients may be modelled and predicted values of
defibrillation voltage, energy and resistance compared to values measured at
the time of implantation.
In conclusion, the boundary element model was successful in modelling the
electric field in the torso and in predicting implantable defibrillator performance.
The model has potential to be used in research and development and
in clinical settings.

Item Type: Thesis - PhD
Authors/Creators:Gale, TJ
Keywords: Electric countershock, Implantable cardioverter-defibrillators
Copyright Holders: The Author
Copyright Information:

Copyright 1995 the Author - The University is continuing to endeavour to trace the copyright
owner(s) and in the meantime this item has been reproduced here in good faith. We
would be pleased to hear from the copyright owner(s).

Additional Information:

Includes bibliographical references (leaves 158-169). Thesis (Ph.D.)--University of Tasmania, 1996

Item Statistics: View statistics for this item

Actions (login required)

Item Control Page Item Control Page