Intelligent non-invasive modeling of ultrasound-induced temperature in tissue phantoms



Since ancient times that raising the temperature
of human cells (hyperthermia) is known as a tool
for the destruction of tumor masses, well before the
existence of a molecular understanding of cancer cells.
The use of these therapies can be traced back to 500
B.C., in ancient Greece, where Hippocrates, a physician
whose work is considered a cornerstone in western
medicine, pointed [5] that if a tumor "cannot be cut,
it should burned. If it cannot be burned, then it is in-
curable". Nowadays these therapies are commonly used
for rehabilitation purposes; their use in treatment of
oncological diseases is a major driver for researching,
with outstanding possible benets for the society. E-
cient hyperthermia practice demands knowledge about
the exact amount of heating required at a particular
tissue location, as well as information concerning the
spatial heating distribution. Both of these processes are
then required to be accurately characterized. Until now,
ultrasound heating treatments are being monitored by
magnetic resonance imaging (MRI), recognized as being
capable of achieving a 0:5 oC=cm3 temperature resolution
[14], thereby imposing a gold standard in this eld.
However one can notice that MRI-based techniques,
besides the inconvenient instrumental cost, obliges the
presence of a team of clinicians and limits the hyperthermia
ultrasound treatment area due to the space restrictions
of an MRI examination procedure. This work
introduces a novel non-invasive modelling approach of
ultrasound-induced temperature in tissues. This comes
as a cost eective alternative to MRI techniques,capable
of achieving a maximum temperature resolution of 0:26 oC, clearly inferior to the MRI gold standard
resolution of 0:5 oC=cm3. In order to derive the model,
as in-vivo studies raise ethical issues, a phantom was
utilized, whose composition re
ects the ultrasound reactions
of human tissues. Previous works have investigated
the the possibility of non-invasive temperature
estimation (NITE) in multilayered phantoms by applying
soft-computing methods ??, where it was indicated
the use of backscattered ultrasound (BSU) features to
construct estimators, based on soft-computing methods.
From all the BSU features, the temporal echoshifts
have received a major attention ??. Thee proposed
estimators make use of temperature-dependant
features to estimate the evolution of the temperature,
e.g. temporal echo-shifts are due to changes of speedof-
sound and expansion of the medium. At the present
work we study the possibility of using b-spline neural
networks (BSNN) as reliable non invasive estimator of
temperature propagation in phantoms. The results exposed
demonstrate the feasibility of using intelligent
non-invase techniques as a practicable solution with a
wide range of applications.


Non-invasive temperature estimation, ultrasound, neural networks, tissue phantoms


Non-invasive temperature estimation ultrasound neural networks tissue phantoms

Related Project

Link: Linking Excellence in Biomedical knowledge and Computational Intelligence Research for personalized management of CVD within PHC


Biomedical Signal Processing and Control, Elsevier 2016

Cited by

No citations found