REPOSITORY > RESULTS

Doctoral dissertation

Low-power wireless biomedical signal processing

Author(s): Marko Pavlin (Author), Franc Novak (Supervisor)

Thesis defense date: 02.10.2015

Organization: MPŠ - Mednarodna podiplomska šola Jožefa Stefana

PID: 20.500.12556/ReVIS-15069

Views: 6 | Downloads: 0

Abstract

In this dissertation I am dealing with topics related to procedures and devices which
are part of those procedures with severely ill patients in hospitals. At the beginning I
introduce the reader to the problem and to facilitate the understanding and tracking
solutions given in the dissertation. In the introductory part I present the reasons
why I considered the work in this area. The bottom line is that the users should be
provided with easier and better quality in patient monitoring work when using sensors
and monitors. One of the ways to achieve this is the use of wireless links between
the patient and instrumentation. However, the solution is not entirely trivial. The
described solutions address the problems in the field of individual parameters that
are monitored. While main consideration being low power consumption, there are
also other focusing aspects, like miniaturization and overall efficiency. One chapter is
dedicated to battery operated power sources including energy harvesting solutions.
In addition to the more reliable and continuous treatment of a patient, the presented
work provides solutions to improve the accessibility of the patient, reduces the risk of
infections and improves accuracy and reliability when compared with conventional
cable connections. All those general aspects and improvements are presented in the
introductory part, which is finally concluded with a description of biomedical signals,
their sources and signal dynamics. The following parameters are addressed in my
dissertation: blood pressure, heart rate, oxygen saturation, respiration, temperature
and biochemical analysis. Each following chapter starts with the discussion about
stage of the art in the specific field and presentation of the work of other authors. In
the second half of each chapter, I describe my solutions and possibilities to optimize
signal path for lower consumption. Chapters are finally concluded with the prototypes,
measurements and discussions of the test results.
The first parameter is the blood pressure. I focused on invasive measurement. Sensors
are usually disposable, for single use only. Consequence to that is these sensors
have to meet one requirement: a low price. At the beginning of the chapter, I first
describe pressure sensors followed with the analysis of the entire signal chain. I propose
the optimization for lower power consumption. Blood pressure being a normally periodic
signal, can have sporadic random readings caused by abnormal patient condition.
For this reason, the blood pressure signal from the sensor is continuously sampled and
transmitted in real time. To reduce the power consumption, the complete signal path
should be designed with a low power consumption approach, because it remains active
the whole time. I proposed some optimisations, prepared the prototype and finally
evaluated it with measurements.
In the next chapter I described the bioelectrical signals. The most important representative
of this group is the electrocardiogram. All other measurements of bioelectrical
signals are just a variation of these basic measurements. I started the sensor chain in
such sensors and circuits at the basic signal source, a single cell. Its bioelectric activity
generates small pulses and I thought it was important to know the whole background,
I also briefly studied and described the whole signal generation mechanism. After the
general introduction to bioelectric signals I proposed a miniature, low power circuit for
measuring those signals. I also evaluated and presented dynamic energy consumption
x
of individual regimes of operation. I also introduced galvanically isolated contactless
biopotential sensors. They are important in some special cases and cover a certain
segment of the application where contact electrodes can not be used (e.g. in severely
burned patients).
Biooptical measurements are an important branch in the treatment of patients. I
started with the presentation of biooptical effects and how some parameters can be
measured. The basic building blocks of an optical sensor are described, followed by a
detailed description of the prototype implementation and some approaches to reduce
power consumption. As in other chapters, I also presented the results of the biooptical
sensor measurements.
Breathing rate represents one of the four main vital signs. The most common
method to detect the breathing rate and volume is to measure bioimpedance. When
ECG is measured, the measurement of the impedance of the chest can be done using
the same electrodes and does not represent a big challenge. Because of that I focused
on the non-contact measurement using high-frequency methods. To test this idea, I
constructed a prototype, which has operated in the microwave domain. I described my
idea and principle of operation in detail and finally presented the results of testing.
It might be useful to know when the ill patient coughs. Therefore I also described
the prototype of the cough sensor using accelerometer, which I evaluated and finally
presented the results of power consumption measurements.
In the field of temperature measurement at first glance it seems that there is nothing
to explore or develop, because the thermometer is one of the oldest instruments.
Of course, this is not true. I focused my research on the very sensing element. I
prepared two sets of the prototypes and compared two different technologies. As a
result I made simulations on prepared models and compared calculated results with
the measurements on real parts.
I have dedicated the last chapter in the field of sensors to biochemical sensors. I
discussed two types of the sensors: blood sugar and pH. Both types are the basis for
measuring the whole multitude of parameters with sensors that operate on a similar
principle. Similarly as in the previous chapters I first investigated the existing situation
in this area and described the work of other authors. I constructed my own solutions
and built two prototypes. Finally I measured power consumption and discussed the
results.
Finally, I conclude my thesis with two chapters about power sources and wireless
interfaces. The primary source of energy for all presented sensors is a rechargeable
battery. I investigated the possibilities to extend the operating time with energy harvesting.
I constructed and tested the prototype using an indoor solar cell. I researched
the possibility of using such an indoor source with actual measurements.
Even more Finally I described two wireless interfaces with low power consumption.
I focused on two types of wireless: high data rate interface with higher power consumption
and low data rate interface with lower consumption. I found out working with
both interfaces to be the most optimal way of operation. The balance between both
interfaces is acquired by selectively switching according to actual needs.
The concluding chapter describes first the integrated prototype, where I gathered all
the presented solutions in one single package. The multiparameter wireless prototype
was assembled in my home workshop with extremely low budget approach (it costed
me less than three hundred euros for all the presented work). One of my goals was not
only low consumption of the energy, but also resources to achieve the goal of developing
new medical devices. I hope the presented work proves that.

Attachments

Cite this work