Transducers, sensors, and instrumentation in

clinical biomechanics
P.A. Payne

Department of Instrumentation and Analytical Science, UMIST, P.O. Box 88, Manchester, M60
IQD, UK

Presented at the 28th Annual Scientific Meeting of the Biological Engineering Society,
Salford, UK, 7-9 September 1988

ABSTRACT
Successful measurements in anyJield are dependent on the availability of appropriate transducer materials and the associated
instrumentation. In recentyears there has been a most welcome advance in both these areas. If we consider&t the transducer
developments that have recently taken place, these have much to do with the discovery and application of new materials such as
electroactive polymers, jibre optic devices and many others. Instrumentation has largely benejitted from the microelectronics
revolution. Our ability to process and ultimately display data has improved to a remarkable extent. Indeed, the designer of
instrumentation is under enormous pressure to convert the data into the digital domain as early as possible simply because the
resulting instrument will usually be easier to design and construct, more accurate, more reliable, smaller and possibly more
important still, cheaper. This paper reviews some of these developments, gives a number of examples of applications in the
clinical biomechanics area and makes some predictions for future developments.

Keywords: Biomechanics, transducers, instrumentation

INTRODUCTION (m.r.i.) systems, which are capable of providing

images of internal organs of the body and, under
The science of measurement has emerged over
certain circumstances, providing information on the
recent years as a subject deserving the same recogni-
cell chemistry. Low cost m.r.i. systems may have to
tion as other major branches of science. Indeed, most
await commercial exploitation of high temperature
physical science endeavour is fundamentally de-
superconductors’.
pendent on measurement 3. This observation is also
Clinical biomechanics instrumentation has tradi-
true in the case of medical and biological measure-
tionally relied upon fairly simple devices. In this
ments. However, in addition to the normal con-
context the transducer employed is critical. The
straints that apply in the design of measurement
terms sensor and transducer are frequently treated as
systems, we must take into account such things as the
though they are interchangeable, although, as has
anatomy of the measurement site, the physiological
been pointed out2,3, there are distinct differences
activity and the safety of the subject under investiga-
between them. A transducer is involved in energy
tion.
conversion from one form to another, normally
The theory and design basis for medical and
electrical, whereas a sensor simply recognizes the
biological measurement systems is dependent on
presence of the measurand. Most so-called sensors
many disciplines such as human or animal physio-
are, in fact, transducers and it is difficult to identify
logy and anatomy, solid state physics, electrical,
any useful device that acts only as a sensor.
electronic and mechanical engineering, acoustics,
and fluid dynamics. The subject is therefore both
complex and interdisciplinary. Therein lies its
fascination for many of its practitioners.
RECENT DEVELOPMENTS IN
In the context of medicine, the ideal measurement
TRANSDUCERS
system would be low cost, non-invasive (no req,&re-
ment to mechanically penetrate the skin and free- Transducers rely on materials that act as energy
dom from forms of energy kiiown to be hazardous), converters. Table I gives examples of a number of
rapid, precise, reliable and safe both for the patient energy conversions, together with the transducer
and the user. It would not require data recording for ‘materials that may be used for the purpose. It can
subsequent replay and analysis and wduld use low be seen that many uses have been found for piezo-
energy sources, especially for long-term ambulatory electric materials. Miniature vibration transducers
monitoring. This ideal medical instrument awaits and accelerometers have been developed based on
invention. However, many of the requirements piezoelectric materials. These are commercially
noted are met by magnetic resonance imaging available from a number of companies worldwide.

0 1989 Butterworth & Co (Publishers) Ltd

0141-5425/89/OaO18W5 $03.00
180 J. Biomed. Eng. 1989, Vol. 11, May
 Transducers and instrumentalion in biomechanics: P. A. Payne

Table 1 Examples of some transduction principles (adapted from reference 4)

Variable Tranduction Example

principle

‘l‘rmperaturr Thermoelectric Thermocouple; thermopile

Thermocapacitive Change of capacitance (of, for example, ceramic capacitors)
Thermoresistive Thermistors; platinum resistance thermometer
Thermal expansion Glass thermometer; bimetallic strip; quartz crystal (frequency change)
Radiation Thermopile; thermistor; photovoltaic semiconductor; photoconductive
semiconductor
Thermochemical Liquid crystals (temperature sensitive chemical changes)
Pyroelectric PZT (polarization change with temperature)

Dispacement
(a) Low mechanical impedance types Resistive Potentiometer (rotational and linear)
Capacitive Change in capacitance
Inductive Change in inductance
Transformer AC excited differential transformer
Electromagnetic Movement of coil through a magnetic field (no DC response)
Optical Moire fringe; interferometry; reflection change
Ultrasonic Pulse transit time
(b) Higher mechanical impedance types Piezoresistance Very small displacements
Mechanoresistance Metal foil and wire strain gauge; mercury gauge
Velocity Magnetic induction Moving coil (gives output proportional to velocity)
Doppler effect Ultrasonic/optical electromagnetic frequency change

Accrleration F=ma Measurement of the force due to the movement of a known mass (e.g.
using piezoelectric, magnetostrictive or mechanoresistive transducers)

Force Elastic displacement Primary: displacement of an elastic member. Secondary: displacement

transducer as listed above
Piezoresistive Semiconductor (e.g. silicon)
Piezoelectric Barium titanate (ferroelectric ceramic-lacks DC response)
Magnetostrictive Induced voltage type (lacks DC response); variable inductance type
Balance Electromagnetic force balance (null-type system)

Pressure Elastic displacement Diaphragm deflection (Bourdon tube, bellows, etc.) measured with a
displacement transducer
Force balance Manometer; electromagnetic force balance

Flow Pressure difference Orifice type

Mechanical Rotameter: driving an electromagnetic EMF generating secondary
transducer
Thermal Heat transport from a heated element (e.g. self-heated thermistor)
Electromagnetic Electromagnetic flow EMF generated by conducting liquid flowing
through a magnetic field
Ultrasonic Pulse transit time; Doppler shift; cross correlation
Optical Laser Doppler shift
Indicators Dye dilution; thermal dilution; conductivity dilution
Transit time Marker transit time (bubbles, solids, etc. in carrying fluid)

Table 2 Comparison of parameters for various piezoelectric materials

Piezoelectric materials
As indicated in Table 1, piezoelectric materials are Material Acoustic d33 4
used in many ultrasonic applications. One of the impedance (p C N _ ‘) Electromechanical
(Kg me2 s-‘) coupling
major transducer material developments is currently coefficient
occurring in this field. Since the report of relatively
high piezoelectric activity for polyvinylidene PZT5A 33.7 375 0.49
fluoride (PVDF) in 196g5, considerable effort has PVDF 4.0 20 0.20
gone into trying to understand how this effect arises (uniaxially drawn)
and attempting to improve on the performance. The PVDF 4.0 SO
(biaxially drawn)
ultimate aim is to approach the performance of VDF-TrFE’ 4.51 145 0.30
conventional ceramic materials such as lead zir- copolymer
conate titanate (PZT)‘. A comparison of some prop- Copolymer blends” 4.6 180
erties of PVDF, PZT and more recently de-
veloped polymer piezoelectric materials is given in
Table 2. A distinctive property of these polymer
materials is their low characteristic acoustic im- electrical transmission lines are considered, it can be
pedance. This may be shown to be equal to the concluded that a transducer material whose char-
product p,c (reference 9) where p0 is the density of acteristic acoustic impedance is close to that of the
the material being studied and c is the propagation transmission medium is the best for use in an
velocity of a plane acoustic wave in the material. aqueous environment. In this context, all the
If analogies between acoustic transmission and polymer materials score well. The most recently

J. Biomed. Eng. 1989, Vol. 11, May 181

 Transducers and instrumentation in biomechanics: P.A. Payne

Table 3 Transmit and receive efficiency parameters for various strutted with simple photolithography. No particu-
piezoelectric materials lar steps need to be taken to reduce inter-element
coupling because the material is inherently lossy at
ultrasonic frequencies.
PVDF has also found a role in making pressure
Quartz 1.0 1.0 1.0
measurements under the foot13,14, an application in
PZT5A 70.0 0.2 1 14.6
LiNbO, 2.8 0.54 1.51 which its ability to deform easily is a major advan-
PVDF (biaxial) 6.9 1.35 9.3 tage. Another intriguing application is in using
VDF-TrFE 7.4 1.89 14.0 PVDF to help mimic the skin’s ability to sense
Poly-blend - - 17.1 contact pressure15. Again, the advanced forms of
polymer piezoelectric material will also give rise to
No dimensions are quoted for this tabulated data, instead all
parameters are related to a quartz standard
improvement in these applications. It should be
added that much further work is necessary in order
to understand how some of the effects giving rise to
developed polymer blend materials are beginning high piezoelectric activity come about. This may not
to come close to an ideal ultrasound transducer appear so important to the potential user of the
material. This direction of research may well reveal material until it is realized that from such an under-
even better materials in due course. standing may come ways of further improving the
Much has been written on the subject of com- performance of transducers fabricated from these
parisons between conventiona ceramics (e.g. PZT5) materials.
and polymer materials. Numerous performance
indices are used to support the prejudices of particu- Fibre optic transducers
lar authors, but in this context the critical criterion is A new communications industry has grown up
the go-and-return performance of the material in a worldwide based upon the exploitation of optical
standard transducer configuration. Comparative fibres as light waveguides. This has been so successful
measures of performance” of this form have been that the cost of both silica and polymer based fibres
used. They have proved to be simple, rapid and has fallen greatly. This has allowed the measurement
useful in developing and exploring the performance scientist to seriously suggest a wide variety of appli-
of new materials. A performance index that explores cations of these devices. A few of these have become
the go-and-return aspects of piezoelectric materials commercially available.
has been suggested by Callerane et al.” and is There are two main approaches in the use of
discussed by Silk ‘*. Silk sugg es t s t hat the Callerane optical fibres for making measurements. In one
approach is more valid for the comparison of per- approach the fibre is simply used as a means of
formance between materials of fundamentally dif- guiding light energy to the end of the fibre where a
ferent characteristics. The Callerane et al.” transduction process takes place. In the other ap-
performance index Yr Y, is given by: proach the intrinsic properties of the fibre are ex-

Y,Y, = -.-
x: t ploited, perhaps over a long distance, giving rise to a
distributed measurement. In either case, the reasons
l-K2 ‘9
I c33 for using optical fibres are essentially the same. The
where Yr is a transmit efficiency parameter, Y, a material itself is low cost, light in weight and can be
receive efficiency parameter, K, the electro- of very small dimensions (a few micrometres in
mechanical coupling efficiency (longitudinal wave), diameter). However, the most compelling reason is
cj; ultrasonic velocity at constant displacement and t their almost complete immunity to interference from
the material thickness. As Silk points out, some of the electromagnetic radiation. Additionally, they will
results of using such a performance comparison are also link in well to an optical communication or
surprising. Table 3 extracts some figures from refer- computation systemt6.
ence 12 and compares these with figures for Y, Y, Optical fibres have been exploited for physical17,
for the more recent copolymers and polymer blends. chemicalI and biochemical’g measurements. These
Analysis of these figures shows that PZT5A is are all areas of some interest in biomechanics. Con-
outstanding as a transmitting material, but less good siderable effort has gone into temperature measure-
as a receiver; PVDF is a poor transmitter, but good ment systems based on some form of interferometer
in the receive mode; VDF-TrFE approaches PZT5A method. These may be of interest where an invasive
on the basis of Yr Y, and polymer blend materials measurement of local temperature is required. Pres-
are the best. sure sensing can also be achieved on a miniature
These data and comparisons suggest that a major scale. This may also be exploitable for invasive
change in the approach to ultrasound probe design measurements. In the area of chemical measure-
may be imminent. Considerable advantages accrue ments considerable efforts have gone into measuring
from the use of polymer materials for both single dissolved gases. Transducers for pH (ref. 20) PO, and
element and multi-element devices. No matching PcO? (ref. 22) have been developed. Many of these
layers are required due to the low acoustic imped- have application in bio-mechanics. Due to their
ance value of these materials. They are compliant small size, robust construction and low cost, they
and can be readily fashioned into shapes not possible may eventually become the transducer of choice in
for ceramics. They can be produced in very thin measurements of blood and tissue gases. Further
sheet form to provide a basis for very high frequency work on the bio-compatibility of these devices will
devices. Finally, multi-element devices can be con- have to be carried out if they are to be used for long-

182 J. Biomed. Eng. 1989, Vol. 11, May

term implantation. Optical fibre transducers for Custom-made transducers on silicon will be expen-
other species, such as glucosez3, have also been sive.
developed. These may eventually play an important
role in continuous monitoring for insulin delivery. Conducting polymer transducers
Considerable work has gone into understanding the
phenomenon of electrical conduction in certain
OTHER TRANSDUCER DEVELOPMENTS polymer materials such as polypyrrole**. The pos-
sibility of employing such materials in a measure-
Resonator devices
ment mode has also been explored. Some intriguing
Mechanical resonators, often cylindrical, in which a possibilities have emerged, for example, an attempt
piezoelectric driver excites the resonance and the to model the human olfactory system using an array
presence of the measure and disturbs the resonant of polypyrrole transducerszg. Materials such as this
frequency, have found wide application24. Measure- change conductivity when exposed to a range of
ments of flow, mass flow and density are possible. An gaseous species. By ‘doping’ the surface of a device
adaptation of these devices may prove useful for in with suitable thin layers of polymers, transducers
vitro work in medicine. with reasonable specificity can be obtained. Con-
Of more direct interest are resonant transducers siderable work has been carried out to achieve
based on quartz crystal devices. Quartz is chosen for reproducible synthesis conditions for such devices,
this application because of its wide availability and particularly in the context of devising anaesthetic
very high magnification factor (@. By the addition gas monitors3’. There is good reason to believe that a
of a thin coating of an appropriate material to the new family of transducers will eventually emerge
quartz resonator, it is possible to selectively adsorb from this activity.
gases and create what is effectively a mass-balance
capable of detecting micro- or even picogrammes of
adsorbentz5. The problem with such devices is their INSTRUMENTATION
sensitivity to water vapour. Therefore a second The microelectronics revolution has had a major
reference transducer is usually employed. There is influence on instrumentation in recent years. It has
also the problem of poisoning by an unexpected gas meant that some measurements previously thought
species which may entirely destroy the transducer. impossible are now commonplace. Good examples of
this are measurements which rely upon complex
signal processing and spectral analysis.
Silicon based transducers More recent developments have been associated
The enormous advances26 made in fabricating inte- with the reduction in cost and improvements in
grated circuits on silicon can also be readily ex- performance of small personal computers. These
ploited for measurement. Although silicon is not an devices by themselves do not make a measurement
ideal transducer material, the various techniques by possible, but very often they permit a measurement
which it can be purified, selectively doped and to be more conveniently conducted, to be more
etched to incredible accuracies mean that excellent precise and to be conducted without close super-
transducers can be produced which can also contain vision. How well this is accomplished is dependent
considerable computing or signal processing power on the quality of the software. This is a further area
adjacent to the sensing element. Both physical quan- of expertise now required of a measurement scientist.
In the past, many measurements have been
tities (mass, acceleration, pressure, etc.) and, by
using a MOSFET whose gate area is suitably ex- limited by the performance of the analogue-to-
posed via a membrane”, chemical or biochemical digital converter. Recently many new devices have
quantities can be measured. emerged giving rise to the possibility of, for example
The addition of computing power close to the 125 MHz, seven-bit conversion3’. These appear to
transducer provides the possibility of delivering out- cater for almost any requirement in the areas of
put data in a digital form. This alone is of great use measurement discussed in this paper. The integra-
because data in digital form can be employed im- tion of the transducers with the remainder of the
mediately by the computer which is usually one part measurement system is a design problem, especially
of any major measurement system. However, an if transducers with some built-in computational
additional benefit, is that the transducer can begin power are available. The problem is similar to that
to act independently when required. With the of deciding how soon to convert data into the digital
appropriate design, the transducer can be made self- domain. Today the solution is probably to do so as
adapting. It can then change its transfer function to soon as possible.
compensate for non-linearities or for temperature,
humidity or other environmental perturbations. The
CONCLUSIONS
term ‘smart sensor’ has been applied to such devices.
Some measurements of interest in biomechanics To a large extent the microelectronics revolution
might be possible via these devices, although in went at such a fast pace that it left sensors and
many cases, due to physical size, in vivo measure- transducers behind. Only in recent years, as the
ments are unlikely. There is also the question of cost, realization that measurement science represents an
because for biomechanics any standard, and there- important area of activity, have things begun to
fore inexpensive, device might not be suitable. change. New materials are now being explored and

J. Biomed. Eng. 1989, Vol. 11, May 183

 Transducersand instrumentationin biomechanics:P.A. Payne

new approaches to measurements in biomechanics 15. De Rossi D, Domenici C, Lazzeri L, Nannini A. A new skin-
can be anticipated. Some of these have been dis- like sensor for contact stress measurement in biomechanics.
cussed in this paper, but many others exist or will Proc 28th Ann Sci Meeting on Clinical Applications of Biomechan-
its, Salford, UK, 1988 (abstract).
come about. Without doubt there is an exciting
16. Usher MJ. Sensors and Transducers. Basingstoke: Macmillan,
period of activity ahead.
1985: 121-5.
17. Jones JDC, Jackson DA. Research advances in fibre optic
interferometric and polarimetric sensors. Laser Focus
REFERENCES October 1985, 143-6.
1. Maddox J. Solid state physics; liquid-nitrogen-temperature 18. Narayanaswamy R, Sevilla III F. Optical fibre sensors for
superconductors arrive. _Nature 1987; 325: 756. chemical species. J Phys E Sci Instrum 1988; 21: 10-7.
2. Payne PA. Sensors and their applications. J Phys E Sci 19. Li PYF, Narayanaswamy R, Payne PA. Biosensors based on
Ins&urn 1983; 16: 947-51. chemically sensitive fibre optic devices. Proc 3rd Int IMEKO
3 Usher MJ. Sensors and Transducers. Basingstoke: Macmillan Conf on Measurement in Clinical Medicine, Edinburgh, 1986;
1985: 3. 5965.
4. Cobbold RS. Transducers JOY Biomedical Measurements. New 20. Kirkbright GF, Narayanaswamy R, Welti NA, Fibre-optic
York: Wiley Interscience, 1974. pH probe based on the use of an immobilised calorimetric
5. Kawai H. The piezoelectricity of polyvinylidene fluoride. indicator. Analyst 1984; 109: 1025-8.
Japan 3 Appl Phys 1969; 8: 975-6. 21. Peterson JI, Fitzgerald RV, Buckhold DK. Fibre optic
6. Pointon AJ. Piezoelectric devices. IEE Proc 1982; 129 A: probe for in vivo measurement of oxygen partial pressure.
285-307. Anal Chem 1984; 56, 62-7.
7. Koga H, Ohigashi H. Piezoelectricity and related properties 22. Zhujum Z, Seitz WR. A carbon dioxide sensor based on
of vinylidene fluoride and trifluoroethylene copolymers. 3 fluorescence. Anal Chim Acta 1984; 160: 305-Q.
Afig Phys 1986; 59: 6. 23. Schultz JS, Mansouri S, Goldstein IJ. Affinity sensor: a new
8. Banks RE, Smith S, Nicholls M, Moss F, Payne, PA. technique for developing implantable sensors for glucose
Piezoelectric copolymer blends and a method for determin- and other metabolites. Diabetes Care 1982, 5: 245-53.
ing the activity of piezoelectric copolymers and blends 24. Langdon, RM, Resonator sensors. 3 Phys E Sci Instrum 1985;
thereof. UK patent application, 1988, 8811003.6. 18: 103-15.
9. Hill CR, ed. Physical Principles of Medical Ultrasonics. Chiches- 25. Alder JF, McCallum JJ. Piezoelectric crystals for mass and
ter: Ellis Horwood, 1986: 58. chemical measurements. A review. Anabst 1983; 108;
10. Payne PA, Moss F, Nicholls M, Banks RE, Smith S, Jawad 1169-89.
SM. A comparison of properties of PVDF and VDFTrFE 26. Middelhoek S, Noorlag DJW. Silicon microtransducers. 3
copolymer materials for transducers. Proc Ultrasonics Interna- Phys E Sci Instrum 198 1; 14 : 1343-52.
tional ‘85 ConzGuildford: Butterworths, 1985; 405-10. 27. Turner APF, Karube I, Wilson GS, eds. Biosensors: Funda-
11. Callerane J, Tancrell RH, Wilson DT. IEEE Cat. 79CH mentals and Applications, Oxford : Oxford University Press,
1482-9SU. London: IEEE, 407. 1987.
12. Silk MG. Ultrasonic Transducers for Non-destructive Testing. 28. Skotheim TA, ed. Handbook of Conducting Polymers, Vol 1 and
Bristol: Adam Hilger, 1984. 2, New York: Marcel Dekker, 1986.
13. Derbyshire B, Platts RGS. A shapeable foot-pressure mea- 29. Persaud K, Dodd G. Analysis of discrimination mechanisms
suring system. 3 Biomed Eng 1989; 11: 258-264. in the mammalian olfactory system using a model nose.
14. Dario P, Starita A, Bergamasco M, Francesconi R. Piezo- Nature 1982; 299: 352-5.
electric polymer-based platform system for the measure- 30. Payne PA, Harris NK, Bartlett JG. Gas sensor. UK patent
ment of foot-ground pressure distribution. PYOCBES 28th Ann application, 1988, 8807618.7.
Sci Meeting on Clinical Ap$ications of Biomechanics, Salford, 31. Data sheet TP4196, Teledyne Philbrick, Dedham, MA,
UK, 1988 (abstract). USA, 1988

184 J. Biomed. Eng. 1989, Vol. II, May