Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02220361 1997-11-06
P~K~ROUND OF THE INVENTION
i) Field of the Invention
This invention relates to a method and
apparatus for determining the thickness of bones of the
skull in vivo.
ii) Description of Prior Art
Cranial bone grafts have become the substrate
of choice in rehabilitation and reconstruction of the
craniomaxillofacial skeleton; and the calvarium is
generally accepted as the best donor site. Advantages of
the calvarium as a donor site for bone reconstructive
surgery include accessibility, the inherent contour and
abundance of harvestable bone, improved graft volume
survival and the inconspicuousness of the donor site.
Knowledge of calvarial thickness at the donor
site would be of significant assistance to surgeons
harvesting calvarial grafts.
Computerized tomography provides a reasonable
estimate of cranial thickness, however, this imaging
modality lacks precision when extrapolating for in situ
assessment.
It has been demonstrated that calvarial bone
thickness is subject to regional variation and thus far
accurate methods for the in situ or in vivo measurement
of skull thickness are not available.
Previously it has been suggested that 6 mm of
parietal bone thickness is the threshold for safe in situ
calvarial harvesting, while others have suggested that a
2 mm thickness margin of the diploic space should be the
limiting factor in obtaining a safe separation plane in
split cranial harvests.
Despite the limited number and nature of
predictive studies on the assessment of skull thickness,
the importance of recognizing the variation between
minimum and maximum skull thickness at a particular site
CA 02220361 1997-11-06
is established. This variation can approach 8 mm. The
variations and lack of predictability of bone thickness
based on tables and charts of average data places
predictability well beyond the margin of safety required
in the context of consistently harvesting cranial bone
without compromising the patient.
Knowledge of the thickness of the maxilla in
preparation for cranial facial surgery or dental surgery
would also be of value, so as to avoid penetration
through the maxilla and potential damage or infection of
underlying organs.
Ultrasound has been used in other environments
in evaluating thickness or changes in thickness of an
industrial article such as result from disintegration or
flaking or separation of parts of the article. U.S.
Patent 5,351,544 describes a measuring apparatus which
employs high frequency ultrasound to determine the
thickness and/or flaking state of a specimen along its
depth. U.S. Patent 5,440,929 describes an ultrasonic
device for measuring thickness of a bottom plate of an
oil storage tank. U.S. Patent 3,985,022 describes a
technique for ultrasonic measurement of a work piece.
U.S Patent 5,009,103 describes an ultrasonic thickness
measuring method and apparatus. U.S. Patent 5,585,563
describes a method for determining the thickness of a
specimen such as an optical lens without mechanical
contact.
All of these prior methods employing ultrasound
are directed to evaluating articles of manufacture where
the opposed surfaces of the article are accessible or can
be located. None of these prior Patents is concerned
with measurement or evaluation of bone or the like, in
vivo.
CA 02220361 1997-11-06
The bones of the skull include the cranium
bones and the facial bones, for example, the calvarial
and the maxilla.
SUMMARY OF THE INVENTION
This invention seeks to provide a method and
apparatus for determining the thickness of bones of the
skull in vivo.
In particular the invention seeks to provide
such a method and apparatus as a preparatory step to
harvesting calvarial grafts for use in reconstructive
surgery.
Further the invention seeks to provide such a
method and apparatus for determining the thickness of the
maxilla in preparation for cranial facial surgery or
dental surgery.
In accordance with the invention there is
provided a method of determining the thickness of bones
of the skull in vivo comprising: i) transmitting pulsed
ultrasonic waves having a frequency of 1 to 3 MHz into
bone, having outer and inner opposed faces, in vivo, in a
zone where it is desired to determine thickness of the
bone, between the inner and outer opposed faces, ii)
receiving the combined echoes of ultrasonic waves
reflected at said outer face, and ultrasonic waves
transmitted through said bone from said outer face and
reflected at said inner face, iii) converting the
received echoes from ii) into electrical signals, iv)
windowing the electrical signals, v) computing the power
spectrum of each windowed electrical signal, vi)
differentiating the power spectrum to prove a wave plot
of frequency and amplitude, vii) measuring the distance
between adjacent peaks of the wave plot, and viii)
determining the thickness of the bone between said inner
and outer opposed faces of said zone from the distance
measured in step vii).
CA 02220361 1997-11-06
In accordance with another aspect of the
invention there is provided an apparatus for determining
the thickness of bones of the skull in vivo comprising:
a) an ultrasonic transducer for generating pulsed
ultrasonic waves having a frequency of 1 to 3 MHz, b)
receiving means for receiving reflected echoes of the
ultrasonic waves, c) means for converting the reflected
echoes into electrical signals, d) window processing
means for windowing the electrical signals, e) a spectral
analysis unit for developing power spectra of the
windowed signals, f) means for differentiating the power
spectrum of each signal to provide a wave plot of
frequency and amplitude, g) a peak detector unit for
determining the distance between peaks of said wave plot,
and h) comparator means for calculating the thickness
from the distances determined by the peak detector unit
in g).
DESCRIPTION OF Pn~nK~ EMBODIMENTS
The invention may be applied to exposed bone or
to bone which has a tissue covering, for example, a
covering of skin or mucosa of the oral cavity, depending
on the site of the bone under investigation.
In either case it is necessary to determine the
zone of the plot of the reflected waves, within which the
bone falls. In particular the zone is identified by
windowing using a generalized cosine window, for example,
a Hamming Window or a Hanning Window. This zone is
selected based on known data of bone thicknesses and is
selected so as to encompass and extend beyond the bone
inner and outer faces.
In the case where the bone is covered by a
tissue covering such as skin, account must be taken of
the thickness of the skin, in determining the windowed
zone embracing the inner and outer faces of the bone.
This may be achieved in accordance with the invention by
CA 02220361 1997-11-06
a preliminary step in which pulsed ultrasonic waves
having a frequency of 6 to 8 MHz, preferably 7 MHz are
transmitted through the tissue covering into the bone;
echoes from the tissue surface and the bone outer and
inner faces are received and converted into electrical
signals. From these signals the thickness of the tissue
covering, or in other words, the distance to the bone
outer face through the tissue covering is determined and
from this is determined the windowing zone for the bone
thickness determination.
The distance to the bone outer face through the
tissue covering can be determined by a time of flight
measurement. More especially, the time of flight from
the transducer to the point of reflection from the bone
outer face is determined using a peak detection until
which identifies peaks in the electrical signal derived
from the reflected echoes to identify the bone outer face
and the tissue surface.
In connection with this latter embodiment of
the invention the apparatus further includes an
ultrasonic transducer for generating pulsed ultrasonic
waves having a frequency of 6 to 8 MHz. This may be a
transducer separate from the transducer employed to
generate the pulsed ultrasonic waves having a frequency
of 1 to 3 MHz; or a transducer may be employed which has
an adjustment or control enabling shift between the two
classes of pulsed ultrasonic wave required in this
embodiment of the invention.
In the differentiated power spectrum employed
to determine the bone thickness, the distance between
adjacent peaks of the wave plot is measured. This
measurement may be of the distance between adjacent
negative peaks, or the distance between adjacent positive
peaks. More especially the distance between adjacent
major peaks is employed.
CA 02220361 1997-11-06
BRIEF n~-CCRTPTION OF DRAWINGS
FIG. 1 illustrates schematically an
apparatus of the invention for carrying out the method of
the invention,
FIG. 2 shows a windowed original wave signal
plot of amplitude with time;
FIG. 3 shows a power spectrum plot of
amplitude against frequency derived from the window of
Fig. 2;
FIG. 4 shows a differentiated power spectrum
derived from the plot of Fig. 3;
FIG. 5 illustrates schematically an
apparatus of the invention in a different embodiment; and
FIG. 6 shows a windowed wave signal plot of
amplitude with time for the apparatus of Fig. 5.
n~.~rRTpTIoN OF ~n~r~nn~ EMBODIMENTS
WITH n~n NCE TO THE DRAWINGS
With further reference to Fig. 1, an apparatus
schematically illustrated includes one or more
ultrasonic transducers (TDR) 12 each having a transmitter
(TX) 14 and a receiver (RX) 16. The dotted lines
indicate one or a plurality of transducers 10.
Receiver 16 communicates with a windowing/-
filtering unit 18 which communicates with a spectral
analysis unit 20 which communicates with a power spectrum
computing station 22.
Computing station 22 communicates with a
filtering unit 24 which communicates with a peak
detection unit 26.
Peak detection unit 26 communicates with a
comparator unit 28 and with a display unit 30.
Comparator unit 28 also communicates with display unit
30.
In operation, the apparatus 10 provides a
numerical value for bone thickness based upon the signals
CA 02220361 1997-11-06
8 --
received by receiver 16 from transmitter 14 of the
transducer 12.
The invention is further described by reference
to the method as applied to measurement of the thickness
of cranium bone, in vivo.
The transducer 12 suitably with a stand-off
medium is placed against an exposed clean area of the
cranium bone which is a potential site for bone
harvesting; low frequency ultrasound waves of 1 to 3,
preferably about 2 MHz are generated by transducer 12.
Some of the ultrasound waves are reflected by
the outer exposed surface of the cranium bone and some
are transmitted through the bone and are reflected at the
concealed inner face of the cranium bone opposed to the
outer exposed face. The combined reflected waves produce
an echo signal which is received by receiver 16.
A standoff medium such as an aqueous gel may be
disposed between the transmitter 14 and the outer exposed
surface of the cranium bone.
A stand off medium is an acoustically
transparent medium, for example, water or a water-soluble
gel, which provides a separating distance between the
transmitter 14 and the clean surface.
Receiver 16 converts the echoes into an
electrical signal which signal may be amplified and
stored in a memory (not shown) for future processing by
the windowing/filtering unit 18 or may be fed directly
for processing by the windowing/filtering unit 18.
In unit 18 the electrical signal from receiver
16 or from memory storage is processed or windowed using
a generalized cosine window (e.g., Hamming Window or
Hanning Window) for a region that embraces the bone
thickness region. This is illustrated in Fig. 1 which
shows the electrical signal from receiver 16 and a
CA 02220361 1997-11-06
windowed zone W. The data within window zone W are used
for subsequent processing.
Thus the windowed zone W is a zone which bounds
the outer and inner faces of the bone; thus the
extremities of the windowed zone are beyond the outer and
inner faces of the bone, respectively.
The unit 18 filters the data of the windowed
zone W, typically with a bandpass filter having a filter
range from 10,000 to 2,000,000 Hertz. This filtering
removes electronic background noise having a frequency
above 2MHz and below 10,000 Hz. The filtered signal is
processed in spectral analysis unit 20 to produce a
normalized power density spectrum as illustrated
schematically in Fig. 3. This shows the signal in the
form of a plot of amplitude against frequency of the
echoed waves.
The normalized power density spectrum is
differentiated at power spectrum computing station 22 to
provide a differentiated amplitude by frequency display.
The signal from station 22 is optionally filtered,
typically with a low pass filter to smooth out the signal
by removing electronic background noise and the output is
processed in a peak detection unit 26, which determines
the distance between adjacent major negative peaks fi-fi-
1~ fi+l-fi~ fi+2-fl+l etc-, as illustrated in Fig. 4, the
frequency fi-l, fi~ fi+l, fi+2~ fi+3~ etc- at each
sequential negative peak is also determined corresponding
to each Of fi-l~ fi, etc-
The thickness D of the cranium skull isdetermined by comparator unit 28 in accordance with
equation I:
D = V/2(fi+l ~ fi)
where
CA 02220361 1997-11-06
-- 10 --
V is the estimated average velocity of
sound in bone,
fi and fi+l identify frequencies of any two
adjacent negative peaks f~ fi, etc-
The thickness D may be calculated in comparatorunit 28 using a single pair of adjacent frequencies fi-l,
fi, fi+l, etc. or as an average based on multiple pairs
of adjacent frequencies, for example, fi-l and fi; fi and
fi+l; fi-l and fi+3, etc.
The method can be carried out in a like manner
in which peak detection unit 26 determines the distance
between adjacent major peaks, with the frequency at each
sequential positive peak being determined corresponding
to each of the major positive peaks, instead of the major
negative peaks.
Verification of the accuracy of the method of
the invention was carried out employing a porcine skull
and comparing the thickness determined by the method of
the invention with the thickness determined by digital
calipers. In each case the average of 10 measurements
was employed. The average thickness of the skull
achieved with the method of the invention was 2.96 mm
whereas the average achieved with the digital caliper at
the same location was 3.11 mm, a difference of only 0.15
mm.
With further reference to Fig. 5 there is shown
a variation of the apparatus of Fig. 1. The apparatus 10
of Fig. 1 is employed in an embodiment in which the
transducer 12 is placed against exposed bone, with or
without a stand off medium, to determine the thickness of
the bone.
In a second embodiment, employing an apparatus
such as is illustrated schematically in Fig. 5, the bone
thickness measurement is conducted through the overlying
CA 02220361 1997-11-06
skin surface or other surface such as the mucosa of the
oral cavity.
With reference to Fig. 5 the components having
the same function as in Fig. 1 are identified by the same
integers as in Fig. 1, raised by 100. Thus Fig. 1 shows
an apparatus 110 having one or more transducers 112 each
having a transmitter 114 and a receiver 116.
It will be seem that the apparatus 110 of Fig.
differs from the apparatus 10 of Fig. 1 in the
inclusion of a tissue/bone peak detection unit 119.
With reference to the embodiment of Fig. 5, the
thickness of the skin or the distance of the bone from
the transducer 112 is determined in order to identify the
precise region of the windowing unit 118 and for the
subsequent bone thickness measurement. Typically a 6 - 8
MHz sampling range is used with a preferred value of 7
MHz. Given the velocity of sound in biological tissue,
1540 m/sec., the wavelength of a 7 MHz ultrasound wave is
approximately 2.2e - 4 m. Thus, in a 1 mm thick sample
of skin on top of a bone approximately four ultrasonic
oscillations would occur before the entering ultrasonic
wave would be reflected from the bone. This gives rise
to both an entering reflection from the skin surface and
a reflection from the bone surface.
An example of these reflections appears in Fig.
6. The distances from the transducers 112 to the bone
through the skin can be determined using a time-of-flight
measurement. The time-of-flight from the transducers 112
to the reflection from the bone surface is determined
using peak detection unit 119. The detected peaks (see
Fig. 6) pl (surface) and p2 (bone) are then used to
determine the starting point for the windowing unit 118.
The windowed zone is identified as W, in Fig.
6, and it is this zone W which is filtered in unit 118.
The procedure described with reference to Fig. 1 is
CA 02220361 1997-11-06
thereafter followed employing the windowed zone W
developed as in Fig. 6.
EXAMPLE
In vivo validation of the ultrasonic thickness
method of the invention was conducted using a live
Landrace pig (25kg +/- 2kg). This animal was made
available prior to euthanasia for the primary
experimental purpose to study the effects of hepatic
xenografts. It has been previously sedated with an
intramuscular injection of diazepam (2 mg/kg) and
anesthetized with inhalation of isofluorane 1-2%, mixed
with 95% oxygen and 5% carbon dioxide. Intravenous
sodium pentobarbital and prophylactic ibeprenorphine were
administered to ensure appropriate anesthesia and
analgesia, respectively.
The scalp of the pig was incised and retracted
to the level of the suborbital rims bilaterally. A
periosteal elevator ensured removal of a remaining soft
tissue overlying the calvarium. The point chosen for
sampling was chosen to reflect the variability in
thickness of the porcine skull and permanently marked for
later caliper measurement.
Thickness of the porcine skull was determined
using two different techniques - direct digital caliper
measurements and ultrasonic thickness measurement using
the methods described above. Ultrasonic measurements
were performed first.
Ultrasonic measurements were performed using a
Matec (model SR9000 [Matec Corporation, Nautick, MA,
U.S.A.]) microcomputer based pulser/receiver. A single
crystal (1.27 cm diameter) unfocussed, broadband
transducer was attached to the pulser/receiver. The
center frequency of the transducer was 1.0 MHz. with a
6dB bandwidth of 1.84 MHz. The transducer was placed on
an acoustic standoff device. This device was a
CA 02220361 1997-11-06
cylindrical tube containing de-ionized water. The output
of the pulser/receiver was digitized at a sampling rate
of 25 MHz using a Tecktronix ~model TDS 250) digital
oscilloscope whose output was connected to the GPIB bus
(IEEE 448) of the microcomputer Ten (lO) samples were
stores on the microcomputer hard disk. These digital
samples were then windowed using a generalized cosine
window and filtered using a bandpass from lO kHz to 2.0
MHz. The power density spectrum was calculated for the
windowed region; see Fig. 3. Power density spectrum
minima are identified as fi-l~ fi, fi+l, fi+2, fi+3- The
stores power density spectrum was then differentiated and
plotted in spectral format; see Fig. 4. The negative
peaks were then detected, see fi-fi-l~ fi+l-fi~ fi+2-fi+l
in Fig. 4. These peaks were used to determine the
thickness of the bone sample using formula (I) above.
fi-fi-l = 543,000 Hz D = 3.00 mm
fl+l-fl = 555,000 Hz D = 2.93 mm
fi+2-fi+l = 551,000 Hz D = 2.95 mm
After completion of the ultrasonic measurements
a full thickness bone wedge was created using a sagittal
saw. The bone segment underwent calvarial thickness
measurement at the same marked points used for the
ultrasound measurements. A digital caliper, Mitutoyo
(model 500 [Mitutoyo Limited, London, UK]) with a
resolution of 0.01 mm was utilized to make 4 measurements
at the marked point.
RESULTS
The average of the 4 caliper measurements of
the wedge bone samples was 3.11 mm while the average for
the 10 ultrasound samples was 2.96. The difference
between the two measurements was 0.16 mm.
