A Method of Measuring Eye Movemnent Using A Scieral Search Coil in A Magnetic Field-274 PDF
A Method of Measuring Eye Movemnent Using A Scieral Search Coil in A Magnetic Field-274 PDF
A Method of Measuring Eye Movemnent Using A Scieral Search Coil in A Magnetic Field-274 PDF
[7] T. H. Bullock, "Neuron doctrine and electrophysiology," Science Inc., Gendale, Calif. Semi-Annual Rept. No. 4, pp. 27-43;
vol. 129 pp. 997-1002; April 1959. January, 1962.
[8] T. H. Bullock and S. Hagiwara, "Intracellular potentials in [18] E. R. Lewis, "An electronic model of the neuron based on the
pacemaker and integrative neurons of the lobster cardiac dynamics of potassium and sodium ion fluxes," in Neural Theory
ganglion," J. Cell. and Comp. Physiol.,vol 50, pp. 25-48; August, and Models: Proc. of the Ojai Conference, Stanford University
1957. Press., Stanford, Calif.; in press.
[9] S. Hagiwara and T. H. Bullock, "Intracellular recording from [19] H. Grundfest, "Excitation triggers in post-junctional cells," in
the giant synapse of the squid," J. Gen. Physiol., vol. 40, pp. "Physiological Triggers," T. H. Bullock, Ed., American Physio-
565-577; March, 1957. logical Society, Washington, D.C., pp. 119-151; 1957.
[10] T. H. Bullock and C. A. Terzuolo, "Diverse forms of activity in [20] A. L. Hodgkin and A. F. Huxley, "The components of membrane
the somata of spontaneous and integrating ganglion cells," J. conductance in the gaint axon of loligo," J. Physiol., vol. 116,
Physiol., vol. 138, pp. 341-364; 1957. pp. 473-496; 1952.
[11] T. Otani and T. H. Bullock, "Effects of presetting the membrane [21] A. L. Hodgkin and A. F. Huxley, "A quantitative description
potential of the soma of spontaneous and integrating ganglion of membrane current and its application to conduction and
cells," Physiological Zoology, vol. 32, pp. 104-114; April, 1959. excitation in nerve," J. Physiol., vol. 117, pp. 500-544; 1952.
[121 E. Florey, "Recent studies on synaptic transmitters," Am. Zo- [22] G. Hoyle, "Comparative physiology of conduction in nerve and
ologist, vol. 2, pp. 45-54; February, 1962. muscle," Am. Zoologist., vol. 2, pp. 5-25; February, 1962.
[131 P. Fatt, "Physics of nerve processes," Repts. on Progress in [23] F. F. Hiltz, "Analog computer simulation of a neural element,"
Physics, vol. 21, pp. 112-143; 1958. IRE TRANS. ON BIO-MEDICAL ELECTRONICS, vol. BME-9, pp.
[14] J. C. Eccles, "The Neurophysiological Basis of Mind," The
Carendon Press, Oxford, England; 1952. 12-20; January, 1962.
[15] J. Del Castillo and B. Katz, "On the localization of acetylcholine [24] L. D. Harmon, "Artificial neu,ron," Science, vol. 129, pp. 962-
receptors," J. Physiol., vol. 128, pp. 157-181; 1955. 963; April, 1959.
[16] P. Fatt and B. Katz, "An analysis of the end-plate potential [25] L. D. Harmon, "Studies with artificial neurons, I: Properties
recorded with an intra-cellular electrode," J. Physiol., vol. 115, and functions of an artificial neuron," Kybernetik, vol. 1, pp.
pp. 320-370; 1951. 102-107; December, 1961.
[17] E. R. Lewis, "The Resistor-Capacitor Integrating Circuit: an [26] F. Jenik, "Pulse processing by neuron models," in "Neural Theory
Important Functional Unit in Neural Simulation," Lab. for and Models: Proc. of the Ojai Conference," Stanford University
Automata Research, Librascope Division, General Precision, Press, Stanford, Calif.; in press.
Summary-With the subject exposed to an alternating magnetic advantages over existing methods that make it applica-
field, eye position may be accurately recorded from the voltage gen-
ble to a wide variety of physiological studies involving
erated in a coil of wire embedded in a scleral contact lens worn by
the subject. Using two magnetic fields in quadrature phase and two eye movements (for a review see Riggs et al.1 or Alpern2).
It has the accuracy and sensitivity of the method em-
coils on the lens, one may measure horizontal, vertical and torsional
eye movements simultaneously. The instrument described has an ploying a mirror on a scleral lens3 without requiring an
accuracy and linearity of about 2 per cent of full scale, a resolution
individually fitted contact lens and it can be used to
of 15 seconds of arc and a bandwidth of 1000 cyles per second.
study the largest and the smallest movements of which
INTRODUCTION the eye is capable by merely changing the gain of elec-
rlf HE VESTIBULO-OCULAR reflex, which aids tronic amplifiers. The ability to measure all three com-
in stabilizing images on the retina, requires a nice ponents of eye movement is inherently lacking in some
methods such as that of corneal reflection and the elec-
interplay between the six semicircular canals and
the twelve extraocular muscles. The study of this reflex tro-oculogram. The magnetic field method described
should have at its command an instrument for accu- head here is also unaffected by translatory movements of the
rately measuring eye movement in all three degrees of (or eye) thereby eliminating, for many purposes,
freedom simultaneously. This need has led to the follow- the head clamp or dental biteboard.
ing development. The instrument described has other
1 L. A. Riggs, J. C. Armington, and F. Ratliff, "Motions of the
retinal image during fixation," J. Opt. Soc. Am., vol. 44, p. 315; 1954.
*
Received July 10, 1963. This investigation was supported by 2 M. Alpern, "Muscular mechanics," in "The Eye," H. Davson,
Public Health Service Research Grant AM-05524 from the National Ed., Academic Press, New York, N. Y., vol. 3; 1962. See especially
Institute of Arthritis and Metabolic Diseases. ch. 5, sec. I.
t Department of Medicine, The Johns Hopkins University, I F. Ratliff and L. A. Riggs,
"Involuntary motions of the eye dur-
Baltimore, Md. ing monocular fixation," J. Exp. Psychol., vol. 40, p. 687; 1950.
Authorized licensd use limted to: IE Xplore. Downlade on May 10,2 at 19:326 UTC from IE Xplore. Restricon aply.
138 IEEE TRANSACTIONS ON BIO-MEDICAL ELECTRONICS October
frequency (27r X 5000) and A sin ¢ is the area of the coil Fig. 1 A simplified schematic of the use of a scleral field coil
projected onto the x-y plane which links the flux. for obtaining the vertical (k) component of eye position.
Numerically if N is 10 and H, is 2.19 gauss peak, then a
vertical gaze 0 of 10° yields a voltage e, of 2.2 mv rms.
Since the noise level of the amplifier G is below 2 ,uv, this
already indicates the ability to resolve at least one-
thousandth of this movement or 36 seconds of arc. If the
eye looks down by - 10, the phase of the voltage re-
verses. By using a phase detector and comparing the
signal to a voltage e, derived from a fixed coil that inter-
cepts the field H, a polarity reversible dc signal e< is pro-
duced which is proportional to sin 0.
One coil on the lens and one magnetic field would per-
mit the measurement of only one degree of freedom.
Before proceeding to three degrees of freedom it is
necessary to define appropriate angles in a coordinate Fig. 2-The coordinate system of the orbit and the globe defining
system. These are shown in Fig. 2. The unprimed axes the angles of horizontal gaze 0, vertical gaze 4 and torsion 4'.
x, y and z are taken as fixed in the orbit. They are the so-
called axes of Fick. The primed axes x', y' and z' are H, sin ct, that is 900 apart from H2 in phase. The phase
fixed in the eyeball and move with it. The angles 0 detector of Fig. 1 will reject all quadrature components
(longitude, yaw, azimuth, horizontal duction), 4 in e1 and will continue to produce a signal ed cc sin ¢ in
(latitude, pitch, elevation, vertical duction) and 4 (roll, spite of the addition of the H, field. If e1 is now amplified
torsion) were first introduced into ocular physiology by and phase detected against e, a signal from a stationary
Fick4 in 1854. There are, of course, many other systems coil intercepting H, only, a dc signal eo is produced pro-
in which to specify eye position (Alpern5) but that shown portional to sin 0 cos 0 which is the projection of the coil
in Fig. 2 in conjunction with the magnetic field system onto the y-z plane. Finally, a second coil is wound on
offers output signals which are related to the angles the lens which effectively lies in the y'-z' plane. The
themselves in a reasonably simple way. In what follows, voltage generated in it, e2, is phase detected against e, to
0, q and 4 will be taken as the definition of the hori- produce a signal proportional to this coil's projection
zontal, vertical and torsional components of gaze. onto the x-y plane, namely sin 41 cos 0. In short, by the
One can now measure 0 by introducing another mag- use of two magnetic fields spatially and temporally
netic field H, directed along the x axis. This will also in- in quadrature and two lens coils in spatial quadrature,
duce a voltage in the coil and to distinguish voltages due one can, by phase detection, produce the three dc
to H, and H. one may use time multiplexing, frequency signals,
coding or phase coding. The latter was chosen since it
seems the simplest method and H, is given the form eo c sin 6 cos 4
ecc sin 4)
e;, ac sin 41 cos 0. (3)
4 A. Fick, "Die Bewegungen des menschlichen Augapfels, "Z. rat.
Medizin- N.F., vol. 4, p. 101; 1854.
5 Alpern, op. cit., p. 16. A more rigorous development will follow.
Authorized licensd use limted to: IE Xplore. Downlade on May 10,2 at 19:326 UTC from IE Xplore. Restricon aply.
1963 Robinson: Measuring Eye Movement Using Search Coil in a Magnetic Field 139
If 0, d> and 1' are all less than 200, sin 0 departs from 0
in radians by less than 2 per cent and cos 'k departs from
unity by less than 6 per cent so that approximately
es oc 0
Authorized licensd use limted to: IE Xplore. Downlade on May 10,2 at 19:326 UTC from IE Xplore. Restricon aply.
140 IEEE TRANSACTIONS ON BIO-MEDICAL ELECTRONICS October
.10AWC i
Fig. 4-The step response of the instrument. Time lines, 10 msec Fig. 5.-Step response of the eye. Mean central latency, 90 msec;
apart; rise time, 0.5 msec; ringing frequency, 1 kc; overshoot less duration of horizontal movement, 45 msec; duration of vertical
than 10 per cent. movement, 55 msec; average slewing speed, 280°/second.
feet from the subject. At the point in time, indicated by Electromyographic evidence,"1 while inconclusive
the bottom trace, the spot is switched from the primary about active checking, does show that a pair of antago-
position to a spot up and to the subject's left at 450 by a nistic muscles exert much more torque during a saccade
slant distance which subtends 150 at the subject's eye. than is necessary to maintain the eye in the new posi-
The central latency of 90 msec is atypically short indi- tion. This is reminiscent of a dead beat posicast control
cating learning and anticipation by the subject. The be- system'2 in which the extra torque is turned off shortly
havior of 14 is in accordance with Listing's law which will before the movement is completed. Fig. 5 indicates that
be discussed shortly. It is difficult to determine if the the turnoff times to the six extraocular muscles are not
latencies in 0 and q5 are the same because of the smooth coincident since the horizontal recti complete their
nature of the initial rise, but within the time resolution movement in 0 10 msec before the vertical recti and
of the record (about 1 msec) they appear the same. obliques as a group finish their action in 4 and 4'. This
However, their subsequent behaviors run different time would then be supporting evidence for two distinct pre-
courses, the horizontal movement (6) taking 45 msec for motor reticular formation nuclei, one of which organ-
completion with a peak slewing speed of 292° per second izes horizontal gaze, while the other organizes vertical
while both vertical (4) and torsional (4') movements take and torsional gaze together."2
55 msec with a peak vertical slewing speed of 2630 per
second. The result is that the eye spins for the bulk of TRACKING
the transient at a slant angle of 33° rather than 450. In Fig. 6, a projected spot from an oscilloscope is
When the a transient is complete, the eye is still about made to move in a circle at 4 seconds per revolution.
30 below the target and finishes by a purely vertical Let a be the angle to the spot measured around the
movement. circle from the left horizontal axis and ,B be the radius of
The time course in Fig. 5 shows neither the overshoot the circle subtended at the subject's eye, in this case
and ringing obtained by Westheimer5 with corneal re- 150. The angles 0 and 4 are then given by
flection or the overdamped approach often found with
the electro-oculogram, neither of which places a burden tan 0 = tan S cos a
on the eye. If the eye were a rigid body, its moment of tan 4 = sin ,B sin a. (6)
inertia would be about 4.12 dyne cm sec2 but since the
vitreous body lags considerably9 the moment of inertia In this case 6 and 4 trace out two sine waves 90° apart
seen by the muscles during a saccade is probably much as shown in the record. The bottom trace is the hori-
less. The moment of inertia of the contact lens is 0.815 zontal component of the circle and acts as a time refer-
dyne cm sec2, so that its added burden is not negligible. ence. The usual corrective saccades and smooth pursuit
This is seen in Fig. 5 in the small initial acceleration. modes"4 are visible. The subject has learned the pattern
Similar results were obtained by Yarbus10 using a suc-
11 J. E. Miller, "Electromyographic pattern of saccadic eye move-
tion contact lens. ments," Am. J. Ophthal., vol. 46, p. 183; 1958.
12 O. J. M. Smith, "Posicast control of damped oscillatory sys-
I G. Westheimer, 'Mechanisms of saccadic eye movements,' tems," PROC. IRE, vol. 45, pp. 1249-1255; September, 1957.
AMA Arch. Ophthal., vol. 52, p. 110; 1954. 13 J. Szentagothai, "Anatomical basis of visuo-vestibular co-
9 A. C. Hilding, "Normal vitreous, its attachments and dynamics ordination of motility," Proc. Internat'l Union of Physiol. Sciences,
during ocular movement," AMA Arch. Ophthal., vol. 52, p. 497; 1954. Leiden, The Netherlands; vol. 1, p. 485; 1962.
10 A. L. Yarbus, "The motion of the eye in the process of changing 14 L. R. Young, "A Sampled Data Model for Eye Tracking Move-
fixation points," Biofizika, vol. 1, p. 76; 1956. (In Russian.) ments," Sc. D. dissertation, Mass. Inst. Tech., Cambridge; 1962.
Authorized licensd use limted to: IE Xplore. Downlade on May 10,2 at 19:326 UTC from IE Xplore. Restricon aply.
1963 Robinson: Measuring Eye Movement Using Search Coil in a Magnetic Field 141
15 J. P. C. Southall, ed., "Helmholtz's Treatise on Physiological The coil voltages in general are
Optics," The Optical Society of America, Rochester, N. Y., trans.
from the 3rd German ed., vol. III; pp. 37-123; 1925. en = Nn(Hi A0)w cos cot - N0(H2-A0)o. sin ct; n = 1, 2
Authorized licensd use limted to: IE Xplore. Downlade on May 10,2 at 19:326 UTC from IE Xplore. Restricon aply.
142 IEEE TRANSACTIONS ON BIO-MEDICAL ELECTRONICS October
,II 1, 2. (12)
Specifically for (9), (10), and using (11) 4. TIGHTLY TWISTED
NO.38 MAUNET > W _U6
2
12.?
COIL 1
el= .A,X,lco[H. sin 0 cos q cos cot- H sin 4 sin cot] \S-LERA
p t ~~~-
-19,
e,-A,AxXC,[Hx(cos 0 cos 4-+ sin 0 sin 1 sin 41) cos wt .%rAIP- } 6.2 ~,12. 7
5 MM
+ H, cos 0 sin 41 sin cot]. (13) ALL DIMENSIONS IN MM
(1,)
eo = elex [L A ATH cG] sin 0 cos 0
Flg. 8 (a) Desigin of suctionl contact leins. (b3 Plan of ,v inding for
e6 = eie- [AA NH woG] sinqf coils 1 anid 2 showing the effec tive sagittal area of coil 2. Onie tUrn
onix is showni, acttual lens carried 10 turnis in each coil.
e,;- Ce- = [,A, XH wG] sin 4 cos A (15)
where G is the lum-iped gain of all ac and dc amplifiers. ably unavoidable with any type of suctioIn lens that
This derivation illustrates the mnethod whereby7 one mayr seats near the limibus, indicates that 30 miin slhoulcd he
calculate sources of error dule to such things as lack of set as a limit on the duration of experimiients. Slit lamiip
phase and spatial quadrature of the magnetic fields, investigation of the cornea after wearing the lenis reveals
lack of spatial quadrature of the lens coils, imlproper no hydration or anoxia. The interior surface of the lenis
lens alignment and phase shifts in the amplifiers. is coated with a contact lens wettinig agent (e.g.,
Soaclens, Burton, Parsonis and Co.) to prevent fogginig.
THE CONTACT LENS A negative pressure of -40 num Hg (fromll a mercury
Fig. 8(a) shows a scale drawing of the lens used. The manometer) holds the lens well and prevents slipping in
optical portion is an afocal or plano lens with a radius of 0 and ¢. Slipping in 4 is much harder to conitrol due to
curvature of 8.8 mm, 0.3 mm thick, 12 miml in diameter the cylindrical symmetry of the lenis. Thle use of 60
of commercial manufacture.'" The scleral portion is mnm Hg helps but is the upper limllit in this directioni,
machined from methyl methacrylate plastic (Lucite) since the cornea begins to bulge slowl\ causinig temll-
into which the lens is cemented. The coils are 10 turns porary curvature inyopia after about 20 iminlutes.
of No. 44 magnet wire cemented in slots in the scleral Scalloping the surface of the lens in contact with the
portion. Fig. 8(b) shows the winding scheme which en- sclera is also an aid but the degree to which this can be
ables coil 2 to have an effective area vector along the x' carried is not vet known. Smooth ridges 0.2 mm-hiigh
prevent slow drifts in 4 and irreversible displacemiients
axis. The air pocket over the cornea is 1 mm thick and
is vented by a laterally placed pipe leading to a very due to an obicularis oculi twitch but during a comiiplete
soft thin polyvinyl tubing. The scleral skirt is limited blink, the force of the lids on the suction pipe coupled
radially to 5 mm to prevent interference with the with Bell's phenomenon will displaceis also a source
the lens in 41 to a
new position. The suction tubing of
fornices, the superior being the least accommodating.
The scleral radius of curvature is already in general annoyance since it may apply torques to the eve despite
clinical use and in view of the corneal clearances allowed its extreme flexibility.
Some thought is being given to
this lens should be applicable to the adult human eye methods of evacuating
the lens without the presence of a
without the need of special fitting. The lens is applied suction tube which would
alleviate both of the above
discussion of contact lenses used
with the aid of a corneal anesthetic (e.g. Ophthaine, E. R. problems. An excellent
Squibb and Sons). The effect wears off in about 15 min, in eye movement studies, especially the "suck on" vari-
but once the lens is in place it may be worn beyond this ety, has been presented by the coil
Barlow.'7
time. Although the limbal diameter of the lens is 14 mmll, Lead assembly is made to by mleans of pockets
ends are held with conducting
anterior chamber drainage is impeded. Intraocular into which the bared wire
pressure rises from normal (15 mm Hg) to 22 mm Hg in
silver epoxy. This enables repair if the leads break. To
20 min with a lens suction of 40 mm Hg and returns to prevent flux linkages
in the departing leads, they are a
normal in about 10 minutes. This consideration, prob- twisted pair of fine magnet wire (No. 38).
Once outside
17 H. B. Barlow, "Slippage of contact lenses and other artifacts in
16 Acknowledgment is dtue to Danker and \W6hlk, Inc. for their relation to fading and regeneration of supposedly stable retinal
generotus assistance. images," Qucart. J Exp. Psychol., vol. 15, p. 36; 1963.
Authorized licensd use limted to: IE Xplore. Downlade on May 10,2 at 19:326 UTC from IE Xplore. Restricon aply.
1963 Robinson: Measuring Eye Movement Using Search Coil in a Magnetic Field 143
lk
Authorized licensd use limted to: IE Xplore. Downlade on May 10,2 at 19:326 UTC from IE Xplore. Restricon aply.
144 IEEE TRANSACTIONS ON BIO-M4EDICAL ELECTRONICS Oc lober
FIELD STRENGTH AND UNIFORMITY the transfer function after feedback between themii is
The field strength midway between two field coils of
radius R, coaxially aligned and spaced a distance 2d I(s) GY(s)
(20)
apart, each carrying NI ampere turns is E(s) rs
t + G--GI(s)
coo 2
4wrXNR2
H(0= oersteds. (16)
10(R2 + d2)3/2 where G is amplifier gain (36.4), r is total load resistanice
(14 Q) and wo is the resonant frequency (4.8 kc). The
If one calculates the rate H/.Ho at which H changes as feedback ratio is
one moves a distance 6 along the axis or 6 off the axis or
the rate of increase (from zero) /Hn/Ho of the flux conm- 1 r
ponent Hn perpendicular to the axis upon sinmultane- 2Q 2woL
ously moving 6 along and 6 off the axis, it is possible to 1
derive the conditions on R and d that simultaneously Y (s) =
minimize all three sources of error. This condition is 1
sL + r +
SC
R = / /8 d - 1.07d. (17)
V 7 is the load admittance. For s =jw and G>>1 (20) is ap-
proximately
This is so nearly unity that for convenience we shall set I (s) wo 2
R = d whence (16) becomes = - o- -
E(s) co r
4irNI 1 xI
H0 = - - = 0.444 (18) which indicates that the field current lags the channiiel
10V/23 R R
input voltage by 900. Since the z field is driven fromii a
and the error in field uniformity (first order approxima- voltage developed across the 2 Q resistor in the x circuit
tion) is found to be and consequently in phase with the x field current, the z
zH 9 a \2 current is locked in by the feedback to be in cluadrature
8k. R) (19) with the x current. Without such feedback, with a Q
Ho of 40, a drift of 0.25 per cent in oscillator frequency
Eq. (19) is a criterion for coil design R based on ac- would cause a phase shift of 11.30 from quadrature be-
ceptable uniformity (AH/Ho vs 6) while (18) dictates tween the two fields HX and Hz. To the phase demiiodu-
the needed ampere turns (and so power) to then create a lators this would appear as a physical skewinig of the
given field strength. For the case at hand, R = 1 foot was fields in space by 11.30 which means that 0 aind ¢. would
chosen to accommodate the subject's shoulders. From be measured in reference to a skewed coordiniate sy s-
(19), if 6 is 1.13 inches the error is 1 per cent which tem. Further analysis of (20) shows that the rate of
limits, for this error, the eye to a cube 2.26 inches on an phase shift with detuning is attenuated by 1/(2G2) after
edge centered about the midpoint between the field feedback. Quantitatively this is a factor of 2640 which
coils. From (18), at 1.5 amperes, the field strength is practically eliminates the effect of oscillator or conm-
1.75 oersted rms (R must be expressed as 30.5 cm). ponent drift on field quadrature.
Since this yields a suitable resolution of the instrument Phase shift in the cables or ac amplifiers will cause a
the choice of R - 1 foot is satisfactory. Other considera- similar skewing of the coordinate system or what is the
tions may dictate a different choice for R. For example, same thing, apparent cross modulation between the
if more freedom of movement of the subject (human or angles 0, 0 and if. The choice of what shall be called a
animal) were desirable, R could be increased by 10 purely horizontal nmovemnent (0) or a purely vertical
allowing 22.6 inches of movement in a 1 per cent uni- movement (0) is somewhat arbitrary but it is convenient
form field. The field strength and resolution would drop to establish them in the nornmal sense in the field of
by 10 (2.5 min of arc for the latter) unless more power vision of the subject. To accomiiplish this, at the begin-
was applied. On the other hand it might be feasible to ning of an experiment, the subject is showni a field coIn-
decrease R to the point where the field coils could be taining a dot in the primary position and four dots lo-
worn in a helmet by the subject giving him "tether" cated above, below, left and right of the primary posi-
mobility. tion. These dots are then taken as the definition of
The magnetic fields are maintained in quadrature by horizontal and vertical. After the lens is in place
the series feedback shown around the power amplifiers. the subject looks left and right, and up and down at
The feedback winding consists of two turns wound the test dots while the demodulator excitation volt-
around either field coil of a pair. The turns ratio is thus ages eZ, e,O and e4,p are phase shifted until cross nmodula-
80:1 which is twice the Q of the load. It can be shown tion is eliminated. Phase shifting is accomplished by in-
that if E is the input voltage coming from either the jecting in series with a given voltage, another voltage in
oscillator or T1 and I is the field current in that channel, quadrature whose magnitude can be controlled by the
Authorized licensd use limted to: IE Xplore. Downlade on May 10,2 at 19:326 UTC from IE Xplore. Restricon aply.
1963 Stratbucker et a!.: Magnetocardiogram 145
phase adjust potentiometers shown in Fig. 9. The ex- voltage gain, a frequency response of 10 kc and a full
citation voltages are picked up across two turn windings scale output of + 2.5 volts across 50 Q. This offers an
(5 volts per turn at a field current of 1.5 amperes) on the available current of +50 ma full scale which is fed
appropriate field coil. This voltage is balanced to ground directly to a mirror galvanometer (Mod 1650) in the
through 200 Q resistors according to manufacturers' recorder which is fluid damped to a factor of 0.64, has a
recommendations. The required excitation voltage at bandwidth of 1900 cycles per second and a sensitivity
the carrier preamplifier should be 4.5 volts rms and of + 10 cm deflection for +37 ma. An oscilloscope
should be stepped down by fixed attenuators not shown monitor is helpful. If eo and e, are placed on the hori-
in Fig. 9. The local oscillator in the Sanborn Co. carrier zontal and vertical plates of the oscilloscope one obtains
preamplifiers may be deactivated simply by removing its a frontal projection of the movement of a hypothetic
plug-in coupling transformer. The carrier preamplifier spot on the center of the pupil.
contains a dc amplifier after demodulation and filtering
which creates a single ended output voltage of + 3 volts ACKNOWLEDGMENT
full scale across 1000 Q. Channel gain is varied by The author wishes to thank Dr. M\I. E. Langham of
means of a fixed step attenuator at the front end of the Wilmer Institute for ophthalmological counsel and
ac amplifier. The Sanborn power supply Mod 350- guidance and Dr. C. F. Hazlewood and Mr. W. J.
500AP contains a transistor dc amplifier with unity Sullivan for cooperating in the experiments.
Summary-Experiments have been conducted which reveal the Time-varying electric fields associated with the con-
existence of a detectable magnetic field associated with cardiac elec- duction process in the heart (or any other irritable tis-
trical activity. The relationship between the magnetic record and the
electrocardiogram has been explored and it is shown that under cer- sue) create the familiar time-varying potentials at
tain conditions of axis orientation the voltage induced into a toroidal points within and on the boundary of the medium sur-
sensing element around the heart has the form of the first time de- rounding the tissue. Maxwell's equations' predict the
rivative of the electrocardiogram. A formula based on Maxwell's existence of a magnetic field associated with any time-
equations has been developed to relate these two phenomona. varying electric field. The concept of a biologically pro-
duced magnetic field was suggested in 1958 by Valen-
INTRODUCTION tinuzzi2 in his classic treatise on magnetobiology. Ap-
E ARLY DEVELOPMENT of electrocardiography parently the suggestion was not subjected to experi-
was based upon the concepts of static field mental verification. Although considerable literature
theory. The fundamental nature of this depend- has accumulated on the influence of magnetic fields on
ence has continued to be recognized to the extent that biological systems, it appears that the first successful
the most recent contributions to this important branch recording of the magnetic field associated with mnem-
of medicine deal almost exclusively with the concepts of brane electrical activity occurred in 1960 when Seipel
electric potential and electric field intensity. and Morrow3 reported the detection of a magnetic field
I S. Ramo and J. R. Whinnery, "Fields and Waves in
Modern
*
Received April 19, 1963; revised manuscript received August 20, Radio," John Wiley and Sons, Inc., New York, N. Y., 2nd ed., pp.
1963. The research reported in this paper was supported in part by 177-206; 1953.
the U. S. Public Health Service, Grant No. HE0133710. 2 M. Valentinuzzi, "Curso de Magnetobiologia y Complementos
t Department of Physiology and Pharmacology, College of de Magnetoquimica," School of Medicine, Montevideo, Uruguay.
Medicine, University of Nebraska, Omaha, Nebr. Published in English by Tech. Info. Center, Space and Info. Div.,
t Department of Electrical Engineering, University of Nebraska, North Am. Aviation, Inc., Downey, Calif., p. 106; October, 1961.
Lincoln, Nebr. 3J. H. Seipel and R. D. Morrow, "The magnetic field accompany-
1 Aberdeen Proving Ground, Md. Formerly with NIH Biomedi- ing neuronal activity. A new method for the study of the nervous
cal-Electronics Trainee, University of Nebraska, Lincoln, Nebr. system," J. TVashington Acad. Sci., vol. 50, pp. 1-4; October, 1960.
Authorized licensd use limted to: IE Xplore. Downlade on May 10,2 at 19:326 UTC from IE Xplore. Restricon aply.