 |
Description  |
|
|
This invention relates to an oximeter, and is more in particular directed
to the provision of a reference oximeter and method for operating the
same.
BACKGROUND OF THE INVENTION
Pulse oximetry has become an accepted method of oxygen determination in the
last five years. However, up to this point, all devices on the market have
used the transmission method of detection. The pulse oximeter of the
present invention is based upon a reflectance method. Both of these
methods are based upon several related facts.
First, the concentration of blood in a given location of the body varies
with each pulse of the heart. This variation can be measured by optical
methods by introducing a light source near the skin and detecting either
the reflected or the transmitted light intensity. This intensity is
directly related to the localized blood concentration.
Secondly, the wavelength of the light source determines the effect that
oxygen saturation has on the reflected or transmitted intensity. The
wavelength which does not change intensity with saturation but only with
concentration is called the isobestic wavelength. This isobestic condition
occurs at several wavelengths. By using this wavelength as a reference and
by comparing it to a second wavelength in the red portion of the spectrum,
it is possible to determine the oxygen saturation of the blood
non-invasively.
Current pulse oximeters using the transmissive method require the light
from the emitters to pass through the tissue to the detector on the
opposite side. This requires the sensor to be placed on an area of the
body where the distance from the emitters to the detectors is fairly small
(about an inch at most) and not obstructed by opaque tissue layers, such
as bone. This limits the placement of the sensors to areas such as the
finger tips, ear lobes, or the bridge of the nose.
SUMMARY OF THE INVENTION
The pulse oximeter of the invention uses a reflectance method where both
the emitters and the detector of the sensor are placed next to each other.
The light penetrates the tissue and is reflected back by the various
layers of the skin and by the hemoglobin in the blood. By using this
method, the sensor can be placed almost anywhere on the body where blood
flow is sufficient. The light reflected back by the layers of the skin is
fairly constant (or steady-state) whereas the light reflected back by the
blood changes with each pulse depending upon the amount of blood in the
tissue. Also, the hemoglobin in the blood changes coloration due to the
amount of oxygen.
Pulse oximetry uses the amplitudes of the pulsation signals to determine
the oxygen concentration of the blood rather than the steady-state levels
of the signals. However, the steadstate levels must be nearly equivalent
at all times to be able to measure oxygen saturation from the pusitile
signals. Also, the steady-state, or DC, level of the signals must be
adjustable to allow the sensor to compensate for factors such as skin
pigmentation, skin thickness, and sensor coupling to the skin. This is
achieved through the use of a voltage source to control the drive of the
emitters in the sensor. The voltage source is controlled by the voltage
level output from a digital to analog converter which gets its input from
the microcontroller. This allows the microprocessor to select one of a
multiple of brightness levels, for example 64, for the emitters. If the
skin is heavily pigmented of if the blood flow in the area is limited, the
microcontroller can increase the brightness of the emitters to compensate.
Since the characteristics of the light emitting diodes used as the
emitters can change slightly with continued use or temperature change, it
is necessary to employ an integrator to measure and compensate for changes
in DC levels. The time it takes to compensate for these changes can also
be controlled by the microcontroller. When the brightness of the emitters
needs to be changed, the microcontroller changes the brightness control
voltage and also the time constant of the integrator from one second to
0.01 second. The DC levels are thus quickly changed to the new level and
the new pulsatory signals are ready for sampling in less than one second.
The first bandpass stage is also changed to a low pass filter to allow for
the DC shift to settle out. This circuit is a key element in allowing this
instrument to function properly on a wide variety of patients and over a
wide range of conditions.
Briefly stated, in accordance with one feature of the invention, a
reflectance oximeter is provided comprising a red light source, an
infrared light source, a photodetector for receiving light from said
sources reflected from tissue, first and second control circuits for
energizing said red and infrared light sources respectively at different
instants and signal processing means coupled to said photodetector and
synchronized with said energizing means for determining oxygen saturation
in said tissue. The process of reflection will be understood to include
some degree of diffusion. The signal processing means comprises means for
generating first and second feedback signals corresponding to DC component
of Red and IR light respectively received by the photosensor. Means are
provided for applying the first and second signals to the first and second
control circuits. The first and second control circuits comprise means
employing the feedback signals as negative feedback signals in the
energization of the sources. The control circuits may comprise time
constant circuits, and means for varying the time constant of application
of the feedback signals thereto. The system may comprise a source of a
brightness control voltage, the control circuits further comprising means
responsive to the brightness control voltage for controlling the
energization of the red and infrared sources as a function thereof.
In a further feature of the invention, the reflectance oximeter comprises a
red light source, an infrared light source, a photodetector for receiving
light from the sources reflected from tissue, first and second control
circuits for energizing the red and infrared light sources respectively at
different instants and signal processing means coupled to the
photodetector and synchronized with the energizing means for determining
oxygen saturation in the tissue. The signal processing means comprises
filter means for selecting the a.c. component from the signal output of
the photodetector, and a controllable gain circuit for controlling the
gain of the output of the signal processing means. The gain circuit
comprises a plurality of amplifiers of different gain connected to
simultaneously receive the same signal output from the filter means, an
output circuit, and means for selectively connecting individual ones of
the amplifiers to the output circuit. The plurality of amplifiers may
comprise at least four amplifiers connected to receive the a.c. component
of signals corresponding to infrared light and at least four amplifiers
connected to receive the a.c. component of signals corresponding to red
light. The amplifiers of each group of amplifiers have relative gains of
1, 1.25, 2 and 4.
In a still further feature of the invention, a reflectance oximeter
comprises a red light source, an infrared light source, a photodetector
for receiving light from the sources reflected from tissue, first and
second control circuits for energizing the red and infrared light sources
respectively at different instants, and signal processing means coupled to
the photodetector and synchronized with the energizing means for producing
first and second output signals corresponding respectively to reflected
red light and reflected infrared light for determining oxygen saturation
in the tissue. The signal processing means further comprises means for
detecting relative variation in the first and second signals, and means
for inhibiting the determination of oxygen saturation when the relative
variation exceeds a determined level.
In accordance with another feature of the invention, a reflectance oximeter
comprises a red light source, an infrared light source, a photodetector
for receiving light from the sources reflected from tissue, first and
second control circuits for energizing the red and infrared light sources
respectively at different instants, whereby the photodetector receives
pulses of reflected light corresponding separately to reflected red light
and reflected infrared light, and signal processing means coupled to the
photodetector and synchronized with the energizing means for determining
oxygen saturation in the tissue. The processing circuits comprise means
for sampling the maximum and minimum values of each pulse received from
the photodetector, and means for producing output signals responsive to
the levels of the last maximum and last minimum values of pulses
corresponding to reflected red and infrared light. The processing circuits
may comprise means for determining the difference between the maximum and
minimum levels sensed for each pulse.
In a still further feature of the invention, a reflectance oximeter
comprises a red light source, an infrared light source, a photodetector
for receiving light from the sources reflected from tissue, first and
second control circuits for energizing the red and infrared light sources
respectively at different instants, signal processing means coupled to the
photodetector and synchronized with the energizing means for determining
oxygen saturation in the tissue, and means for matching the d.c. levels of
the output signals from the photodetector corresponding to red and
infrared light. The matching means comprises means for adjusting the
brightness of at least one of the sources as a function of the
corresponding d.c. level output from the photodetector. The matching means
may comprise a brightness control circuit for the sources, a feedback
circuit connected to feed back the d.c. components of the outputs of the
photodetector to the control circuit. The control circuit may further
comprise means for controlling the relative brightness of the light
sources.
Still further in accordance with the invention, a reflectance oximeter
comprises a sensor having a red light source, an infrared light source,
and a photodetector for receiving light from the sources reflected from
tissue. The sensor comprises a housing having an aperture, a sensor
carrier, the light sources and photodetector being mounted to have active
light emitting and light receiving surfaces respectively at one side of
the carrier. An electrical connection arrangement is provided for the
sources and photodetector within the housing, and resilient means are
provided for mounting the carrier in the aperture, whereby the carrier
floats with respect to the housing. The resilient means may comprise means
for biasing the carrier to have a uniform linear pressure for
displacements of the carrier through a determined range, and may comprise
a membrane sealingly holding the carrier in the aperture. The arrangement
may further comprise an interconnection cable extending into the housing,
with the electrical connection arrangement comprising a flexible
connection board having electrical leads thereon, the sources and
photosensor being mounted on the flexible board, and the leads being
connected to the cable. The flexible board is preferably mounted to
resiliently yield to pressure applied to the carrier, and may be U-shaped.
In accordance with another feature of the invention, a reflectance oximeter
comprises a sensor having a red light source, an infrared light source,
and a photodetector for receiving light from the sources reflected from
tissue. The sensor comprises a housing comprising a sensor carrier. The
light sources and photodetector are mounted to have active light emitting
and light receiving surfaces respectively at one side of the carrier
externally of the housing. A light barrier coats the sources and
photodetector on substantially all surfaces except the active light
emitting and receiving surfaces. The light barrier means may comprise
silver paint.
In another feature in accordance with the invention, a sensor arrangement
for a reflectance oximeter comprises a red light source, an infrared light
source, a photodetector for receiving light from the sources reflected
from tissue, and a housing having a sensor carrier. The light sources and
photodetector are mounted in the carrier to have active light emitting and
light receiving surfaces respectively externally of the housing. The
sources and photosensor are positioned along a straight line with the
infrared source being positioned between the red source and photosensor.
The axis of infrared sensor is preferably substantially 0.128 inches from
the axis of the photosensor, and the axis of the red sensor is preferably
substantially 0.085 inches from the axis of the infrared sensor.
In one method in accordance with the invention for reflectance oximetry
wherein red and infrared light sources are separately sequentially
energized, and reflected light from said sources is sensed to produce red
and infrared reflectance signals respectively, the method comprises
separating the a.c. and d.c. components of said reflectance signals,
determining oxygen saturation from said a.c. signals, and adjusting the
brightness of light from said sources to maintain said a.c. signals within
a predetermined range.
In a further method in accordance with the invention for reflectance
oximetry wherein red and infrared light sources are separately
sequentially energized, and reflected light from said sources is sensed to
produce red and infrared reflectance signals respectively, the method
comprises separating the a.c. and d.c. components of said reflectance
signals, determining the difference between the maximum and minimum values
of each pulse of said a.c. component, and determing oxygen saturation from
said difference by comparison of said difference with a look up table. The
step of determining the difference preferably comprises determining the
difference twice for each of the pulses.
In order that the invention may be more clearly understood, it will now be
disclosed in greater detail with reference to the accompanying drawings,
wherein:
BRIEF FIGURE DESCRIPTION
FIG. 1 is a perspective view of the arrangement of the LED's and sensor in
accordance with one embodiment of the invention;
FIG. 2 is a cross-sectional view of the sensor housing;
FIG. 3 is a perspective view of the sensor housing;
FIG. 4 is a block diagram of the system of the invention;
FIG. 5 is a circuit diagram of the power supply;
FIG. 6 is a circuit diagram of the alarm control circuits;
FIG. 7 is a circuit diagram of a voltage output circuit;
FIG. 8 is a connection diagram for clarification;
FIG. 9 is a circuit diagram of the timing circuits;
FIG. 10 is a circuit diagram of the circuits controlling the operation of
the LED's;
FIG. 11 isa circuit diagram showing processing of the output signalsof the
photodetector;
FIG. 12 isa circuit diagram showing part of the bandpass filters;
FIG. 13 isa circuit diagram showing the gain control circuits;
FIG. 14 isa further interconnection circuit diagram;
FIG. 15 isa circuit diagram of the digitial circuitsof the system of the
invention, including the microcontroller;
FIG. 16 isan interconnection diagram;
FIG. 17 isa further interconnection diagram;
FIG. 18 isa front view of the control panel;
FIG. 19 isa circuit diagram of the display panel;
FIG. 20 isa memory map of the microcontroller;
FIG. 21 isa memory map of the system;
FIG. 22 isa table of constants, showing variouslimits thereof;
FIG. 23 isthe Oximeter look-up table;
FIG. 24 isa timing circuit that may be used in the system;
FIG. 25 isa timing diagram;
FIG. 26 isa diagram illustrating constant levels;
FIG. 27 isan overall flow diagram for operation of the system, and;
FIGS. 28-54 are flow diagramsillustrating the operation of the invention.
DETAILED DESCRIPTION OF THE INVENTION
The sensor used with the reflectance oximeter in accordance with the
invention preferably consistsof the following:
a red LED (660 nanometers), and infrared LED (925 nanometers), and a
phototransistor (800 nanometerspeak sensitivity). Multiples of each of the
RED and INFRARED LED's are preferably employed.
The Infrared LED 10 ismounted in the center with the Red LED 11 on one side
and the Phototransistor 12 on the other side. The spacing between each
sensor and the arrangement isvery critical. Saturation hasbeen obtained
with the sensorsin other locationsbut the consistence and repeatability
from site to site ismaximized with the illustrated arrangement.
Sensorshave been placed in a triangular pattern with good resultsbut the
spacing between the three partsand their relationship to each other is
critical.
The sensors are pulsed (turned on and off) at a rate of 256 times per
second with a duty cycle of about 1/8. Each LED is turned on in
synchronism with the phototrans is tor and the sample and hold input
circuit of the instrument. Thus the instrument sees the light reflected
back from the skin from one LED at a time.
The attachment of the sensor to the skin is also very critical. A small
amount of force (approximately 10 grams ) must be maintained to adequately
hold the sensors in contact with the skin and yet not apply too much
pressure to cause the area to be traumatized and thus cause blood flow
restrictions. Tests have been conducted to determine if the sensor must be
in contact with the skin. Results have shown that the sensor need not be
in contact. Saturation has been obtained from as far away as 1/2 inch.
Although housings of a rigid plastic have been used with good results, the
most favorable concept is one of more flexability, one that would more
easily and completely conform to the surface of the skin on which it was
placed. Thus, either of two designs have been found to allow for proper
operation. In one the whole housing is flexible and semi compliant with
the sens or area on a floating membrane to insure proper pressure and
contact. The other is of a harder type housing with the same sensor area
as for the floating area membrane. The whole sensor housing is attached to
the skin by means of a removable two sided adhesive foam donut pad. In
order that the sensor be reusable, this foam pad can be removed and a new
pad applied, thus insuring proper adhesion to each patient.
FIG. 1 shows the preferred sensor spacing and configuration. The spacing
between the Infrared LED 10 and the Red LED 11 is not extremely critical,
but the spacing between the Infrared/Red LED's and the phototransistor 12
is quite critical.
This spacing is as detailed on the drawing. The configuration is also to
some degree critical. As mentioned above, the arrangement has been
successfully employed with three sensors in a straight line, or in a
triangular pattern. Various arrangements have produced limited good
results, but the illustrated arrangement detailed on the drawing has
produced the best, desired results. The drawing shows the wire connections
for the sensors. The cathodes of the Red and the IR and the emitter of the
Phototransistor are all tied to common. The anode of the Red, the anode of
the IR, and the collector of the Phototransistor are brought back to the
circuit separately. The RED and the IR are connected to their appropriate
drivers and the collector of the Phototransistor is tied to the input
operational amplifier and sample and hold circuit.
The signals need to be shielded from the RF and hence a ground plane over
the conductors is required. An outer shield is provided over two of the
conductors to help isolate them from the RF and the noise generated from
the two LED's turning on and off.
The sensor has approximately 18" of cable 13 with a shielded plug attached
to the end of the cable using a technique called insulation displacement.
This plug plugs into a jack or the extension cable. The extension cable is
designed to encase the plug with the exception of the very front chrome
surface. The housing is molded of thermal plastic, such as a polyethylene,
polycarbonate, pellathane or a styrene.
At the other end of the extension cable is another plug. This latter end
plugs into the front of the machine. All of the plugs and jacks used are
grounded from noise and RF. The pc board was designed and fabricated to
allow the wires to mount to the jack, and also to provide shielding.
FIG. 2 illustrates the sensor housing utitlizing the hard concept. The
sensors are allowed to float on a silicon rubber gasket type membrane.
Referring to FIG. 2 an adhesive pad 14 is designed to be removed and a new
one applied for each patient. The pad is approximately 0.062 thick and
shaped like a donut.
The adhesive pad 14 has a carrier on both sides of the foam to aid in the
adhesion and removal process. This carrier may be a very thin polyethylene
material. This allows the adhesive to be removed from the patient and the
sensor without tearing the pad. The pad is aligned and registered on the
sensor with the use of a small raised guide tab. This small tab is
approximately 1/2 the thickness of the pad and will force proper
positioning of the pad without error. If the pad were to be applied
nonconcentric, a portion of the sensor side of the pad could interfere
with the movement of the floating diaphragm. This tab may be increased in
height to just slightly higher than the pad thickness. This forms a light
barrier around the sensing area of the skin to help block ambient light
from reaching the phototransistor and causing potentially erroneous
readings. The adhesive has a protective layer both top and bottom that
will be peeled away when the pad is applied to the sensor and the patient.
This protective layer is preferably split in the center to allow easy
removal.
Reference numeral 15 is the sensor (LED's/phototransistor) mounting frame
or carrier. The three sensor elements are mounted from the back side, the
small dome lens thereof pushing through appropriate circular holes in the
frame. They are mounted using an adhesive, such as UV cure epoxy. This LED
carrier 15 provides many functions. The carrier is molded out of a dark,
light blocking thermal plastic. The light cannot be permitted to enter the
phototransistor except through the small domed lens area. The carrier will
provides exact and proper positioning of the three sensors, and provides a
proper test bed for testing the assembly, etc. The small domes protrude
through the top of the carrier approximately 0.010. This will cause the
sensor domes to always be in contact with the skin just slightly more than
the whole top carrier piece. To increase their effectiveness, and to
maximize their power and detection abilities, a silver reflective paint is
applied on the outside of the RED, IR, and photo-transistor bodies. This
paint covers all areas without interfering with the domed lens area. With
the LEDs (RED/IR) this causes any and all emitted energy to be sent out
through the dome area and not through the sides and bottom. On the
phototransistor, it does two things. First, it helps block any ambient
light, or light emitted from the RED and IR leds, and second, it helps
capture all received light onto the die. On the RED and IR the paint also
helps to block light from the phototransistor.
The silicon rubber gasket 16 is one of the key elements of the design. The
silicon rubber gasket provides the following functions.
1. It provides a moisture barrier/dirt/contamination barrier from the
outside to the inside.
2. The part allows the LED carrier to "float". This floating action will
help insure proper skin contact at all times. If the housing is moved
slightly, the inner carrier will float keeping the sensors in contact with
the skin and eliminating a large amount of movement artifact.
3. Being of silicon rubber, it can easily be cleaned with normal hospital
cleaning solutions. The silicon part is attached to the LED carrier in the
manner of of a rubber band. It is stretched slightly, slipped over the top
or bottom of the carrier 15 and put into place. The design is such that
the silicon rubber will properly hold the carrier in place and provide all
the required functions.
4. The design of the part is such as to afford a very small constant,
linear force to the skin. This force is approximately 10 grams. The part,
as the carrier is pushed down, will roll downward. This will continue to
roll with a consistent linear motion for well over 0.100. Calculated max
movement would be approximately 0.050 under normal use conditions.
5. This part is designed to apply a specific amount of preload to the
carrier. This preload is set at approximately 0.020. This will force the
sensor to always be in contact with the skin.
6. The rounded edge of the part is a ball 17 which allows easily an
entrapment technique.
The top housing member 18 is designed to afford proper alignment of the
pad, a stable surface for the pad to contact a light shield from outside
ambient light, and half of the entrapment technique for the silicon rubber
part 16.
The bottom 19 of the housing is nice and rounded to give a soft appealing
look (See FIG. 3). The bottom housing also provides the other half of the
entrapment for the silicon. The cord or cable 13 is mounted through a
small extended opening off the bottom of the housing. The cable is bonded
into the bottom housing with some epoxy based technique. The top and the
bottom housing parts are connected using a solvent bonding operation.
The flex strip 20, having circuit traces for the sensor elements is
attached to the LEDs/phototransistor providing the proper connections and
is routed down and then soldered to the cable. The flex strip must be very
flexible and not contribute or alter the linear movement of the LED
carrier. Hence, the flex may be 0.001 kapton, with 1 oz copper trace and
0.001 kapton. Flexibility would be similar to that of one or two sheets of
paper.
It has been found that all LED's and photodetectors do not operate
satisfactorily in the environment of the invention. Specific devices that
have the necessary saturation, color and intensity characteristics for
correct operation of the system of the invention herein disclosed are
BN501 (infrared LED), BR2262 (red LED) and PS502 (phototransistor), all
products of Stanley Optoelectronics of Japan.
Referring now to FIG. 4, therein is illustrated a block diagram of the
oximeter aparatus in accordance with the invention. The oximeter is
controlled by a microcontroller 30 coupled via the data bus 31 to a
program memory 32. The microcontroller 30 is connected to a serial bus 33,
to apply digital signals to the digital to analog converter 34 and receive
signals from the analog to digital converter 35.
The digital to analog converter 34 applies a brightness control voltage to
voltage control circuits 40, 41 for the red and infrared LEDs, the control
circuits 40, 41 applying operating current to the red and infrared
emitters 11, 10 by way of electronic switches 42, 43, respectively. The
switches 40, 42 are controlled by a timing circuit 44 which also controls
the operation of red and infrared sample and hold circuits 45, 46,
respectively, so that the output of the sample and hold circuit 45 occurs
synchronously with the turning on of the red emitter switch 42, and the
output of the infrared sample and hold circuit 46 occurs synchronously
with the turning on of the infrared emitter switch 43.
The output of the phototransistor 12 is applied by way of the current
amplifier 47 to the sample and hold circuits 45, 46, and thence to band
pass filter and gain circuits 50, 51, respectively and thence to
controlled gain circuits 52, 53, respectively. The outputs of the gain
circuits, which may include four to one selectors 54, 55, is applied to
the analog to digital converter 35, and thence on to the serial bus. The
four to one selectors or multiplexers 54, 55 are controlled by the
microcontroller 30, to select the desired gain level from multiple gain
amplifiers 56, 57, respectively.
The outputs of the sample and hold circuits 45, 46 are applied as DC level
feed-backs to the voltage control circuits 40, 41, respectively, as well
as to the analog to digital converter 35 to enable monitoring of this DC
level by the microcontroller. The band pass filter and gain circuits 50,
51 are controlled by the filter response control line from the
microcontroller, as is the response of the control circuits 40, 41. The
circuits is further provided with a suitable alarm and beeping circuit 60,
random access memory 61, and a display and keypad circuits 62.
On the left-hand side of FIG. 6 is the 120 volt A.C. input to an EMI
filter, thence through a transformer, to the constant voltage constant
current charger CVCC which maintains the 12 volt battery at a full charge
during its usage, as long as it's plugged in. As soon as the wall current
is lost, battery backup enables continued operation for approximately two
hours, or up to eleven hours when LCD displays are employed.
The 12 volt battery 100 is coupled via a power switch 101 to three voltage
regulators. The supply 105 outputs a voltage ABIASto provide a
pseudo-ground for operating the op amps. This output is four volts. The
LM317 supply 106 is connected as an 8 volt regulator and supplies the
upper rails for all of the analog circuitry. The LM340T supply 109
supplies the microprocess or board with 5 volts.
FIG. 7 shows all of the circuitry necessary to control the tone change for
the speaker or a saturation change, and to control how long of a beep to
have for the pulse. This is done with a 556. It also includes a
multivibrator to generate the tone for the alarm. The part 4046, U10, is a
voltage-controlled oscillator to produce the tone desired for the
saturation value. The controlling voltage is DAC Saturation which comes
from the DAC of the microcontroller board. The pulse is controlled by
signal PA6 that goes to the trigger of U8A to give about a 20 millisecond
pulse output to the 4066 switch U9D. U9D acts an an analog switch to
control whether the signal is present or not. After being supplied with AC
bias the signal goes to a 3340 connected as a voltage controlled
attenuator 110. The control voltage for the 3340 is from line DAC 3 and,
along with the output from the microcontroller via U9D, control is
provided for the magnitude of the signal, plus controlling the volume.
U8A includes a 556 connected in a multivibrator mode to generate a tone.
That tone is controlled by the 4066, U9A. The signal PA7 gates the 4066 to
control whether the alarm tone is on or not. The software controls when
the alarm is on. The beep signal is not mixed with it, so this circuit
makes sure there is only one signal there. The alarm signal is applied to
a 3340, 111, to enable control of the volume of the alarm with the signal
called DAC 2. This is a buffered output from the DAC on the
microcontroller board. Both of these signals are fed into an LM386 audio
amplifier U7. The LM386 is operated by the supply voltage VBAT and drives
the speaker through AC coupling via C23.
FIG. 7 is a diagram of a circuit for a voltage output to enable
verification of the functionality of the unit and also to give an output
proportional to saturation, a pulse wave form, for clinical research. The
output goes to a DB25 connector on the back panel. The unit may have three
analog outputs, one for pulse, one for saturation, and one for the wave
form, or variation as desired.
FIG. 8 shows self-explanatory connections between the two 44 pin connectors
in the circuit.
FIG. 9 shows the analog circuits 44 concerned with the timing generation
for turning on and off the LEDs and turning on and off the sample and hold
at the appropriate time.
The circuit of FIG. 10 is concerned with controlling the voltages applied
to the IR and the red LEDs. DAC 1 is a signal from the connector bus, and
ranges from 0 to 5 volts. This signal controls the level of brightness of
the LEDs. It is amplified in U9C to give a signal that ranges from 0 to 8
volts to provide a little gain. The signal output of U9C is applied via a
serial 200 k resistor to the non-inverting input of U7A and also via a 200
K resistor to the non-inverting input of U7B. This is part of a feedback
loop and it forms an integrator which allows control of the brightness
The other part of the feedback loop is feedback IR, which is the signal
FBIR. This signal is applied via a 1 meg resistor, or alternatively via a
resistor, to the inverting input of U7A, or into Pin 6 of U7B. When
considering FBRED, the feedback red signal is similarly applied via a 1
meg resistor or alternatively, via a 10 k resistor, to the inverting input
of U7B, U7A and U7B are connected as integrators with 1 mfd mylar
capacitors C12, C9 forming the integrating part of these circuits. The
time constant of the feedback loops can be changed by activating the 4066
switches U8A, U8B and paralleling the 1 meg resistors R85, R86 with the 10
k resistors R7, R15. This sets the integrator to a much faster mode for
faster signal acquisition. This mode control is effected by the signal
PA3, which is controlled by the microcontroller and permits changing the
time constants, not only in the feedback loop, but as will be apparent,
also in the bandpass filters .
The output signals of U7A and U7B are directed via voltage followers U7D,
U7C, to the V-adjust input of LM317 regulators U2, U1. These circuits
perform voltage control functions to produce regulated output voltages,
controlled by the V-adjust inputs, with an upper limit of that is about
1.2 volts below VBATSW. R9 is a load to maintain the proper current output
of the regulator, and smoothes the output. R8, R17 39 are current limiting
resistors. Q1 and Q2 are switches to turn on the signals to the LEDs. The
diodes D1, D2, D4, D5, D7, and D8 protect the FET's from static
electricity or discharges that might occur in that area. The timing
signals will be discussed in the following paragraph.
P1, pin 2 applies the signal to the IR emitter, and P1, pin 1 applies the
signal to the red emitter. This is a voltage controlled signal.
Referring now to FIG. 11 the photodetector input to the circuit is applied
to P1, pin 4. This is the collector side of the phototransistor. Pin 3 is
the isolated ground on emitter side of the phototransitor. R70 is a
control potentiometer for the gain of the current to voltage convertor,
formed by the Opamp U20A. R68 is a current limiting resistor to protect
U20A from any inputs which may exceed the rails potential. The signal
output of U20A are applied to a red sample and hold circuit 45 and an IR
sample and hold circuit 46. The sample and hold circuits include the 4066
analog gates U17B, U17D, and capacitors C51, C49. The signals now are
essentialy in the form of an AC signal riding on top of the DC. U20B and
U20C are voltage followers to eliminate any loading of C51 or C49.
The signal output of U20B, U20C is essentially a DC signal with a small AC
component, and that DC signal is fed back as the above-discussed FBIR or
FBRED. These signals are also fed through a voltage divider back to an A
to D convertor on the microcontroller board. This enbables checking to
make sure that the DC levels are properly matched. It's crucial to the
system to maintain a match between the Red and the IR signals in so far as
their DC components are concerned. Becaue of thi, feedback is employed
which maintains that match contant independently of any degrading of the
parts themselves or any minor fluctuations that might occur in contacting
the skin, etc.
The signals are now AC coupled via C40 and C39 to the first stages of
bandpass filters, U21A, U21B. The feedpack paths of the filters, which
include 1.8 meg resistors, must be shorted by analog switches U17 under
the control of the PA3 signal. This makes it a lowpass filter and allows
compensation for the DC component. PA3 from the microcontroller board, is
the same signal that is used to control the signals from the feedback loop
constant for faster compensation.
After filtering in U21A and B, the signals pass through very large
polyester capacitors C41, C44 and serial resistors R74, R82 to a bandpass
stage formed by U21D and U21C. After filtering in these stages, the
signals are essentially AC signals. These signals are applied to LTC1062
circuits U19, U8 which are 5 pole lowpass filters, to eliminate any high
frequency components. Since the LTC1062 circuits cannot drive substantial
capacitive loads, their outputs are applied to a voltage followers formed
by U16A and U16B.
Resistor R74 which is a 100 k potentiometer. This resistor is used to
adjust the two stages U21C, U21D so that the levels of the IR stage, at
the top of the figure, and the red stage at the lower portion of the
figure, can be essentially matched. This matching is essential since the
two bandpass filters, all of the filtering, all of the gain on our AC
signal, must be identical in the two circuits. This matching may
alternatively be effected in the microcontroller, to render adjustment of
R74 unncessary, so that the unit can be calibrated by running a test
program with a clibration header connected thereo.
The signal outputs CNT1, CNT2 are applied to the circuit of FIG. 12. This
circuit includes the bandpass filters, Linear Technologies LTC1060T--U12
and U15. These are 2-poled band or highpass/lowpass filters configured to
have a gain of 56 over 39, in mode 3. The Q is adjusted by R38, R47 and
adjusts the center frequency. The center frequency was set to about 102
beats per minute. The filter is used to just provide a few more poles on
the high and the low end.
The Tbus signal applied to the filters is a signal which is at 256 hz that
clocks these switched capacitor filters. The signal outputs of the filters
U12, U15 are applied to a further LTC1060 filter U10. Once the signals
have passed through these three filter stages, they exit the circuit as
GAIN1 and GAIN2. U10 may have a gain of 1, and a Q of 1.
Referring now to FIG. 13 of the analog circuit, the RED or the IR signal
pass through voltage followers U16D, U9D, and then are each applied to
four stages of amplification, each with a different feedback resistor, so
each gain is different. These gains in one example were 1, 2, 4 and 8, and
in another example they are about 1, 1.25, 2, and 4. The outputs of these
amplifiers U11, U13 are fed to 4052, i.e. a dual 4 to 1 selector, U14A, B.
The selector is controlled by signals PB4 and PB5 from the microcontroller
board. In this system, the 4 by 1 selectors are used to enables elections
of any desired gain signal. While an adjustable gain system may provide
the necessary adjustment function, such systems generally have too long a
setting time. The present system, however, aquires the correct signals
almost instantly. In the illustrated system, the 4 channels operate
simultaneously and the desired one is picked using a 4 to 1 selector. The
4 to 1 selector also controls the RED signal. Both of these | | |
