 |
Description  |
|
|
FIELD OF THE INVENTION
This invention relates to automated blood pressure monitors employing the
oscillometric method of detection, and more particularly to artifact
rejection and analysis methods which improve the overall timing response
of the unit.
BACKGROUND OF THE INVENTION
The oscillometric technique for blood pressure measurement was developed
long ago, but has only found extensive utility more recently, with the
advent of inexpensive integrated circuits and even less expensive
microprocessor controlled automated blood pressure monitoring systems.
Thus, while indirect, manual oscillometric measurement is described in The
Southwestern Veterinarian, Vol. 23, Summer of 1970, No. 4, pp. 289-294,
practical oscillometry tends to be found in numerous commercial systems
wherein a dedicated microprocessor with stored program control operates by
controlling cuff pressure, sensing pressure fluctuations caused by heart
beats, and performing the requisite calculations which yield heart rate,
mean arterial pressure, and systolic and diastolic pressures. The advent
of very large-scale integration portends a change from software systems
back to hard wired systems, wherein an entire specially configured machine
circuit may be included on but a single chip, but in all events such
increasing sophistication and decreasing cost provides further supports
for the increasing popularity of automated blood pressure monitors.
Of particular interest with respect to the principles of the present
invention are the concepts set forth in U.S. Pat. Nos. 4,360,029 and
4,394,034 to M. Ramsey, III, which are commonly assigned with the instant
invention. The Ramsey patents derive from common parentage, the former
including apparatus claims and the latter including method claims, their
division having been made in response to a restriction requirement during
the prosecution. Both patents, however, carry common disclosures of
apparatus and methods for artifact rejection in oscillometric systems,
which have been in practice in the commercially successful DINAMAP* brand
monitors, which are manufactured and marketed by Critikon, Inc., of Tampa,
FL, the assignee hereof. In accordance with the Ramsey patents, an
inflatable cuff is suitably located on the limb of a patient, and is
pumped up to a predetermined pressure. Thereupon, the cuff pressure is
reduced in predetermined decrements, at each level of which pressure
fluctuations are monitored. These typically consist of a DC voltage with a
small superimposed variational component caused by arterial blood pressure
pulsations (referred to herein as "oscillatory complexes"). Therefore,
after suitable filtering to reject the DC component and to provide
amplification, pulse peak amplitudes above a given baseline are measured
and stored. As the decrementing continues, the peak amplitudes will
normally increase from a lower amount to a relative maximum, and
thereafter will decrease. The lowest cuff pressure at which the
oscillations have a maximum peak value is representative of mean arterial
pressure. Systolic and diastolic pressures may be evaluated either as
predetermined fractions of mean arterial pressure, or by more
sophisticated methods of direct processing of the oscillatory complexes.
The Ramsey patents devote considerable effort and disclosure to the
rejection of artifact data and hence to the derivation of accurate blood
pressure data. Indeed, as is apparent from FIG. 2 of the Ramsey patents,
the most insubstantial portion of the measurement cycle (denominated "T3")
is devoted to the execution of complex detection at the various pressure
levels, measurement of signal peaks of true complexes, and processing
those peaks in accordance with artifact rejection algorithms so in
sequence to identify the pressure level at which the peaks are a relative
maximum, that is, mean arterial pressure. As taught by the Ramsey patents,
about thirty seconds of a total cycle of less than 40 seconds are devoted
to these "T3" functions.
It is a primary object of the principles of the present invention to reduce
the time required to proceed through the pressure decrementing steps,
correspondingly to provide data more rapidly to the administering
physician, and to increase patient comfort by speeding the deflation and
decreasing the time at which arterial occlusion takes place.
Corespondingly, such improvements in speed and comfort should be had
without penalty to overall accuracy.
It is a subsidiary but related object to make such improvements to systems
of the class described in the Ramsey patents, with but minimal hardware or
software variation, thereby to provide data and performance which is
consistent with the considerable mass of clinical data already available
for such systems.
SUMMARY OF THE INVENTION
The principles of the present invention are premised on the proposition
that full scale artifact rejection criteria should be applied at the
beginning of each inflation--deflation cycle, but that after a
predetermined succession of "true" complexes are identified, abbreviated
artifact rejection criteria may be utilized. In particular, in accordance
with the principles of the present invention, at early pressure decrements
in the cycle, as per the teachings of the Ramsey patents at least two
complexes per level are detected, their respective peaks are identified
and compared with one another, and they are identified as "true" complexes
if they are within a specified range of one another, for example 20%. In
accordance with the principles of the present invention, if such true
complexes are identified in this fashion for a certain number (e.g. 2 or
3) of successive levels, abbreviated criteria may be adopted, for example
at subsequent decrement levels simply detecting a single complex, and
identifying its peak as a "true" peak at that level simply if it is in
relative time correspondence with previously detected complexes (i.e.
within a "heart rate window"), and is within a specified amplitude range
of a next previous peak (e.g. 50%). If these heart rate timing and
amplitude criteria are not met, however, the system automatically reverts
to the more detailed artifact rejection schemes, involving detection of
multiple complexes at each pressure decrement level. As desired, the
principles of the present invention may be applied together with other,
more sophisticated artifact rejection methods as taught by Ramsey, for
example the time rate of change criterion which in accordance with the
Ramsey patents is applicable simultaneously with use of amplitude criteria
as described above. Through application of the principles of the present
invention, the cycle time may be reduced by the amount of time saved at
each level at which one, rather than more than one, complex is determined
prior to the next decrementation of cuff pressure. In practical systems,
this time savings may be on the order of 5-10 seconds per cycle, which
otherwise may have been in the range of 25-30 seconds.
DESCRIPTION OF THE DRAWINGS
FIG. 1 replicates FIG. 6 of the cited Ramsey patents, which in turn
discloses a block diagram of the peak-to-peak calculator and artifact
rejection circuits of those patents.
FIG. 2 shows in block diagramatic form a method embodying the principles of
the present invention, the procedures therein being particularly amenable
to software implementation.
FIG. 3 shows an illustrative modification of the FIG. 1 apparatus in
accordance with the principles of the present invention.
BEST MODE FOR CARRYING OUT THE INVENTION
U.S. Pat. Nos. 4,360,029 and 4,349,034 to M. Ramsey III are incorporated by
reference herein. In particular, the principles of the present invention
may be incorporated with relatively minimum difficulty into systems and
methods set forth in the Ramsey patents.
It is worthy of note that the Ramsey patents are configured in terms of
apparatus embodiments, the functional blocks of which are each set forth
in extensive detail, including function. It is to be understood that the
disclosure of the Ramsey patents is sufficiently precise to be made and
used without undue experimentation either in the form of a dedicated hard
wired system, in the form of a suitably programmed microprocessor based
system, or as a hybrid of both. Indeed, widespread availability of
relatively low-cost microprocessor and associated memory apparatus has
made that embodiment the implementation of choice. Further, as of the date
of filing hereof, software implementation is the best mode contemplated by
the inventors hereof. If is to be understood, however, that such
implementation should no way be taken as limiting, and in the very near
future, the availability of low-cost specially designed very large scale
integration (VLSI) may once more make hard wired systems the embodiment of
choice. For emphasis, it deserves restatement that the disclosures set
forth herein, together with those in the referenced Ramsey patents, are
sufficient under the statute for either mode of embodiment.
Reference is first had to FIG. 1 (FIG. 6 of the Ramsey patents), which
describes a relevant subsystem of the overall system and which is
described in considerable detail in Cols. 8-10 of the Ramsey patent. In
turn, numerous functional blocks of the subsystem of FIG. 1 are configured
disclosed in greater detail in the Ramsey patents. It is to be noted that
even in accrdance with the principles of the present invention, at the
initiation of each measurement cycle, the self-same procedures as taught
by Ramsey may be executed, until a certain number of true complexes have
been identified (e.g. at 2 or 3 successive decrement levels), whereupon
the more simplified, abbreviated artifact rejection criteria of the
present invention are employed. Thus, in accordance with the principles of
the present invention, the cuff is applied to the patient and inflated to
a predetermined start level, and at each such level the heart pulse signal
p(i) is calculated and delivered to a peak-to-peak detector. The
peak-to-peak detector 100 operates on the envelope of the oscillatory
complex, and at each such cuff pressure level, respective first, PP(a),
and second, PP(b) peaks are detected, and are coupled to an amplitude
check circuit 102. The amplitude check circuit 102 determines whether the
successive such peaks are within a certain range of one another, for
example within 20% of one another. If they are, the complexes are
designated as being "true" complexes, the amplitude check is satisfactory,
and a logical 1 output signal is generated. If not, a logical 0 output
signal is generated. It will be appreciated that the amplitude check
circuit 102 is effective to detect pressure artifacts such as accidental
striking of the cuff in synchronism with arterial pulses.
The purpose of the 20% tolerance requirement is to reject those beats which
are widely variable in amplitude from beat to beat such as could be caused
by premature ventricular contractions. Not only does this eliminate the
acceptance of variable data during heartbeat irregularities, but it is
also quite useful for reducing the influence of pulsation artifacts caused
by subject motion or outside interference such as physician bumping
against the cuff as could occur in an operation.
It will be noted that simultaneously a slope check 105, a period check 106,
and a rise time check 107 are being conducted. These optional further
criteria, together with the envelope check 103, provide a joint and
accurate artifact rejection scheme. In all events, however, at this stage
of the processing, successive complexes are detected at each pressure
level and the slope check, period check, rise time check, and the like are
subservient to the basic amplitude discrimination from the amplitude check
102 and the envelope check 103. In accordance with the principles of the
present invention, once these checks have "passed" for two or three
successive pressure levels, as desired, it is not necessary to continue to
require the redundant complex amplitude checks, but rather simply to
utilize a decrement to decrement envelope check, together with a period
check such as set forth in 106.
The operation of the principles of the present invention in conjunction
with the apparatus of FIG. 6, as well as the balance of the apparatus of
the referenced Ramsey patents, will be better appreciated upon
consideration of FIG. 2. FIG. 2 will be recognized as a flow chart from
which the principles of the present invention are to be executed, which
flow chart may be taken directly to code, in the case of a software style
system, or which may be implemented as shown in FIG. 3 as modifications to
the hardwire system of FIG. 1. The procedures of FIG. 2 assume the rest of
the system operating as described, and therefore simply highlight the
modifications needed for incorporation of the principles of the present
invention.
An initial inflate cuff step 201 assumes all of the preparatory T1 and T2
steps of the Ramsey patents, and the initiation of the T3 steps.
Additionally, a cuff pressure level flag marker "n" is set at 1, and a
post-man iteration marker "X" is set at 0. Thereupon, at 202 and 203,
successive complexes are detected, and their peak levels are evaluated. In
each case, a suitable procedure is disclosed in the Ramsey patents.
Thereupon, also as disclosed in the Ramsey patents, the successive peaks
are compared with one another, to determine whether the second varies by
less than 20% from the first. If not, no "true" complex has been detected,
and further data must be gathered. Assuming the peaks do compare
favorably, that is that the amplitude criteria are met, the "yes" branch
is taken from comparison step 204, and, as desired, other artifact
criteria may be employed, as represented by step 205. These will be
recognized as including, as desired, one or more of a slope check, period
check, rise time check, envelope check, and the like. Assuming these
artifact criteria are not met, the "fail" step is taken, to recycle and
search for correct data in similar instance to the "no" exit from compare
step 204. Suitable iteration abort criteria are applied at 212 to insure
that the system does not get stuck in an endless loop.
Assuming that the other artifact criteria at 205 are met, that is that all
artifact criteria are met, the "pass" branch is followed from step 205 to
step 206, at which the associated of value PP(n) is calculated stored,
that being recognized as the average of PP(a) and PP(b), pursuant to the
teachings of the Ramsey patents. The flag marker n is incremented by 1,
and at 207 the cuff pressure is decremented.
To this point, the operation has been more or less standard in the fashion
of the Ramsey patents. At the next step, however, comparison 208, the
procedure determines how many prior pressure decrement levels in
succession have yielded true complexes and stored valid average peaks. In
the method shown, two such steps are adequate to follow the "yes" branch
(and thereby to initiate the abbreviated procedures in accordance with the
principles of the present invention), but it is understood that a
different number of levels, for example three or more, may also be
utilized. Assuming that two such values are not in storage, and that the
principles of the present invention are not yet ripe for application, the
"no" exit is followed from 208, and the whole cycle is to be repeated for
the new pressure level subject always to the iteration abort criteria.
If the "yes" exit is followed from step 208, the operation at step 209
refers generally to the execution of those procedures necessary for
evaluating mean arterial pressure. That is, the mean arterial pressure
procedures are completed to determine the pressure at which the peaks are
a relative maximum. Optionally, these may be executed exactly as set forth
in the Ramsey patent. Other modes of doing so will of course be equally
appropriate.
In any event, the next step, 210, determines whether mean arterial pressure
has been evaluated. If so, the "yes" branch is followed. Assuming not,
however, the "no" branch exits from step 210, and the principles of the
present invention are called into play. As shown at step 213, a single
complex detection and peak determination is conducted at the new level,
but rather than the more complex averaging and comparison as shown at 204,
the peak is assigned the PP(n+1) value. At 214, this value is compared
with the average valid peak from the previous level, to determine whether
or not the variation is within specific acceptable limits, in the example
shown, 50%. If not, indicating that the principles of the present
invention will not be acceptable to continue defining peaks, the "no"
branch is followed, back up to the initiation of the entire process from
step 202. If so, however, the "yes" branch is followed, back up to the
optional "other artifact criteria" step at 205. For example, in preferred
embodiments of the principles of the present invention, a period check
will be adequate further criterion to distinguish the data from artifacts.
Assuming all is well, the loop continues, and will continue to do so
through steps 206, 207, 208, 209, 210, 213, 214 and so on, until mean
arterial pressure has been determined at 210, whereupon the yes exit is
followed.
Once mean arterial pressure is so determined, it is desirable to check a
predetermined further number of levels in order to assist in the diastolic
pressure evaluation. As earlier stated, the flag variable "X" was set in
order to allow for this check. From the yes branch of decision box 210,
the routine moves to another decision box 216, determining whether X
equals N, the number of levels at which complexes are to be checked past
means arterial pressure. If fewer than "N" loops have been accomplished
(i.e. fewer than the predetermined number of levels past means have been
investigated), the no branch is followed from steps 216, the "X" variable
is incremented by one at 217, and the routine returns to steps 213 and 214
in order to provide the peak measurement functions for each such level.
Once the "N" levels past means have been checked, the routine flows from
decision box 216 via the yes branch, thereby to complete the routine. As
shown at box 211, the actual systolic and diastolic computation routine
may be conducted. In the unlikely event that means arterial pressure is
the only data desired, the routine is at an end.
It will be appreciated from the foregoing that, in the case of software
embodiments of the Ramsey methods, but minimal software changes in
accordance with FIG. 2 will be required, and that such changes are well
within the ability of those of ordinary skill. In the event, however, that
one wishes to modify the hardwire version as shown in the Ramsey patent,
recourse is had to FIG. 3 as follows. In particular, FIG. 3 is configured
similarly to FIG. 1, except that a few additions have been provided in
order to accommodate the principles of the present invention. These
additions are represented by a switch 301, an or-gate 302, and a "good
pulse counter" 303. Operation of these elements in conjunction with the
rationale of the present invention as discussed with FIG. 2 will result in
gainful operation of the principles of the present invention in hard wired
form.
As in FIG. 1, signal flow at the input is provided in parallel fashion to a
peak to peak detector 308, a derivative calculation 310, a period check
element 313, and a rise time check 314. The latter three elements are
coupled, as in the FIG. 1 embodiment and in the aforementioned Ramsey
patent, to an and-gate 307. Another input to the and-gate 307 is from a
slope check 312 via the derivative calculation 310, again in accordance
with the prior art.
The signal from the peak to peak detector is coupled to an averager (e.g.
integrator) 309, an amplitude check 311, and to one pole 318 of the bypass
switch 301. The other pole 317 of the bypass switch 301 is coupled to the
output of the averager 309. Therefore, depending upon the position of the
switch 301 (under control of the "good pulse counter" 303) the envelope
check circuit will receive either the output of the averager 309, or the
peak signal directly from 308.
In the FIG. 3 embodiment, the peak signal from 308 through the amplitude
check 311 is coupled directly to one input of the or-gate 302, the other
input of which is the output of the good pulse counter 303. In turn, the
envelope check signal 316 and the output 306 of the or-gate 302 are also
coupled to the and-gate 307. It is the output of the and-gate 307 which
enables the good pulse counter to generate a logical 1 at its output, and
thence to control the bypass switch 301 as well as the or-gate 302. The
counter 303 itself is a simple gated counter which, upon receipt of two
successive logical 1's from the and-gate 307, generates a logical 1 at its
output thereby both to energize the or-gate 302, and to switch the bypass
switch 317 to defeat the amplitude test after two good pulses have been
received. Similarly, a logical 1 from the counter 303, after two good
pulses have been received, cause the bypass switch to be moved to the
lower position 318 and thus also to bypass the averager 309.
In operation, at the initiation of each cycle the switch 301 is at the
uppermost position 317, thereby including the averager 309 in the
processing. The counter 303, having been reset automatically by the delay
unit, has its output at a logical zero. Since the amplitude check 311 has
not yet yielded any favorable indication, the output 306 of the or-gate
302 is also at a logical zero state, and the output of and-gate 307 is
similarly at a logical zero. Incoming signals are therefore processed in
the same fashion as in the FIG. 1 apparatus. At some point, however, all
of the checks, including the peak to peak and envelope check, the
amplitude check, the slope check, the period check, the rise time check,
and the cuff pressure check are satisfied (i.e. all outputs are logical
1's, which are conveyed to inputs of the and-gate 307), and the system
output as well as the input to the counter 303 goes to a logical 1. If and
only if the next consecutive complex also satisfies all checks, and yet
another logical 1 is presented thereby to the counter 303, the output of
the counter 303 goes to a logical 1 and the switch 301 is moved to its
bypass mode, that is closed to the lowermost terminal 318. For the next
sequence of signals, so long as the various other checks are satisfied,
the averaging routine will be obviated. When and if the balance of the
checks are not satisfied, or in the alternative the passage of the delay
at 325 (which upon operation of inverter 324 commences with the first such
logical 1), straight detection is permitted.
It will be appreciated that, in accordance with the embodiment of FIG. 3,
some or all of the various slope checks, period checks, rise time checks,
and the like may either be dispensed with in accordance with the needs of
the designer, or enhanced and made further complicated in accordance
therewith. In either event, just as in the case of the prior art, those
checks form different and sometimes useful artifact rejection criteria
which may in some instances be useful in conjunction with the principles
of the present invention, and in others may not. In either event, the
principles of the present invention are deemed to have independent
utility. Further, it will be appreciated that the principles of the
present invention themselves are as characterized by the spirit and scope
of the claims appended hereto, and that numerous alternative embodiments
will occur to those of ordinary skill without departure from the spirit or
scope of the principles of the present invention.
* * * * *
|
|
 |
|
 |
Description |
|
