Electrical computers: arithmetic processing and calculating – Electrical digital calculating computer – Particular function performed
Reexamination Certificate
1998-12-08
2001-06-19
Mai, Tan V. (Department: 2121)
Electrical computers: arithmetic processing and calculating
Electrical digital calculating computer
Particular function performed
Reexamination Certificate
active
06249796
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a technique for reducing the settling time of a high pass filter caused by sharp transitions in the filter input signal, and more particularly to a technique for quickly and accurately detecting a pulsatile signal component in a pressure signal produced by an automatic blood pressure gauge.
2. Description of the Prior Art
As noted above, the present invention is generally applicable to reducing the settling time of high pass filters used for detecting a low level pulsatile signal in the presence of a sharp transition in the input signal. However, for illustrative purposes the invention will now be described in a specific apparatus, namely, a non-invasive blood pressure (NIBP) measuring device. It should be kept in mind, however, that the invention is not limited to such a particular apparatus or use.
FIG. 1
illustrates a conventional automatic blood pressure gauge, which includes a resilient inflatable cuff
2
connected via a tube
4
to an electric pump
6
. The cuff is placed about the arm of a patient (not shown) and the pump is controlled by a microprocessor
8
to inflate the cuff with a fluid, such as air, to a preset pressure. A deflation valve
9
is also connected to tube
4
and controlled by microprocessor
8
to deflate the cuff during the blood pressure measuring process. In addition, a pressure transducer
10
is connected to the cuff
2
via a tube
12
, which measures instantaneous air pressure levels in the cuff. The electrical pressure signal produced by the transducer is applied to an analog to digital (A/D) converter
14
for digitization and then to microprocessor
8
for analysis to determine the instantaneous pressure of the cuff as well as for detecting the blood pressure pulses of the patient. Analysis of the pressure signal generally comprises band-pass filtering and processing by the microprocessor to detect the pulsatile component caused by the beating of the patient's heart, to produce values representing mean, systolic and diastolic blood pressure measurements of the patient. These values are then conveyed to the user of the NIBP device using, for example, a display
16
In operation, the cuff is affixed to the upper arm area (or other extremity) of the patient and then inflated to a pressure greater than the suspected systolic pressure, for example, 150 to 200 millimeters of mercury (mmHg). This pressure level collapses the main artery in the arm, effectively stopping any blood flow to the lower arm. Next, the cuff is slowly deflated and the electrical signal provided by the pressure transducer is analyzed to detect cuff pressure variations caused by the blood pressure pulses of the patient, where the pulses are mechanically coupled to the cuff. In general, the pulse component of the pressure signal has a relatively low amplitude, on the order of one percent of the total pressure signal, and is therefore somewhat difficult to quickly and accurately detect and measure.
Proper operation of the automatic blood gauge require s prompt and accurate detection and measurement of the low level pulsatile component of the cuff pressure signal. The microprocessor in the automatic blood pressure gauge typically executes an algorithm employing a high-pass filter (HPF) to extract from the average cuff pressure signal, the heart-rate pressure pulsations coupled to the cuff by the blood vessels in the patient's arm. In the embodiment illustrated herein, the filter is a 4th order Bessel filter with a −3 dB cut-off frequency of 0.75 Hz.
There are several points during the automatic measurement process where a sharp transition occurs in the average slope of the cuff pressure signal. These include the transition from the end of cuff inflation to the pressure-hold stage oust before the start of a slow deflation stage), and the pressure transition from the hold stage to the deflation stage.
FIG. 2
illustrates such a cuff pressure signal after it has been digitized at 50 samples per second, wherein the transition from the pressure-hold stage to the deflation stage occurs near time sample
820
. Transitions such as these create a relatively large step-response (i.e., disturbance) in the output of the high-pass filter (HPF), which can take several seconds to settle out. The exact length of the settling time depends on the cut-off frequency of the HPF, with lower cut-off frequencies yielding longer settling times.
FIG. 3
illustrates the sampled pressure signal at the output of the HPF. As shown herein, the settling time effect causes distortion in the amplitude of legitimate pulses that appear at the output of the HPF after the occurrence of the transition, thereby delaying the accurate detection and measurement of the blood pressure of the patient. The distortion effect is at its maximum at time sample
831
, and continues to about sample time
950
(note, the increasing amplitude of pulsatile components is representative of the cuff pressure approaching the mean arterial pressure of the patient).
Since the NIBP measurement process depends upon an accurate analysis of pulse amplitude, one approach to the problem of pressure transitions is to ignore the distorted pulses. However, measurement time is also an important consideration for NIBP gauges because blood perfusion is reduced to the limb during the measurement process.
Ideally, the effects of sharp transitions in the cuff pressure signal should be quickly removed from the output of the HPF so that the accurate measurement of the blood pressure pulses can recommence.
SUMMARY OF THE INVENTION
A method and apparatus to reset a digital HPF for improving its settling time in response to a sharp transition in its input signal, by rapid input and clocking through the HPF of new data samples after occurrence of the sharp transition. The new data samples comprise an artificial ramp representative of the average slope of the input signal that should be maintained after occurrence of the sharp transition. The artificial ramp can be generated using “a priori” knowledge of an expected average slope of the input signal, or alternatively, can be generated using an instantaneous measurement of the slope of the input signal at the time the HPF is to be reset.
In the event that “a priori” knowledge of the slope of the input signal is available, and the instantaneous measurement of the slope of the input signal at the time the HPF is to be reset differs therefrom, it is desirable to generate a ramp whose slope provides a smoothed transition from the latter slope measurement to the former slope, since an abrupt transition here can also create undesirable distortion effects.
REFERENCES:
patent: 4542475 (1985-09-01), Acampora
patent: 4875106 (1989-10-01), Heitman
patent: 5388221 (1995-02-01), Meitner et al.
patent: 5420535 (1995-05-01), Nebuloni et al.
patent: 5740096 (1998-04-01), Shou et al.
patent: 5867410 (1999-02-01), Smallcombe et al.
patent: 6023571 (2000-02-01), Matsumoto et al.
patent: 195 18 528 (1996-01-01), None
patent: 0 343 370 (1989-11-01), None
patent: WO 94 23427 (1994-10-01), None
Burke Alexander J.
Mai Tan V.
Siemens Medical Systems Inc.
LandOfFree
Real-time technique for reducing the settling time of a high... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Real-time technique for reducing the settling time of a high..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Real-time technique for reducing the settling time of a high... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2449567