Surgery – Diagnostic testing – Measuring or detecting nonradioactive constituent of body...
Reexamination Certificate
2002-01-14
2003-11-11
Winakur, Eric F. (Department: 3736)
Surgery
Diagnostic testing
Measuring or detecting nonradioactive constituent of body...
C702S075000, C702S076000, C704S214000, C704S208000, C704S267000, C600S309000, C600S310000, C600S336000
Reexamination Certificate
active
06647280
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to the field of signal processing. More specifically, the present invention relates to a numerical transform for obtaining a fundamental period from a signal, particularly for a numerical transform method suitable for determining the period and filtering noise from a physiological signal. The present invention is particularly suitable for use in oximetery applications.
BACKGROUND
In medical or physiological monitoring, physiological measurements required to determine parameters such as blood pressure and blood oxygen saturation levels are often dependent on a valid comparison of monitored waveforms or data to the patient's heart pulse.
For example, commercially available pulse oximeters generally measure energy attenuation or light absorption in biological tissue to determine blood oxygen saturation levels. One common type of pulse oximeter employs light in two wavelengths to obtain the arterial oxygenation level, wherein light in the red range and light in the infrared range are directed at the patient and detected by a photodetector. With each cardiac cycle, there is cyclic light absorption by tissue beds. During diastole, absorption is a result of venous blood, tissue, bone and pigments. During systole, light absorption is increased by the influx of arterialized blood into the tissue bed. The resulting pulsatile variation absorption through time is called the plethysmographic signal.
The oximeter determines the difference between background absorption during diastole and peak absorption during systole at both red and infrared wavelengths, as this difference corresponds to the absorption caused by arterialized blood. Since oxygen saturation determines the red:infrared light absorption ratio, differences in this ratio are used to compute the arterial oxygen saturation, a value derived empirically, generally through a calibration curve.
Accurate measurement of arterial oxygen saturation, therefore, is highly dependent upon an accurate reading of a pulse waveform. The pulse waveform is typically detected by a sensor disposed on an extremity or, in the case of adults, on the nose or ear. These sensors, however, often do not provide an accurate reading due to the motion artifacts caused by muscle movement, vasoconstriction associated with hypotension, shivering, motion of the body site where a sensor is affixed, or other types of internal or external movement during the measurement process. Other sources of noise are also problematic, in particular electromagnetic, measurement, and intrinsic error sources can also cause noise problems. These noise factors can cause the properties of light or energy attenuation to vary erratically. Traditional signal filtering techniques are frequently totally ineffective and grossly deficient in removing these motion-induced effects from a signal. The erratic or unpredictable nature of motion induced signal components is the major obstacle in removing or deriving them. Thus, presently available physiological monitors generally become inoperative or inaccurate during time periods when the measurement site is perturbed.
Furthermore, the problems associated with detecting proper heart pulses and evaluating blood oxygenation levels are significantly more difficult with respect to fetal monitoring. The fetal oximeter sensing function takes place in a physically constrained environment (the uterus), subject to fetal motion, substantial pressure variations (contractions), and interference by the presence of a variety of fluids in the environment (amniotic fluid, meconium, maternal blood). The sensing must be done using reflectance, rather than transmissive, pulse oximetry, further compromising the signal to noise ratio of the measurement. Despite many attempts at sensor design to improve signal quality, the problem of successfully monitoring in the presence of these many factors has not been solved in the prior art. Furthermore fetal physiology, is characterized by typically low arterial oxygen saturation (often below 50%), and a much lower difference between arterial and venous saturation. For this reason, it is doubtful that some of the advanced algorithms for noise rejection, designed with the assumption of more mature physiology, will function.
Lastly, these inventions tend to either not treat the question of pulse rate extraction from the plethysmographic waveform, or present solutions not ideal for the fetal physiology and environment. Fetal heart rate (FHR) measurement is currently the most important parameter monitored in utero to detect signs of fetal distress. Typically, FHR is determined by intermittent manual auscultation through the mother's abdomen by a trained caregiver, or by means of an ECG electrode screwed into the fetal scalp. Thus a means of reliable non-invasive pulse rate determination in fetal monitoring would offer a significant improvement over the prior art.
For example, one method for dealing with noise problems is described in U.S. Pat. No. 5,588,425 to Sackner et al. Here, a narrow range of systolic upstroke times are empirically defined and waveforms are deemed valid only if they fall within the predetermined range. The system, therefore, depends on a consistent, narrow range of upstroke times for all patients to be effective. These values may not apply adequately to fetal monitoring situations. Furthermore, irregular pulses which may provide valuable diagnostic information, are ignored.
Another method for removing noise from a pulse waveform signal is disclosed in U.S. Pat. No. 5,934,277 to Mortz. Mortz discloses a system wherein statistical techniques such as linear regression and correlation are used to filter noise from the signal. This system requires a significant amount of pre-calculation filtering to assure that the input signals from both the red and infrared are consistent. Furthermore, “good data” must be identified before a calculation can be made. For example, if the processed signal contains a significant variation in evaluated points, the system will determine that “good data” has not been achieved. If the data is insufficiently “good” an alarm is set for the user. This system, therefore, is only capable of processing data in a well-defined “good” range, and cannot filter noise factors sufficiently to provide a useable signal in many applications, particularly in fetal applications.
U.S. Pat. No. 5,687,722 to Tien et al. also discloses a system based on a regression algorithm. Here, incoming data is filtered into a number of data windows. A ratio value indicative of oxygen saturation in the patient is statistically analyzed to determine the best estimate for the correct ratio value. Although this algorithm may be capable of extracting oxygen saturation information from non-optimal signals, it does not yield a pulse rate value. In fetal monitoring, where alternative sources of continuous pulse rate information are inconvenient, these situations algorithms of this type are insufficient.
Although the problem has been described mainly with reference to pulse oximetry systems, a similar problem exists in conjunction with a number of different types of physiological monitoring including electrocardiographs, blood pressure, capnographs, heart rate, respiration rate, and depth of anesthesia, for example. Other types of measurements include those which measure the pressure and quantity of a substance within the body such as breathalyzer testing, drug testing, cholesterol testing, glucose testing, arterial carbon dioxide testing, protein testing, and carbon monoxide testing, for example. In all of these types of monitoring, the ability to derive an accurate pulse waveform is extremely important in providing an accurate physiological reading.
SUMMARY OF THE INVENTION
The present invention relates generally to a method of signal processing for use with physiological monitoring or other types of monitoring which involve signals containing a relatively low dominant frequency (order of Hz to tens of Hz), with intermittent broad-band noise of significant amplit
Bahr Dennis E.
Reuss. James L.
Kremer Matthew J
OB Scientific, Inc.
Peinhart Boerner Van Deuren s.c.
Winakur Eric F.
LandOfFree
Method and apparatus for processing a physiological signal does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for processing a physiological signal, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for processing a physiological signal will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3175570