Surgery – Diagnostic testing – Measuring or detecting nonradioactive constituent of body...
Reexamination Certificate
2002-05-14
2003-10-07
Winakur, Eric F. (Department: 3736)
Surgery
Diagnostic testing
Measuring or detecting nonradioactive constituent of body...
C600S323000, C600S481000, C702S019000
Reexamination Certificate
active
06631281
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to the recognition of a useful signal in a measurement signal.
In general, the measurement of signals can be roughly divided into a) the recognition of individual, more or less singular events and b) the monitoring of more or less frequently recurrent, essentially periodical signals. In both cases, superimposing disturbances limit the confidence level of the measurement and it is desirable to avoid, suppress or filter these disturbances.
Periodical signals are herein understood to mean signals in which the useful signal has at least a periodical component at least in a given time slot, but whose frequency may be time-dependent.
Particularly in the medical field of patient monitoring, the recognition of the useful signal and suppression of disturbances is essential, because disturbances lead to false interpretations of the measured values or may render the measurement as a whole unusable.
A measurement which has proved to be very sensitive to disturbing influences, is the pulsoximetric determination of the oxygen content of blood, because pulsoximetry is often more affected by motion artifacts than by the pulse signal determining the blood oxygen content. Pulsoximetry relates to the non-invasive, continuous determination of the oxygen content of blood (oximetry), based on the analysis of the photospectrometrically measured pulse. To this end, it is necessary that a pulse curve (plethysmogram) is available in the case of a plurality of wavelengths. In practice, substantially all apparatuses operate at two wavelengths only, so that low-cost, compact solutions are possible. The photometry principle is based on the fact that the quantity of the absorbed light is determined by the degree of absorption of a substance and by the wavelength. Pulsoximeters utilize the aspect that the arterial blood volume, and only the arterial blood volume, pulsates in the rhythm of the heartbeat. In order to determine the value of oxygen saturation from the determined measured data, a ratio is derived from the measured data, which ratio then represents the oxygen saturation value. The fundamental aspects and fields of use of pulsoximetry are generally known and frequently described, particularly in EP-A-262778 (with a good theory outline), U.S. Pat. No. 4,167,331, or by Kästle et al. in “A New Family of Sensors for Pulsoximetry”, Hewlett-Packard Journal, vol. 48, no. 1, pp. 39 to 53, February 1997.
For the pulsoximetric measurement, particularly methods in the temporal range, adaptive filter spectral analyses and methods in the temporal frequency range have been proposed as methods of recognizing and suppressing artifacts. A detailed description of these methods of suppressing artifacts (which methods are less interesting within the context of this application) is given in the international patent application in the name of the applicant, filed on the same application date (file 20-99-0010).
While the useful signal should remain possibly unaffected in the above-mentioned methods of recognizing and suppressing artifacts, and only the artifacts should be eliminated, the prior art also discloses methods in which, conversely, (only) the useful signal should be filtered from the measurement signal. In addition to the temporal range method (again less interesting within the context of this invention), particularly those methods in which the measurement signals are examined in the frequency range have proved to be advantageous for determining or filtering a periodical useful signal from a more or less disturbed measurement signal. Such methods for use in pulsoximetry are described, inter alia, in U.S. Pat. No. 5,575,284 (Athan), WO-A-96 12435 (Masimo) or EP-A-870466 (Kästle).
According to WO-A-96 12435, a signal is selected as useful signal after transformation of the pulsoximetric measured values in the frequency range by determining the frequency component having the strongest amplitude.
EP-A-870466, by the same inventor and the same applicant, discloses a method of selecting the pulsoximetric signal in accordance with the physiological relevance of the frequency components. After optional suppression of the DC component of the two pulsoximetric raw, or unconditioned, signals (red and infrared), the unconditioned signal values which are present in a continuous time slot are transformed into the frequency range by means of a Fourier transform (here, Fast Fourier Transform—FFT). Ratios of the coefficients of the amplitude spectrum are formed from the transformed unconditioned signals for all frequency peaks. When the infrared spectrum is graphically plotted in the x direction and the red spectrum in the y direction, a representation having needle-like peaks is obtained. These needles correspond to the peaks of the spectra, with very thin needles being obtained for undisturbed signals and the relevant needles of the fundamental and harmonic waves being superimposed. The angle of the needles with respect to the axes corresponds to the saturation value. Since the representation of the spectra is similar to a pincushion in this case, the method described in EP-A870466 is also referred to as “pincushion algorithm”.
To identify the needle representing the pulsoximetric signal, a distance spectrum is first determined in the pincushion algorithm from the complex amplitudes of the red and infrared spectra. The distance spectrum describes the distance between every individual point in the needle diagram from the origin. The individual needles are determined from this distance spectrum by considering the maxima and the attendant foot points. Only those needles which fulfill a series of given criteria are maintained for the further considerations. The reduced selection of needles is subjected to a further classification. Needles representing the useful signal should fulfill the criteria that the peaks fit well in a harmonic frequency range, as many harmonic waves as possible are available, the needles are possibly thin and the frequency of the fundamental wave as well as the saturation value, the perfusion and the pulse rate are within physiological ranges. An overall evaluation for each needle is effected by assigning points or K.O. criteria to each of these criteria. The needle that is given the largest number of points, or in other words, best satisfies the criteria, and has been given at least a minimal number of points is used for determining the output value for the pulsoximetric measured value. Optionally, a comparison with previous output values may be used for plausibility control purposes, and in the case of a significant deviation from the previous output values, the newly determined output value is rejected and no new value will be displayed.
The methods of determining the useful signal by transformation into the frequency range have proved to be clearly less sensitive to disturbances than the filtering method in a temporal range. However, also these methods may yield uncertainties in the frequency range, dependent on the disturbing situation, in which, for safety's sake, no value or only a questionable value can be supplied.
OBJECT AND SUMMARY OF THE INVENTION
It is therefore an object of the present invention to enhance the recognition of a periodical useful signal in a determined measurement signal. This object is solved by the characteristic features of the independent claims. Advantageous embodiments are defined in the dependent claims.
According to the invention, the recognition of a periodical useful signal in a (disturbed) measurement signal is effected in several process steps.
In a first step, there is a transformation of the measurement signal for a given time slot into the frequency range. The Fast Fourier Transform (FFT) is particularly suitable for this purpose, but other arbitrary transformations may be used alternatively.
Optionally, the measurement signal may be filtered before or after the transformation. Preferably, such a filtering is effected, for example, by reducing the DC component (particularly as described in EP-A-87046
Vodopia John
Winakur Eric F.
LandOfFree
Recognition of a useful signal in a measurement 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 Recognition of a useful signal in a measurement signal, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Recognition of a useful signal in a measurement signal will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3135238