Surgery – Diagnostic testing – Measuring or detecting nonradioactive constituent of body...
Reexamination Certificate
2000-09-29
2003-01-07
Winakur, Eric F. (Department: 3736)
Surgery
Diagnostic testing
Measuring or detecting nonradioactive constituent of body...
C600S336000
Reexamination Certificate
active
06505060
ABSTRACT:
FIELD OF THE INVENTION
The present relates in general to processing detector information in a pulse oximetry system and, in particular, to the determination of differential values for use in blood oxygenation calculations with reduced noise sensitivity.
BACKGROUND OF THE INVENTION
In the field of photoplethysmography, light signals corresponding with two or more different centered wavelengths may be employed to non-invasively determine various blood analyte concentrations. By way of example, blood oxygen saturation (SpO
2
) levels of a patient's arterial blood are monitored in pulse oximeters by measuring the absorption of oxyhemoglobin and reduced hemoglobin using red and infrared light signals. The measured absorption data allows for the calculation of the relative concentrations of reduced hemoglobin and oxyhemoglobin, and therefore SpO
2
levels, since reduced hemoglobin absorbs more light than oxyhemoglobin in the red band and oxyhemoglobin absorbs more light than reduced hemoglobin in the infrared band, and since the absorption relationship of the two analytes in the red and infrared bands is known.
To obtain absorption data, pulse oximeters comprise a probe that is releasably attached to a patient's appendage (e.g., finger, ear lobe or the nasal septum). The probe directs red and infrared light signals to the appendage, or tissue-under-test. The light signals are provided by one or more sources which are typically disposed in the probe. A portion of the light signals is absorbed by the tissue-under-test and the intensity of the light transmitted through or reflected by the tissue-under-test is detected, usually by at least one detector that may also be located in the probe. The intensity of an output signal from the detector(s) is utilized to compute SPO
2
levels, most typically via a processor located in a patient monitor interconnected to the probe.
As will be appreciated, pulse oximeters rely on the time-varying absorption of light by a tissue-under-test as it is supplied with pulsating arterial blood. The tissue-under-test may contain a number of non-pulsatile light absorbers, including capillary and venous blood, as well as muscle, connective tissue and bone. Consequently, detector output signals typically contain a large non-pulsatile, or DC, component, and a relatively small pulsatile, or AC, component. It is the small pulsatile, AC component that provides the time-varying absorption information utilized to compute arterial SpO
2
levels.
In this regard, the red and infrared signal portions of pulse oximeter detector output signals each comprise corresponding large DC and relatively small AC components. The red and infrared signal portions have an exponential relationship to their respective incident intensities at the detector(s). As such, the argument of the red and infrared signal portions have a linear relationship and such portions can be filtered and processed to obtain a ratio of processed red and infrared signal components (e.g., comprising their corresponding AC and DC components), from which the concentration of oxyhemoglobin and reduced hemoglobin in the arterial blood may be determined. See, e.g., U.S. Pat. No. 5,934,277. By utilizing additional light signals at different corresponding centered wavelengths it is also known that carboxyhemoglobin and methemoglobin concentrations can be determined. See, e.g., U.S. Pat. No. 5,842,979.
As noted, the pulsatile, AC component of a pulse oximeter detector output signal is relatively small compared to the non-pulsatile DC component. Consequently, the accuracy of analyte measurements can be severely impacted by small amounts of noise. One such type of noise relates to effects on the measured absorption data as a result of undesired variations in the path length of light signals as they pass through the tissue-under-test. Such variations are most typically caused by patient movement of the appendage to which a pulse oximetry probe is attached.
A number of different approaches have been utilized to reduce the deleterious effects of patient motion in pulse oximeters. For example, pulse oximeter probes have been developed to enhance the physical interface between the probe and tissue-under-test, including the development of various clamp type probe configurations and secure wrap-type probe configurations. Further, numerous approaches have been developed for addressing motion contaminated data through data processing techniques. While such processing techniques have achieved a degree of success, they often entail extensive signal processing requirements, thereby contributing to increased device complexity and componentry costs.
Other types of noise include electrical and optical phenomena that cause artifact in the pulsatile component of the measured absorption data. For example, effects due to ambient light and interfering electrical signals can provide significant noise components. Many of these noise sources are not easily filtered out of the detector signal and, therefore, are reflected in the measured absorption data. It will be appreciated that such noise in the measured absorption data can significantly affect blood oxygen saturation calculations if not adequately accounted for in signal/data processing.
The case of calculating derivatives of the measured absorption data signal is illustrative. As noted above, pulse oximetry blood oxygenation calculations are generally based on measuring the relative time varying absorption or optical signal attenuation at two or more wavelengths by the tissue-under-test. Specifically, a ratio of corresponding differential values, such as the normalized derivative of attenuation (NdA) for each of two centered wavelengths or channels may be calculated. The time derivative of the attenuation divided by the attenuation provides the NdA. The ratio of the NdAs for the red and infrared wavelengths, as often employed in pulse oximeters, is directly proportional to SpO
2
.
The NdAs for each wavelength have generally been calculated in two ways. Most commonly, the NdAs have been approximated by measuring a peak to trough amplitude of the pulsatile signal. However, this methodology is sensitive to noise at the data points associated with the peak and trough. Moreover, the response time of such pulse oximeters is limited due to the elongated sampling interval required for NdA calculations. Additionally, this methodology can suffer from reduced accuracy if the delays of the high pass AC filters and low pass DC filters, used to separate the pulsatile and non-pulsatile components of the detector signal in connection with peak and trough identification, are not carefully matched. As noted above, calculation of the NdAs involves dividing the time derivative of the attenuation by the overall attenuation including the DC component. This calculation assumes that the time derivative and DC component are sampled at substantially the same time and, accordingly, any differences in the filter delays can introduce an element of error. Additionally, such filters can take considerable time to stabilize before the oximeter can calculate accurate derivatives.
Another common method of estimating the NdA involves calculating a difference between successive data points of the processed detector signal. This difference is normalized by dividing by an average DC value for the two data points. This methodology avoids many of the disadvantages of peak-to-trough calculations, and generates output for every sample, but the change in signal level is much smaller as between successive samples as compared to peak-to-trough amplitude calculations. As a result, individual measurements are sensitive to noise.
In pulse oximetry systems developed by Datex-Ohmeda, Inc., successive data point calculations are employed but multiple data sets are utilized to determine the NdA ratio used for Sp0
2
calculations thereby improving accuracy. In particular, absorption related values are calculated for each channel for multiple samples over a measurement period. For each sample time during the measurement period, a rat
Datex-Ohmeda Inc.
Kremer Matthew
Marsh & Fischmann & Breyfogle LLP
Winakur Eric F.
LandOfFree
Method and apparatus for determining pulse oximetry... 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 determining pulse oximetry..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for determining pulse oximetry... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3048181