Data processing: measuring – calibrating – or testing – Measurement system in a specific environment – Chemical analysis
Reexamination Certificate
2001-09-07
2004-02-17
Hoff, Marc S. (Department: 2857)
Data processing: measuring, calibrating, or testing
Measurement system in a specific environment
Chemical analysis
C703S002000
Reexamination Certificate
active
06694265
ABSTRACT:
FIELD OF INVENTION
The present invention is directed to detectors, and more particularly to methods and apparatus for determining the boundaries of a detector response profile.
BACKGROUND OF THE INVENTION
Detectors and sensors are used in industry and in research to determine the presence or absence of a molecule, material, chemical, objects or a change in a physical parameter of the environment around or in contact with the detector. By way of example, a thermometer senses changes in temperature. If viewed over time, temperature, one variable, having a value in degrees, can be compared to a second variable, time, measured in seconds, minutes or years. These variables can be graphed or plotted.
As used herein, the term “detector” is used to refer to any instrument or device which creates a signal in response to the presence or absence of a molecule or a change in a physical parameter of the environment in contact with the instrument or device. Common detectors used in industry and research include absorbance detectors, fluorescence detectors, mass spectrometry detectors, chemi-luminescence detectors, refractometry detectors, viscometry detectors, radiation detectors and thermometers.
As used herein, the word “profile” refers to a depiction of a plot of data points usually with lines drawn between the data points whether in electronic form or printed. A “plot” refers to graphical organization of a series of data points, in electronic and printed forms. Such a plot is normally presented on an X-axis and an Y-axis where each axis represents one variable. Typically detectors measure responses at a uniform sample rate. Thus the time difference between adjoining data points is a constant that is termed the sample period. Typically, a plot of data points from a detector will have data points associated with no activity, which result in consistent readings or consistent slope. These consistent readings, representative of no activity or change, are referred to as a baseline. A change in the environment surrounding the detector will alter the profile of the plot creating a “peak” or “valley.” As used herein, the word “peak” refers to any change in the profile, whether plotted as an upwardly projecting or downwardly descending plot. Whether the plot is directed up or down is a matter of choice and the present discussion will address each as a peak for the convenience of clarity. The words plot and profile, as used herein, are not intended to be limited to visually perceived representations. Rather, the words are used to represent how data points are managed or processed to depict information.
Chromatography is the science of separation based upon specific or nonspecific binding of molecules to a stationary phase. Aspects of the present invention have special application in gas and liquid chromatography. In liquid chromatography, a liquid carrying one or more compositions of interest is carried through a solid phase. The compositions elute from the solid phase at different times producing changes in one or more physical parameters measured by the detector. These changes are plotted over time and such graphical representations are known as a chromatograph. Chromatographs typically exhibit peaks that correspond to the compounds that have been separated. It is often desirable to direct fluids containing the compounds to vessels or further processes. By way of example, it may be useful to direct a fluid containing a compound, determined by absorbance, to a mass spectrometer to determine its molecular weight. Also, it may be useful to collect the fluid defining a separated peak into a collection vial. This is an operation known as “fraction collection.” Thus, chromatographic instruments often are equipped with valves for directing compounds from the common stream.
It is accepted practice to analyze a peak in order to obtain two response factors, peak height and peak area. Each of these factors gives a response that is in proportion to the amount of material injected onto the column. But the height and area can be obtained only when the underlying baseline of a peak and the start time and stop time of the peak are known. Ideally a chromatogram consists of a series of peaks, with all pairs separated by a region of baseline. These peaks are termed baseline-resolved peaks. Chromatographs, however, can be complex. A peak from a compound may appear within a cluster of peaks, or merge with other peaks, or appear not as a well-defined peak, but as a shoulder.
In the case of a baseline-resolved peak, the boundaries of the peak are taken to be the start and stop times of the peak's baselines. The start time, also known as “lift-off,” is when the peak first appears above the baseline response. The stop time, also known as “touchdown,” is when the peak response becomes coincident with the baseline response. These times determine the baseline drawn under the peak, which is needed for the determination of height and area. It is useful to determine height and area with computers equipped with appropriate software, such as software including an integration algorithm. For the integration algorithm to be useful, it must determine lift-off and touchdown accurately and reproducibly for peaks of varying heights, and shape asymmetries.
Unfortunately, the times of “liftoff” and “touchdown” are dependent on the height of the peak. As the height of the peak changes, the position of “lift-off” and “touchdown” change. The higher the peak, the further apart these points become. Therefore, dependence on peak height is undesirable, as it requires the practitioner to find a compromise value, or to change the value as the peak heights change. Further, the results, using prior art methods, are dependent on the baseline slope. If the slope of the baseline changes, the position of “lift-off” and “touchdown” change. Problems with this dependence on the baseline slope become significant in the case of small peaks. In the case of small peaks, positive slopes yield start and stop times that occur earlier and negative slopes yield start and stops times that occur later.
In the case of a cluster, the baseline start time is the liftoff time for the first peak in a cluster, and the baseline stop time is the touchdown time for the last peak in the cluster. Their lift off and touch down points must be determined with the presence of adjacent peaks of varying heights and resolutions and shape asymmetries.
Once lift-off and touchdown are established, the next step is to determine boundaries between peaks in a cluster. If a valley separates a pair of peaks, the determination of the boundary is straightforward. Valleys are the local minima between peak apices. The point at the minimum of the valley is the boundary, defining the stop time of the prior peak and the start time of the following peak.
The identification of an appropriate peak boundary for a shoulder is a more difficult problem. A shoulder occurs when two peaks co-elute with low enough resolution such that there is no valley between the peak apices. The shoulder cannot be detected separately from the main peak because there is no valley point separating them. Further, even if the apex of the shoulder is identified, there is no obvious means to demarcate the shoulder from the adjoining peak. The demarcation between the main peaks and the shoulder is hard to define and there is no accepted method within the prior art to demarcate a shoulder from an adjoining peak.
Accurate and reliable determination of lift-off and touch down is essential to accurate and reliable quantitation. Lift-off and touch-down establish a baseline, and it is the baseline that affects all subsequent determinations of peak heights and areas. Accuracy is compromised if the determination of baseline is erroneous or non-reproducible.
The data analysis problem for fraction collection is similar to problems found in quantitation, in that peak boundaries must be determined. Fraction collection, however, because of its real time nature, requires an algorithm that operates in real time. It must be able to handle
Brown, Rudnick, Berlack & Israels, LLP.
Charioui Mohamed
Michaelis Brian
Serio John
Waters Investments Limited
LandOfFree
Method and apparatus for determining the boundaries of a... 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 the boundaries of a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for determining the boundaries of a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3314457