| Patent application number | Description | Published |
|---|
| 20090063102 | METHOD FOR IDENTIFYING A CONVOLVED PEAK - A method for identifying a convolved peak is described. A plurality of spectra is obtained. A multivariate analysis technique is used to assign data points from the plurality of spectra to a plurality of groups. A peak is selected from the plurality of spectra. If the peak includes data points assigned to two or more groups of the plurality of groups, the peak is identified as a convolved peak. Principal component analysis is one multivariate analysis technique that is used to assign data points. A number of principal components are selected. A subset principal component space is created. A data point in the subset principal component space is selected. A vector is extended from the origin of the subset principal component space to the data point. One or more data points within a spatial angle around the vector are assigned to a group. | 03-05-2009 |
| 20090063592 | METHODS FOR DATA PROCESSING - According to various embodiments, variables are grouped in an unsupervised manner after principal component analysis of a plurality of variables from a plurality of samples. A number of principal components are selected. A subset principal component space is created for those components. A starting variable is selected. A spatial angle is defined around a vector extending from the origin to the starting variable. A set of one or more variables is selected within the spatial angle. The set is assigned to a group. The set is removed from further analysis. The process is repeated starting with the selection of a new starting variable until all groups are found. | 03-05-2009 |
| 20090254314 | SYSTEMS AND METHODS FOR IDENTIFYING CORRELATED VARIABLES IN LARGE AMOUNTS OF DATA - Groups of correlated representations of variables are identified from a large amount of spectrometry data. A plurality of samples is analyzed and a plurality of measured variables is obtained from a spectrometer. A processor executes a number of steps. The plurality of measured variables is divided into a plurality of measured variable subsets. Principal component analysis followed by variable grouping (PCVG) is performed on each measured variable subset, producing one or more group representations for each measured variable subset and a plurality of group representations for the plurality of measured variable subsets. While the total number of the plurality of group representations is greater than a maximum number, the plurality of group representations is divided into a plurality of representative subsets and PCVG is performed on each subset. PCVG is performed on the remaining the plurality of group representations, producing a plurality of groups of correlated representations of variables. | 10-08-2009 |
| 20090259438 | RELATIVE NOISE - Relative noise is a single scalar value that is used to predict the maximum value of the expected noise at any point and is calculated from the measured signal and a mathematical noise model. The mathematical noise model is selected or estimated from an observation that includes statistical and/or numerical modeling based on a population of measurement points. An absolute noise for a plurality of points of the measured signal is estimated. An array of values is calculated by dividing each of a plurality of points of the absolute noise by a corresponding expected noise value calculated from the mathematical noise model. The relative noise is calculated by taking a standard deviation of a plurality of points of the array. The relative noise can be used to calculate scaled background signal noise, filter regions, denoise data, detect false positives from features, calculate S/N, and determine a stop condition for acquiring data. | 10-15-2009 |
