Peak analyzing method and waveform processing device

    公开(公告)号:US11965865B2

    公开(公告)日:2024-04-23

    申请号:US17167503

    申请日:2021-02-04

    Inventor: Shin Fujita

    CPC classification number: G01N30/8637 G01N30/7233 G01N2030/027

    Abstract: A peak analyzing method for separating overlapping peaks observed in a second signal waveform representing a relationship between a second parameter and a signal intensity into a plurality of individual peaks originating from different factors based on signal patterns observed in a dimension of a first parameter, the method including at least: performing singular value decomposition on an input matrix expressing three-dimensional data to be processed; estimating characteristic orientations within a space spanned by a plurality of basis vectors by performing a geometric analysis on a trajectory defined by a plurality of weighting vectors in an SVD projection space whose number of dimensions is equal to a lowered rank given by a singular value decomposition process; and deconvoluting signal waveforms in a first matrix of a dimension of the first parameter by a transformation matrix.

    SAMPLE QUANTIFICATION CONSISTENCY AND CLASSIFICATION WORKFLOW

    公开(公告)号:US20240102977A1

    公开(公告)日:2024-03-28

    申请号:US18471808

    申请日:2023-09-21

    CPC classification number: G01N30/8693 G01N30/7233 G01N30/8631 G01N2030/027

    Abstract: The present technology relates to a method and instrument for classifying a sample. The method collects raw data from an analytical instrument (e.g., LC-MS), quantifies consistency parameters from the raw chromatographic data (e.g., peak detection) by assigning criteria and determining probability ratios, and constructs a learning model quantitation step by weighing the presence, absence, or modulation of one or more factors of the one or more samples against a decision criterion for positivity. The method can then apply the learning model to an unknown sample to detect the presence, absence, or modulation of one or more factors in the unknown sample such as the presence, absence, or modulation of a disease state or multiple disease states.

Patent Agency Ranking