# US 2019/0378702 A1 · 3d mass spectrometry predictive classification > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2019/0378702 A1 - Document ID: US20190378702A1 - Status: Published application - Kind: A1 - Application number: 16/003,373 - Filed: 2018-06-08 - Published: 2019-12-12 - Earliest claimed: 2018-06-08 - Assignees: Thermo Finnigan LLC - Applicants: Thermo Finnigan LLC - Inventors: Michael J. ATHANAS; Hossein ARJOMAND - Web page: https://patentgrep.com/patent/US20190378702A1 ## CPC - G16C 20/20: Identification of molecular entities, parts thereof or of chemical compositions - G06F 18/2431: Multiple classes - G16C 20/70: Machine learning, data mining or chemometrics - H01J 49/0004: {Imaging particle spectrometry} - H01J 49/0036: {Step by step routines describing the handling of the data generated during a measurement} - H01J 49/004: {Combinations of spectrometers, tandem spectrometers, e.g. MS/MS, MSn} - H01J 49/02: Details - G16C: COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - H01J: ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS (spark-gaps H01T; electric arc lamps H05B31/00; particle accelerators H05H) ## Abstract A method for analyzing a multidimensional data set includes generating a multidimensional mass spectrometry data set from a sample; and generating an matrix representing the multidimensional mass spectrometry data set such that a first dimension and a second dimension of the multidimensional mass spectrometry data set correspond to a matrix cell location, and an ion intensity corresponds to a matrix cell value; and determining a class of the matrix from a plurality of matrix classes using a trained neural network matrix classifier. ## Claims ### Claim 1 A system for analyzing a multidimensional data set, comprising: a mass spectrometer configured to: generate a multidimensional data set from a sample; and provide the multidimensional data set to an matrix generator; the matrix generator configured to: receive the multidimensional data set; generate a matrix representing the multidimensional data set such that a first dimension and a second dimension of the multidimensional data set correspond to a matrix cell location, and an ion intensity corresponds to a matrix cell value; and provide the image to an matrix classifier; and the matrix classifier configured to: receive the matrix from the matrix generator; and determine a class of the matrix from a plurality of matrix classes. ### Claim 2 (depends on 1) The system of claim 1 wherein (i) the matrix generator is further configured to rescale the matrix prior to providing the matrix to matrix classifier or (ii) the matrix classifier is further configured to rescale the matrix prior to determining the class of the matrix. ### Claim 3 (depends on 1) The system of claim 1 wherein determining the class of the matrix is performed using a trained neural network. ### Claim 4 (depends on 1) The system of claim 1 wherein the first dimension is retention time and the second dimension is mass-to-charge ratio. ### Claim 5 (depends on 1) The system of claim 1 wherein the multidimensional data set is an MS/MS experiment, the first dimension is precursor mass-to-charge ratio and the second dimension is product mass-to-charge ratio. ### Claim 6 (depends on 1) The system of claim 1 wherein the matrix is an image. ### Claim 7 (depends on 6) The system of claim 6 wherein the image is a greyscale image with a single-color value per pixel. ### Claim 8 (depends on 6) The system of claim 6 wherein the image is a color image with at least two color values per pixel, and additional dimensions correspond to the additional color values. ### Claim 9 A method for analyzing a multidimensional data set, comprising: generating a multidimensional mass spectrometry data set from a sample; generating an matrix representing the multidimensional mass spectrometry data set such that a first dimension and a second dimension of the multidimensional mass spectrometry data set correspond to a matrix cell location, and an ion intensity corresponds to a matrix cell value; and determining a class of the matrix from a plurality of matrix classes using a trained neural network matrix classifier. ### Claim 10 (depends on 7) The method of claim 7 further comprising rescaling the matrix prior to determining the class of the image. ### Claim 11 (depends on 7) The method of claim 7 wherein the first dimension is retention time and the second dimension is mass-to-charge ratio. ### Claim 12 (depends on 7) The method of claim 7 wherein the multidimensional data set is an MS/MS experiment, the first dimension is precursor mass-to-charge ratio and the second dimension is product mass-to-charge ratio. ### Claim 13 (depends on 7) The method of claim 7 wherein the matrix is an image. ### Claim 14 (depends on 12) The method of claim 12 wherein the image is a greyscale image with a single color value per pixel. ### Claim 15 (depends on 12) The method of claim 12 wherein the image is a color image with at least two color values per pixel, and additional dimensions correspond to the additional color values. ### Claim 16 A system for assessing an instrument, comprising: the instrument configured to: perform a series of diagnostic routines; and provide results of the series of diagnostic routines to an matrix generator; the matrix generator configured to: receive the results of the series of diagnostic routines; generate an multidimensional matrix representing the results of the series of diagnostic routines with a matrix cell location corresponding to one of the series of diagnostic routines and a time point during the diagnostic routine, and a value of the matrix cell corresponding to a measured value of the diagnostic routine; and provide the multidimensional matrix to a matrix classifier; and the matrix classifier configured to: receive the multidimensional matrix from the matrix generator; and determine a class of the multidimensional matrix from a plurality of matrix classes, the class indicating a corrective action necessary to bring the instrument into nominal operation. ### Claim 17 (depends on 12) The system of claim 12 wherein (i) the matrix generator is further configured to rescale the multidimensional matrix prior to providing the multidimensional matrix to the matrix classifier or (ii) the matrix classifier is configured to rescale the multidimensional matrix prior to determining the class of the multidimensional matrix. ### Claim 18 (depends on 12) The system of claim 12 wherein determining the class of the multidimensional matrix is performed using a trained neural network. ### Claim 19 (depends on 12) The system of claim 12 wherein the multidimensional matrix is an image. ### Claim 20 (depends on 18) The system of claim 18 wherein the image is a greyscale image with a single color value per pixel. ### Claim 21 (depends on 18) The system of claim 18 wherein the image is a color image with at least two color values per pixel, and additional dimensions correspond to the system parameters monitored during the series of diagnostic routines. ### Claim at position 22 -38. (canceled)