Methods for characterizing, classifying, and identifying unknowns in samples
First Claim
Patent Images
1. A method for characterizing an unknown sample, wherein said sample is modeled with a plurality of descriptors, comprising the steps of:
- a. obtaining a plurality of responses from a multichannel instrument, said plurality of responses equal to or greater than said plurality of descriptors, wherein said plurality of responses is related to each of said plurality of descriptors; and
b. determining said plurality of descriptors from said plurality of responses.
3 Assignments
0 Petitions
Accused Products
Abstract
Disclosed is a method for taking the data generated from an array of responses from a multichannel instrument, and determining the characteristics of a chemical in the sample without the necessity of calibrating or training the instrument with known samples containing the same chemical. The characteristics determined by the method are then used to classify and identify the chemical in the sample. The method can also be used to quantify the concentration of the chemical in the sample.
-
Citations
52 Claims
-
1. A method for characterizing an unknown sample, wherein said sample is modeled with a plurality of descriptors, comprising the steps of:
-
a. obtaining a plurality of responses from a multichannel instrument, said plurality of responses equal to or greater than said plurality of descriptors, wherein said plurality of responses is related to each of said plurality of descriptors; and
b. determining said plurality of descriptors from said plurality of responses. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23)
-
-
24. A method for characterizing an unknown sample, wherein said sample is modeled with a plurality of descriptors, comprising the steps of:
-
a. obtaining a plurality of responses from a multichannel instrument, said plurality of responses equal to or greater than said plurality of descriptors, wherein the response from each channel of said multichannel instrument includes a term related to said plurality of descriptors, wherein said term related to said plurality of descriptors contains coefficients for each descriptor; and
b. determining said plurality of descriptors from said plurality of responses. - View Dependent Claims (25, 26, 27, 28, 29, 30, 31)
-
-
32. A method for characterizing an unknown sample, wherein said sample is modeled with a plurality of descriptors, comprising the steps of:
-
a. obtaining a plurality of responses from a multichannel instrument, said plurality of responses equal to or greater than said plurality of descriptors, wherein the response from each channel of said multichannel instrument includes a term related to said plurality of descriptors, wherein said term related to said plurality of descriptors contains coefficients for each descriptor;
b. Defining a matrix P containing said coefficients;
c. determining said plurality of descriptors from said plurality of responses and said matrix P. - View Dependent Claims (33, 34, 35, 36, 37, 38, 39, 40)
a. the response is from said multichannel instrument is included in matrix R where R is equal to C 10(VP+1c)M− - 1
b. said descriptors are determined from matrix V, where V is related to a term of the form {log (C−
1 R M N−
1)−
1c}PT(PPT)−
1;
c. C is a diagonal matrix of the concentrations of the vapors d. M and N are diagonal matrices of particular properties of specific channels of the detector, e. the superscript of −
1 denotes the inverse of the matrix,f. N is a diagonal matrix of the Δ
fs values of the sensors,g. c is a vector of constants, h. PT is the transpose of matrix P, j. PT(PPT)−
1 is the pseudo-inverse of P.
-
-
34. The method of claim 32 wherein
a. the response is matrix R where R is equal to C 10(VP+1c) D− - 1 F;
b. said descriptors are determined from matrix V, where V is equal to {log(C−
1 R D F−
1)−
1c}PT(PPT)−
1;
c. C is a diagonal matrix of the concentrations of the vapors d. D is a diagonal matrix of the polymer densities, e. the superscript of −
1 denotes the inverse of the matrix,f. F is a diagonal matrix of the Δ
fs values of the sensors,g. c is a vector of constants, h. PT is the transpose of matrix P, and i. PT(PPT)−
1 is the pseudo-inverse of P.
- 1 F;
-
35. The method of claim 32 wherein matrix P contains LSER coefficients determined from measurements of thermodynamic partitioning.
-
36. The method of claim 32 wherein matrix V contains solvation parameters for vapors.
-
37. The method of claim 32 wherein matrix R contains responses of acoustic wave vapor sensors with sorbent interactor layers and wherein matrix P contains LSER coefficients determined from measurements of responses of acoustic wave vapor sensors to known vapors.
-
38. The method of claim 32 further comprising the step of utilizing one or more of said descriptors to classify said unknown sample as belonging to a class of chemicals with certain properties.
-
39. The method of claim 32 further comprising the step of utilizing one or more of said descriptors to classify said unknown sample as belonging to a class of chemicals with certain structural features.
-
40. The method of claim 32 further comprising the step of comparing said descriptors to a table of descriptors of known chemicals to determine the identity of said unknown sample.
-
41. A method for characterizing an unknown sample at an unknown concentration, wherein said sample is modeled with a plurality of descriptors, comprising the steps of:
-
a. obtaining a plurality of responses from a multichannel instrument, said plurality of responses equal to or greater than said plurality of descriptors, wherein the response from each channel of said multichannel instrument includes a term related to said plurality of descriptors, wherein said term related to said plurality of descriptors contains coefficients for each descriptor;
b. defining a matrix Pa containing said coefficients and augmented by a vector of ones, c. determining said plurality of descriptors and concentration where from said plurality of responses d. wherein the response is matrix R where R is equal to 10(VaPa1c)D−
1 F;
e. said descriptors and concentration are determined from matrix Va, where Va is equal to {log(R D F−
1)−
1c}PaT(PaPaT)−
1f. Pa is defined as the matrix P augmented by a vector of ones as given in where P is a matrix containing said coefficients g. C is a diagonal matrix of the concentrations of the vapors h. D is a diagonal matrix of the polymer densities, i. the superscript of −
1 denotes the inverse of the matrixj. F is a diagonal matrix of the Δ
fs values of the sensorsk. PaT is the transpose of Pa l. PaT(PaPaT)−
1 is the pseudoinverse of Pa.- View Dependent Claims (42, 43, 44, 45, 46, 47, 48)
-
-
49. A method for characterizing an unknown sample, wherein said sample is modeled with a plurality of descriptors, comprising the steps of:
-
a. obtaining a plurality of responses from a multichannel instrument, said plurality of responses equal to or greater than said plurality of descriptors, wherein said plurality of responses is related to each of said plurality of descriptors; and
b. determining one or more of said plurality of descriptors from said plurality of responses using the method of inverse least squares to perform a regression, where an individual descriptor, y, is modeled as a weighted sum of responses according to y=Xb, where X is the measured response and b is a vector of weights, generally determined by regression b=X+y. - View Dependent Claims (50, 51, 52)
-
Specification