# US 12,762,467 B2 · Methods and systems for polymeric fingerprint analysis and identification

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,762,467 B2
- Document ID: US12762467B2
- Status: Granted
- Kind: B2
- Application number: 17/819,552
- Filed: 2022-08-12
- Published: 2024-02-15
- Granted: 2026-10-13
- Earliest claimed: 2022-08-12
- Assignees: Ul LLC
- Applicants: Ul LLC
- Inventors: Saurav Acharya; Shubhankar Mitra; Karan Sodhi; Rashid Mehdiyev; Dermot Cochran; Annie Ibrahim-Rana; Ray Andrick
- Web page: https://patentgrep.com/patent/US12762467B2

## CPC

- G06N 20/00: Machine learning
- G01N 21/3563: for analysing solids; Preparation of samples therefor
- G06F 18/22: Matching criteria, e.g. proximity measures
- G06F 18/40: Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

An example method includes receiving training data for training of a machine learning model. The training data includes a plurality of pairs of datasets. Each of the pairs of datasets includes a reference dataset and a sample dataset. The reference dataset is indicative of first results of a first plastic sample analysis and the sample dataset is indicative of second results of a second plastic sample analysis. Each of the pairs of datasets further includes an indication, for each of the pairs of datasets, that features of the sample dataset and the reference dataset are a match. The method further includes training a machine learning model based on the training data to determine matches between datasets. The method further includes receiving a new sample dataset. The method further includes determining, using the trained machine learning model, that the new sample dataset matches another dataset.

## Claims

### Claim 1

A method comprising:
receiving, by one or more processors of one or more computing devices, training data for training of a machine learning model, wherein the training data is a first portion of a total available training data and wherein the training data comprises:
a plurality of pairs of datasets, wherein each of the pairs of datasets comprises a reference dataset and a sample dataset, and wherein:
the reference dataset is indicative of first results of a first polymer sample analysis, and
the sample dataset is indicative of second results of a second polymer sample analysis; and
an indication, for each of the pairs of datasets, that features of the sample dataset and the reference dataset are a match;
training, by the one or more processors, the machine learning model based on the training data to determine matches between datasets;
receiving, by the one or more processors, a new sample dataset;
determining, by the one or more processors using the trained machine learning model, that the new sample dataset matches at least one of one or more new reference datasets, the reference dataset of one of the plurality of pairs of datasets in the training data, or the sample dataset of one of the plurality of pairs of datasets in the training data;
receiving, by the one or more processors, a second portion of the total available training data; and
determining, by the one or more processors using the trained machine learning model, whether sample datasets in the second portion of the total available training data match any of reference datasets in the second portion of the total available training data, the reference dataset of one of the plurality of pairs of datasets in the training data, or the sample dataset of one of the plurality of pairs of datasets in the training data.

### Claim 2 (depends on 1)

The method of claim 1, further comprising, in response to the new sample dataset being determined to match the new reference dataset, receiving, by the one or more processors, the new reference dataset.

### Claim 3 (depends on 1)

The method of claim 1, wherein the new sample dataset is indicative of third results of a third polymer sample analysis.

### Claim 4 (depends on 3)

The method of claim 3, wherein the first polymer sample analysis, the second polymer sample analysis, and the third polymer sample analysis are each a same type of polymer sample analysis.

### Claim 5 (depends on 4)

The method of claim 4, wherein the same type of polymer sample analysis is at least one of an infrared spectroscopy analysis, a thermogravimetric analysis, or a differential scanning calorimetry analysis.

### Claim 6 (depends on 1)

The method of claim 1, further comprising pre-processing the training data prior to training the machine learning model.

### Claim 7 (depends on 6)

The method of claim 6, wherein the pre-processing comprises extracting features from the plurality of pairs of datasets.

### Claim 8 (depends on 6)

The method of claim 6, wherein the pre-processing comprises at least one of smoothing curves represented in the plurality of pairs of datasets, removing portions of the plurality of pairs of datasets in which events were not detected, or scaling of the plurality of pairs of datasets.

### Claim 9 (depends on 1)

The method of claim 1, further comprising receiving, by the one or more processors, an input from a user via a user interface, wherein the input indicates whether matches for one or more of the sample datasets in the second portion of the total available training data match were successfully determined.

### Claim 10 (depends on 1)

The method of claim 1, further comprising sending, by the one or more processors, to a display of a user computing device, data indicative of whether a match for the new sample dataset was identified.

### Claim 11 (depends on 10)

The method of claim 10, wherein the data sent to the display further comprises: curve data configured to cause the display to show a representation of:
a first curve representative of the new sample dataset, and
a second curve representative of a dataset determined to match the new sample dataset.

### Claim 12

A system comprising:
a memory;
a display; and
at least one processor coupled to the memory, the at least one processor configured to:
store, on the memory, a trained machine learning model, wherein:
the trained machine learning model was trained with training data comprising:
a plurality of pairs of datasets, wherein each of the pairs of datasets comprises a reference dataset and a sample dataset, and wherein:
the reference dataset is indicative of first results of a first plastic sample analysis, and
the sample dataset is indicative of second results of a second plastic sample analysis; and
an indication, for each of the pairs of datasets, that features of the sample dataset and the reference dataset are a match;
receive a new sample dataset;
determine, based on the trained machine learning model, that the new sample dataset matches at least one of one or more new reference datasets, the reference dataset of one of the plurality of pairs of datasets in the training data, or the sample dataset of one of the plurality of pairs of datasets in the training data; and
send to the display data indicative of whether a match for the new sample dataset was identified, wherein the data sent to the display further comprises curve data configured to cause the display to show a representation of:
a first curve representative of the new sample dataset, and
a second curve representative of a dataset determined to match the new sample dataset.

### Claim 13 (depends on 12)

The system of claim 12, wherein the first curve and the second curve are overlaid on a same x-y graph displayed on the display.

### Claim 14 (depends on 12)

The system of claim 12, wherein the processor is further configured to receive an input from a user via a user interface, wherein the input indicates whether the determination of a match for the new sample dataset is accurate.

### Claim 15 (depends on 12)

The system of claim 12, wherein the training data is a first portion of total available training data, and further wherein the at least one processor is configured to:
receive a second portion of the total available training data; and
determine, using the trained machine learning model, whether sample datasets in the second portion of the total available training data match any of reference datasets in the second portion of the total available training data, the reference dataset of one of the plurality of pairs of datasets in the training data, or the sample dataset of one of the plurality of pairs of datasets in the training data.

### Claim 16

A non-transitory computer readable medium having instructions stored thereon that, upon execution by a computing device, cause the computing device to perform operations comprising:
receiving a dataset, wherein the dataset is a first portion of a total available dataset and wherein the dataset comprises:
a plurality of data pairs, wherein each of the data pairs comprises reference data and sample data, and wherein:
the reference data is indicative of first results of a first plastic sample analysis, and
the sample data is indicative of second results of a second plastic sample analysis; and
an indication, for each of the data pairs, that features of the sample data and the reference data are a match;
training a machine learning model using the dataset;
receiving new sample data; and
determining, using the trained machine learning model, that the new sample data matches at least one of new reference data, the reference data of one of the plurality of data pairs in the dataset, or the sample data of one of the plurality of data pairs in the dataset;
receiving a second portion of the total available dataset; and
determining, using the trained machine learning model, whether sample data in the second portion of the total available dataset match any of reference data in the second portion of the total available dataset, the reference data of one of the plurality of data pairs in the dataset, or the sample data of one of the plurality of data pairs in the dataset.

### Claim 17 (depends on 16)

The non-transitory computer readable medium of claim 16, wherein the first plastic sample analysis is at least one of an infrared spectroscopy analysis, a thermogravimetric analysis, or a differential scanning calorimetry analysis.

### Claim 18 (depends on 16)

The non-transitory computer readable medium of claim 16, wherein the first plastic sample analysis and the second plastic sample analysis are a same type of analysis.

### Claim 19 (depends on 16)

The non-transitory computer readable medium of claim 16, wherein the instructions further cause the computing device to send, to a display, data indicative of whether a match for the new sample data was identified, wherein the data sent to the display further comprises curve data configured to cause the display to show a representation of:
a first curve representative of the new sample data, and
a second curve representative of a dataset determined to match the new sample data.