# US 10,319,574 B2 · Categorization data manipulation using a matrix-assisted laser desorption/ionization time-of-flight mass spectrometer

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

- Number: US 10,319,574 B2
- Document ID: US10319574B2
- Status: Granted
- Kind: B2
- Application number: 15/638,911
- Filed: 2017-06-30
- Published: 2018-02-22
- Granted: 2019-06-11
- Earliest claimed: 2016-08-22
- Assignees: Highland Innovations Inc.
- Applicants: Eung Joon Jo; Yohahn Jo
- Inventors: Eung Joon Jo; Yohahn Jo
- Web page: https://patentgrep.com/patent/US10319574B2

## CPC

- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- H01J 49/0418: {for laser desorption, e.g. matrix-assisted laser desorption/ionisation [MALDI] plates or surface enhanced laser desorption/ionisation [SELDI] plates}
- G06F 16/355: Creation or modification of classes or clusters
- G06N 20/00: Machine learning
- H01J 49/0036: {Step by step routines describing the handling of the data generated during a measurement}
- A61B 5/7246: {using correlation, e.g. template matching or determination of similarity}
- G06F 16/14: Details of searching files based on file metadata
- H01J 49/26: Mass spectrometers or separator tubes
- H01J 49/28: Static spectrometers
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- H01J: ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS (spark-gaps H01T; electric arc lamps H05B31/00; particle accelerators H05H)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)

## Abstract

Methods, systems, apparatuses, and/or computer programs. Mass spectrometer test data (e.g. a blood sample of a medical patient) may be associated with metadata information. The associated metadata information may be associated with the medical patient. A subset of a sample reference library may be based on the associated metadata information. The sample reference library may include a plurality of sets of mass spectrometer reference data, in accordance with embodiments. Embodiments match the mass spectrometer test data with mass spectrometer reference data of the selected subset of the sample reference library. Embodiments determine characteristic information of a source of the mass spectrometer test data (e.g. the medical patient) based on the known characteristics of the matched mass spectrometer reference data.

## Claims

### Claim 1

A method performed by at least one processor comprising:
receiving mass spectrometer test data of a sample, wherein the mass spectrometer test data is generated by a MALDI-TOF MS machine;
associating metadata information of a source of the sample to the mass spectrometer test data;
selecting a subset of a sample reference library based on the associated metadata information, wherein the sample reference library comprises a plurality of sets of mass spectrometer reference data;
matching the mass spectrometer test data with at least one set of the plurality of sets of mass spectrometer reference data of the selected subset of the sample reference library; and
determining characteristic information of the source based on the known characteristics of the matched mass spectrometer reference data,
wherein the selecting the subset of the sample reference library comprises determining overlap between the associated metadata information of the source and metadata information of the plurality of sets of mass spectrometer reference data,
wherein the metadata information of the source and the metadata information of the plurality of sets of mass spectrometer reference data comprises a plurality of categories having parameters,
wherein correlation between the plurality of categories is minimized in order to maximize independence of each of the plurality of categories,
wherein the independence of each of the plurality of categories is determined by a dynamic threshold, and
wherein the dynamic threshold for determining the independence of each of the plurality of categories is determined by at least one of artificial intelligence or deep learning algorithm.

### Claim 2 (depends on 1)

The method of claim 1, wherein:
the mass spectrometer test data has unknown characteristics; and
the plurality of sets of mass spectrometer reference data has known characteristics.

### Claim 3 (depends on 1)

The method of claim 1, wherein:
the sample comprises biological molecules;
the metadata information of the source comprises information about the source of the biological molecules; and
the characteristic information of the source comprises a biological analysis info' ration of the source.

### Claim 4 (depends on 3)

The method of claim 3, wherein the biological analysis information is a medical diagnosis of at least one of a human being, an animal, a plant, or a living organism.

### Claim 5 (depends on 3)

The method of claim 3, wherein the characterization information of the source comprises at least one disease parameter.

### Claim 6 (depends on 5)

The method of claim 5, wherein the at least one disease parameter is a correlation of disease information to the source of the biological molecules.

### Claim 7 (depends on 6)

The method of claim 6, wherein the disease information relates to a disease and the disease information comprises at least one parameter or sub-parameter relating to at least one of:
a name of the disease;
a time stage of the disease;
a type of the disease;
a sub-type of the type of the disease;
severity of the disease; or
information relating to the disease.

### Claim 8 (depends on 3)

The method of claim 3, wherein the characterization information comprises at least one non-disease parameter.

### Claim 9 (depends on 8)

The method of claim 8, wherein the at least one non-disease parameter comprises attributes of the source.

### Claim 10 (depends on 8)

The method of claim 8, wherein the at least one non-disease parameter comprises at least one of:
an indication of a sex of the source;
an indication of an age of the source;
an indication of a disease history of the source;
an indication of a family disease history of the source;
an indication of a smoking history of the source;
an indication of life habits of the source;
an indication of a drinking history of the source;
an indication of sugar levels of the source; and
indication of cholesterol levels of the source; or
information relating to the source.

### Claim 11 (depends on 8)

The method of claim 8, wherein the wherein the at least one non-disease parameter comprises at least one sub-parameter describing at least one attribute of the at least one non-disease parameter.

### Claim 12 (depends on 1)

The method of claim 1, wherein the independence of each of the plurality of categories is based on at least one of a minimum independence threshold, a correlation coefficient index, a cross correlation index, a relative standard deviation analysis, or a distinction threshold to obtain a predetermined degree of diagnostic accuracy.

### Claim 13 (depends on 1)

The method of claim 1, wherein the independence of each of the plurality of categories is determined based on a relative standard deviation value.

### Claim 14 (depends on 13)

The method of claim 13, wherein the relative standard deviation value is at least one of minimized or optimized to maximize at least one of reproducibility or accuracy.

### Claim 15 (depends on 1)

The method of claim 1, wherein the independence of each of the plurality of categories optimizes the selecting the subset of the sample library.

### Claim 16 (depends on 15)

The method of claim 15, wherein the subset of the sample library is optimized by at least one of minimizing the size of the subset of the sample library or maintaining a predetermined degree of diagnostic accuracy.

### Claim 17 (depends on 15)

The method of claim 15, wherein the subset of the sample library is optimized by eliminating irrelevant sets of the plurality of sets of mass spectrometer reference data.

### Claim 18 (depends on 17)

The method of claim 17, wherein the eliminating irrelevant sets of the plurality of sets of mass spectrometer reference data comprises clustering categories of the plurality of categories that are not sufficiently independent to minimize size of the sample library.

### Claim 19 (depends on 18)

The method of claim 18, wherein the clustering comprises:
determining at least one cross-correlation coefficient between at least two of the plurality of categories; and
if the at least one cross-correlation coefficient is greater than a predetermined threshold for the at least two of the plurality of categories, then clustering the at least two of the plurality of categories together to minimize the size of the sample library.

### Claim 20 (depends on 19)

The method of claim 19, wherein the clustering categories comprises intra-category clustering at least two subcategories of one of the plurality of categories.

### Claim 21 (depends on 20)

The method of claim 20, wherein the clustering categories comprises inter-category clustering of at least two subcategories of different categories of the plurality of categories.

### Claim 22 (depends on 21)

The method of claim 21, wherein the predetermined threshold of the at least one cross-correlation coefficient is greater for intra-category clustering than for inter-category clustering.

### Claim 23 (depends on 17)

The method of claim 17, wherein the irrelevant sets comprise sets of mass spectrometer reference data which have a risk of false positive or false negative output results based on the associated metadata information of the source.

### Claim 24 (depends on 17)

The method of claim 17, wherein the irrelevant sets comprise sets of mass spectrometer reference data which are mutually exclusive to the associated metadata information.

### Claim 25 (depends on 1)

The method of claim 1, wherein the correlation between the plurality of categories is determined from at least one of artificial intelligence or a deep learning algorithm.

### Claim 26 (depends on 21)

The method of claim 21, the correlation between the plurality of categories is minimized to optimize at least one of diagnosing, screening, or identifying a medical condition.

### Claim 27 (depends on 1)

The method of claim 1, wherein:
the mass spectrometer test data comprises a mass-to-charge profile output of a mass spectrometer;
the plurality of sets of mass spectrometer reference data comprises a plurality of mass-to-charge profiles of reference samples stored in the sample reference library; and
the matching comprises comparing the mass-to-charge profile of the mass spectrometer test data with the plurality of mass-to-charge profiles of the reference samples.

### Claim 28 (depends on 27)

The method of claim 27, wherein the matching comprises deciding to match the mass spectrometer test data with a set of the plurality of sets of mass spectrometer reference data if there is substantially similar mass-to-charge profiles.

### Claim 29 (depends on 28)

The method of claim 28, wherein the deciding to match the substantially similar mass-to-charge profiles is performed according to predetermined thresholds.

### Claim 30 (depends on 28)

The method of claim 28, wherein the deciding to match the substantially similar mass-to-charge profiles is performed according to dynamic thresholds.

### Claim 31 (depends on 30)

The method of claim 30, wherein the dynamic thresholds for the deciding to match the substantially similar mass-to-charge profiles are determined by at least one of artificial intelligence or deep learning algorithms.

### Claim 32 (depends on 1)

The method of claim 1, wherein the sample reference library is stored in at least one of the MALDI-TOF MS machine, a hard drive, or a cloud database.

### Claim 33

A system comprising:
at least one processor;
a receiving unit configured to receive mass spectrometer test data of a sample using the at least one processor;
an associating unit configured to associate metadata information of a source of the sample to the mass spectrometer test data using the at least one processor;
a selecting unit configured to select a subset of a sample reference library based on the associated metadata information using the at least one processor, wherein the sample reference library comprises a plurality of sets of mass spectrometer reference data;
a matching unit configured to match the mass spectrometer test data with at least one set of the plurality of sets of mass spectrometer reference data of the selected subset of the sample reference library using the at least one processor; and
a determining unit configured to determine characteristic information of the source based on the known characteristics of the matched mass spectrometer reference data using the at least one processor,
wherein the selecting unit selects the subset of the sample reference library by determining overlap between the associated metadata information of the source and metadata information of the plurality of sets of mass spectrometer reference data,
wherein the metadata information of the source and the metadata information of the plurality of sets of mass spectrometer reference data comprises a plurality of categories having parameters,
wherein correlation between the plurality of categories is minimized in order to maximize independence of each of the plurality of categories,
wherein the independence of each of the plurality of categories is determined by a dynamic threshold, and
wherein the dynamic threshold for determining the independence of each of the plurality of categories is determined by at least one of artificial intelligence or deep learning algorithm.

### Claim 34

A computer program product, comprising a computer readable hardware storage device having computer readable program code stored therein, said program code containing instructions executable by one or more processors of a computer system to implement a method of assessing damage to an object, said method comprising:
receiving mass spectrometer test data of a sample;
associating metadata information of a source of the sample to the mass spectrometer test data;
selecting a subset of a sample reference library based on the associated metadata information, wherein the sample reference library comprises a plurality of sets of mass spectrometer reference data;
matching the mass spectrometer test data with at least one set of the plurality of sets of mass spectrometer reference data of the selected subset of the sample reference library; and
determining characteristic information of the source based on the known characteristics of the matched mass spectrometer reference data,
wherein the selecting the subset of the sample reference library comprises determining overlap between the associated metadata information of the source and metadata information of the plurality of sets of mass spectrometer reference data,
wherein the metadata information of the source and the metadata information of the plurality of sets of mass spectrometer reference data comprises a plurality of categories having parameters,
wherein correlation between the plurality of categories is minimized in order to maximize independence of each of the plurality of categories,
wherein the independence of each of the plurality of categories is determined by a dynamic threshold, and
wherein the dynamic threshold for determining the independence of each of the plurality of categories is determined by at least one of artificial intelligence or deep learning algorithm.

### Claim 35 (depends on 27)

The method of claim 27, wherein the mass-to-charge profile comprises data in units of time-of-flight.