# US 2023/0144611 A1 · Identifying false positive data within a set of blast exposure data

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

- Number: US 2023/0144611 A1
- Document ID: US20230144611A1
- Status: Published application
- Kind: A1
- Application number: 18/149,124
- Filed: 2023-01-02
- Published: 2023-05-11
- Earliest claimed: 2020-11-09
- Assignees: Applied Research Associates, Inc.
- Applicants: Applied Research Associates, Inc.
- Inventors: Suthee Wiri; David J. Ortley
- Web page: https://patentgrep.com/patent/US20230144611A1

## CPC

- G01L 5/14: for measuring the force of explosions; for measuring the energy of projectiles
- A42B 3/046: {Means for detecting hazards or accidents}
- G06F 16/9035: Filtering based on additional data, e.g. user or group profiles
- G06F 18/2113: by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
- G06N 20/00: Machine learning
- G06F 2218/00: Aspects of pattern recognition specially adapted for signal processing
- G01L: MEASURING FORCE, STRESS, TORQUE, WORK, MECHANICAL POWER, MECHANICAL EFFICIENCY, OR FLUID PRESSURE (weighing G01G)
- A42B: HATS; HEAD COVERINGS (headbands A41D20/00; head-scarves A41D23/00)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A method, system, and computer-readable media for identifying false positive data within a set of blast exposure data. After debiasing and filtering the blast exposure data, an algorithm identifies predetermined features within the data that may be indicative of false positive data. The predetermined features are used to calculate a false positive score and if the false positive score exceeds a predetermined score threshold, the data is flagged with a false positive flag and may be removed from the set of blast exposure data.

## Claims

### Claim 1

A method of identifying false positive data from a set of blast exposure data using at least one processor of a user device, the method comprising:
receiving, at the user device, the set of blast exposure data comprising raw pressure data over time;
determining a bias associated with the raw pressure data;
debiasing the raw pressure data according to the bias to produce debiased pressure data;
identifying, by the at least one processor, one or more features of a plurality of predefined features within the debiased pressure data,
wherein the one or more features of the plurality of predefined features are associated with one or more predefined false positive classes; and
flagging at least a portion of the set of blast exposure data based on the one or more features of the plurality of predefined features.

### Claim 2 (depends on 1)

The method of claim 1, further comprising filtering the debiased pressure data to produce filtered pressure data.

### Claim 3 (depends on 2)

The method of claim 2, further comprising:
integrating the debiased pressure data over time to produce impulse data; and
integrating the filtered pressure data over time to produce filtered impulse data.

### Claim 4 (depends on 1)

The method of claim 1, wherein the one or more predefined false positive classes comprises a noise class.

### Claim 5 (depends on 1)

The method of claim 1, wherein the one or more predefined false positive classes comprises a sensor error class.

### Claim 6 (depends on 1)

The method of claim 1, wherein the one or more predefined false positive classes comprises a physical impossibility class.

### Claim 7 (depends on 6)

The method of claim 6, further comprising:
identifying, within the set of blast exposure data, a negative impulse associated with the physical impossibility class; and
removing at least an additional portion of the set of blast exposure data relating to the negative impulse.

### Claim 8

One or more non-transitory computer-readable media storing computer-executable instructions that, when executed by at least one processor, perform a method of identifying false positive data from a set of blast exposure data, the method comprising:
receiving the set of blast exposure data comprising raw pressure data over time;
determining a bias associated with the raw pressure data;
debiasing the raw pressure data according to the bias to produce debiased pressure data;
filtering the debiased pressure data to produce filtered pressure data;
identifying one or more features of a plurality of predefined features within any of the raw pressure data, the debiased pressure data, pressure change data, the filtered pressure data, and filtered pressure change data,
wherein the one or more features of the plurality of predefined features are associated with one or more predefined false positive classes; and
flagging at least a portion of the set of blast exposure data based on the one or more features of the plurality of predefined features.

### Claim 9 (depends on 8)

The one or more non-transitory computer-readable media of claim 8, the method further comprising:
determining a false positive score associated with the set of blast exposure data based on the one or more features of the plurality of predefined features.

### Claim 10 (depends on 9)

The one or more non-transitory computer-readable media of claim 9, the method further comprising:
automatically removing at least an additional portion of the set of blast exposure data based on the false positive score.

### Claim 11 (depends on 8)

The one or more non-transitory computer-readable media of claim 8, wherein the plurality of predefined features comprises:
a maximum impulse value feature;
a head to tail difference value feature;
a maximum pressure value feature; and
a baseline-shift feature.

### Claim 12 (depends on 8)

The one or more non-transitory computer-readable media of claim 8, wherein the plurality of predefined features comprises:
a sawtooth signal feature;
a plateau signal feature; and
a sine wave signal feature.

### Claim 13 (depends on 8)

The one or more non-transitory computer-readable media of claim 8, wherein the plurality of predefined features comprises a feature associated with a section of substantially flat pressure within the set of blast exposure data.

### Claim 14

A system comprising:
one or more body-mounted blast sensors; and
at least one processor programmed to perform a method of identifying false positive data from a set of blast exposure data, the method comprising:
receiving the set of blast exposure data comprising raw pressure data over time from the one or more body-mounted blast sensors;
determining a bias associated with the raw pressure data;
debiasing the raw pressure data according to the bias to produce debiased pressure data;
identifying one or more features within the debiased pressure data,
wherein the one or more features are associated with one or more predefined false positive classes;
determining a false positive score associated with the set of blast exposure data based on the one or more features; and
if the false positive score exceeds a predetermined threshold value, flagging the set of blast exposure data.

### Claim 15 (depends on 14)

The system of claim 14, the method further comprising filtering the debiased pressure data to produce filtered pressure data.

### Claim 16 (depends on 15)

The system of claim 15, wherein the debiased pressure data is filtered using a Savitzky-Golay filter.

### Claim 17 (depends on 14)

The system of claim 14, wherein the one or more body-mounted blast sensors include:
a head-mounted blast sensor;
a shoulder-mounted blast sensor; and
a chest-mounted blast sensor.

### Claim 18 (depends on 14)

The system of claim 14, further comprising:
a machine learning model trained to identify false positive data within the set of blast exposure data.

### Claim 19 (depends on 14)

The system of claim 14, the method further comprising:
identifying a peak overpressure within the set of blast exposure data.

### Claim 20 (depends on 19)

The system of claim 19, the method further comprising:
normalizing the raw pressure data based on the peak overpressure.
