# US 10,765,350 B2 · Noninvasive method for estimating glucose blood constituents

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

- Number: US 10,765,350 B2
- Document ID: US10765350B2
- Status: Granted
- Kind: B2
- Application number: 15/460,341
- Filed: 2017-03-16
- Published: 2017-06-29
- Granted: 2020-09-08
- Earliest claimed: 2014-01-14
- Assignees: Analytics For Life Inc.
- Applicants: Analytics For Life
- Inventors: Sunny Gupta; Timothy Burton; Matthew Howe-Patterson
- Web page: https://patentgrep.com/patent/US10765350B2

## CPC

- A61B 5/1455: using optical sensors, e.g. spectral photometrical oximeters
- A61B 5/0075: {by spectroscopy, i.e. measuring spectra, e.g. Raman spectroscopy, infrared absorption spectroscopy (A61B5/0071 takes precedence)}
- A61B 5/14532: {for measuring glucose, e.g. by tissue impedance measurement}
- A61B 5/14546: {for measuring analytes not otherwise provided for, e.g. ions, cytochromes}
- A61B 5/681: {Wristwatch-type devices}
- A61B 5/6898: {Portable consumer electronic devices, e.g. music players, telephones, tablet computers}
- A61B 5/7246: {using correlation, e.g. template matching or determination of similarity}
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)

## Abstract

A light-based method and technique for measuring the static and average plasma glucose concentration over a prolonged period of time. More specifically, the disclosure relates to a method that utilizes mathematical analysis of appendage mobile LED flash IR light transmittance, absorption and scattering by using high resolution mobile camera data to estimate the concentration of glucose and glycated hemoglobin (HbA1c) in millimoles per liter (mmol/L).

## Claims

### Claim 1

A method of performing blood constituent analysis, comprising:
receiving a series of captured images that contain a digital signature of absorbed, transmitted and scattered light from a soft tissue appendage;
determining a power spectral density (PSD) time series data from the digital signature;
performing a model-based analysis to transform the PSD time series data to a noiseless model,
applying a genetic algorithm to the noiseless model to estimate a level of blood constituent as a function of features that are extracted from a database of perfusion videos having associated blood panels; and
communicating the level of blood constituent in a format suitable for display.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
converting the series of captured images to CMOS sectors; and
placing the CMOS sectors into a scatter matrix, an absorption matrix, and a transmittance matrix.

### Claim 3 (depends on 1)

The method of claim 1, wherein the model-based analysis uses a modified matching pursuit (MMP) algorithm.

### Claim 4 (depends on 1)

The method of claim 1, further comprising:
determining Red-Green-Blue (RGB) values from the PSD time series data in accordance with the relationships:
R
=
∫
0
∞
⁢
d
⁢
⁢
λ
⁡
(
T
R
⁡
(
λ
)
⁢
I
⁡
(
λ
)
)
G
=
∫
0
∞
⁢
d
⁢
⁢
λ
⁡
(
T
G
⁡
(
λ
)
⁢
I
⁡
(
λ
)
)
B
=
∫
0
∞
⁢
d
⁢
⁢
λ
⁡
(
T
B
⁡
(
λ
)
⁢
I
⁡
(
λ
)
)
;
reconstructing an approximated PSD by forming an invertible system of equations from the relationships; and
estimating the level of blood constituent using the approximated PSD.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
recording changes in the level of blood constituent over time; and
providing an indication of a pathological state of a subject based at least in part upon the blood constituent level changes.

### Claim 6

A system for performing light-based tissue perfusion blood constituent analysis, comprising:
a capture device that captures a series of images that contain a digital signature of absorbed, transmitted and scattered light from a soft tissue appendage;
a processor that:
receives the series of images and determines power spectral density (PSD) time series data from the digital signature;
performs a model-based analysis to transform the PSD time series data to a noiseless model;
applies a genetic algorithm to the noiseless model to estimate a level of blood constituent as a function of features that are extracted from a database of perfusion videos having associated blood panels; and
a display that displays the level of blood constituent.

### Claim 7 (depends on 6)

The system of claim 6, wherein the processor converts the series of captured images data to CMOS sectors, and places the CMOS sectors into a scatter matrix, an absorption matrix, and a transmittance matrix.

### Claim 8 (depends on 6)

The system of claim 6, wherein the capture device includes a light source that is disposed adjacent to an objective lens of the capture device, the light source transmitting a plurality of wavelengths to enhance the absorption, scattering and transmittance patterns of a blood constituent of interest.

### Claim 9 (depends on 6)

The system of claim 6, wherein capturing the digital signature of absorbed, transmitted and scattered light is performed using a series of CMOS images collected at 15 to 120 frames per second.

### Claim 10 (depends on 6)

The system of claim 6, wherein the processing unit further:
determines Red-Green-Blue (RGB) values from the PSD time series data in accordance with the relationships:
R
=
∫
0
∞
⁢
d
⁢
⁢
λ
⁡
(
T
R
⁡
(
λ
)
⁢
I
⁡
(
λ
)
)
G
=
∫
0
∞
⁢
d
⁢
⁢
λ
⁡
(
T
G
⁡
(
λ
)
⁢
I
⁡
(
λ
)
)
B
=
∫
0
∞
⁢
d
⁢
⁢
λ
⁡
(
T
B
⁡
(
λ
)
⁢
I
⁡
(
λ
)
)
;
reconstructs an approximated PSD by forming an invertible system of equations from the relationships; and
estimates the level of blood constituent using the approximated PSD.

### Claim 11 (depends on 6)

The system of claim 6, wherein changes in the level of blood constituent are recorded over time, and wherein an indication of a pathological state of a subject is provided at least in part upon the blood constituent level changes.

### Claim 12

A method for estimating blood constituent analysis, comprising:
capturing a series of images using a mobile device having a camera that includes a digital signature of absorbed, transmitted and scattered light from a soft tissue appendage;
obtaining a power spectral density (PSD) time series data from the series of images;
performing a model-based analysis to transform the PSD time series data to a noiseless model;
applying a genetic algorithm to the noiseless model to estimate a level of blood constituent as a function of features that are extracted from a database of perfusion videos having associated blood panels; and
presenting the level of blood constituent in a display of the mobile device.

### Claim 13 (depends on 12)

The method of claim 12, further comprising:
converting the series of captured images to CMOS sectors; and
placing the CMOS sectors into a scatter matrix, an absorption matrix, and a transmittance matrix.

### Claim 14 (depends on 12)

The method of claim 12, further comprising illuminating the soft tissue appendage using a light source that is disposed adjacent to an objective lens of the camera, the light source transmitting a plurality of wavelengths to enhance the absorption, scattering and transmittance patterns of a blood constituent of interest.

### Claim 15 (depends on 14)

The method of claim 14, wherein the light source is an LED or a laser diode.

### Claim 16 (depends on 12)

The method of claim 12, wherein the model-based analysis uses a modified matching pursuit (MMP) algorithm.

### Claim 17 (depends on 12)

The method of claim 12, further comprising capturing the digital signature of absorbed, transmitted and scattered light using a series of CMOS images collected at 15 to 120 frames per second.

### Claim 18 (depends on 12)

The method of claim 12, further comprising:
determining Red-Green-Blue (RGB) values from the PSD time series data in accordance with the relationships:
R
=
∫
0
∞
⁢
d
⁢
⁢
λ
⁡
(
T
R
⁡
(
λ
)
⁢
I
⁡
(
λ
)
)
G
=
∫
0
∞
⁢
d
⁢
⁢
λ
⁡
(
T
G
⁡
(
λ
)
⁢
I
⁡
(
λ
)
)
B
=
∫
0
∞
⁢
d
⁢
⁢
λ
⁡
(
T
B
⁡
(
λ
)
⁢
I
⁡
(
λ
)
)
;
reconstructing an approximated PSD by forming an invertible system of equations from the relationships; and
estimating the level of blood constituent using the approximated PSD.

### Claim 19 (depends on 12)

The method of claim 12, further comprising:
recording changes in the level of blood constituent over time; and
providing an indication of a pathological state of a subject based at least in part upon the blood constituent level changes.

### Claim 20

A device, comprising:
a memory that stores computer readable instructions;
a camera that includes a complementary metal-oxide semiconductor (CMOS) sensor; and one or more processors coupled to the memory, the one or more processors executing the computer-readable instructions to:
capture, using the camera, a series of images having a digital signature of absorbed, transmitted and scattered light from a soft tissue appendage;
obtain power spectral density (PSD) time series data from the series of images;
perform a model-based analysis to transform the PSD time series data to a noiseless model;
apply a genetic algorithm to the noiseless model to estimate a level of blood constituent as a function of features that are extracted from a database of perfusion videos having associated blood panels; and
present the level of blood constituent in a display of the device.

### Claim 21 (depends on 20)

The device of claim 20, wherein the device further comprises a light source that is disposed adjacent to an objective lens of the camera, the light source transmitting a plurality of wavelengths to enhance the absorption, scattering and transmittance patterns of a blood constituent of interest.

### Claim 22 (depends on 20)

The device of claim 20, wherein the digital signature of absorbed, transmitted and scattered light is captured by using a digital video made up of a series CMOS images collected at 15 to 120 frames per second.

### Claim 23 (depends on 20)

The device of claim 20, wherein the one or more processors determine Red-Green-Blue (RGB) values from the PSD time series data in accordance with the relationships:
R
=
∫
0
∞
⁢
d
⁢
⁢
λ
⁡
(
T
R
⁡
(
λ
)
⁢
I
⁡
(
λ
)
)
G
=
∫
0
∞
⁢
d
⁢
⁢
λ
⁡
(
T
G
⁡
(
λ
)
⁢
I
⁡
(
λ
)
)
B
=
∫
0
∞
⁢
d
⁢
⁢
λ
⁡
(
T
B
⁡
(
λ
)
⁢
I
⁡
(
λ
)
)
,
wherein the one or more processors reconstructs an approximated PSD by forming an invertible system of equations from the relationships, and
wherein the one or more processors estimate the level of blood constituent using the approximated PSD.

### Claim 24 (depends on 20)

The device of claim 20, wherein the device records changes in the level of blood constituent over time, and wherein the device provides an indication of a pathological state of a subject based at least in part upon the blood constituent level changes.

### Claim 25 (depends on 20)

The device of claim 20, wherein the device is a smartphone.