# US 11,026,606 B2 · Magnetic resonance imaging method to non-invasively measure blood oxygen saturation

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

- Number: US 11,026,606 B2
- Document ID: US11026606B2
- Status: Granted
- Kind: B2
- Application number: 15/416,249
- Filed: 2017-01-26
- Published: 2017-08-10
- Granted: 2021-06-08
- Earliest claimed: 2016-01-26
- Assignees: Ohio State Innovation Foundation
- Applicants: Ohio State Innovation Foundation
- Inventors: Orlando P. Simonetti; Rizwan Ahmad; Lee C. Potter; Juliet Jaison Varghese
- Web page: https://patentgrep.com/patent/US11026606B2

## CPC

- A61B 5/14542: {for measuring blood gases (A61B5/14551 takes precedence)}
- A61B 5/055: involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
- G01R 33/50: based on the determination of relaxation times {, e.g. T1 measurement by IR sequences; T2 measurement by multiple-echo sequences}
- G01R 33/5602: {by filtering or weighting based on different relaxation times within the sample, e.g. T1 weighting using an inversion pulse}
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G01R: MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES (indicating correct tuning of resonant circuits H03J3/12)

## Abstract

Methods for data acquisition and processing of magnetic resonance (MR) imaging to obtain the oxygen saturation (O2sat) of blood using a relationship between transverse relaxation time (T2) of blood and oxygen saturation. The method includes obtaining multiple images at various T2 preparation times. Next, non-linear curve fitting may be used to solve for arterial or venous O2sat. The disclosure provides a calibration-free method for accurate quantitative assessment of blood in the heart and deep vessels, even in locations having limited accessibility with other diagnostic techniques.

## Claims

### Claim 1

A method for determining oxygen saturation (O2sat) of blood using magnetic resonance (MR) image data, comprising:
acquiring multiple transverse relaxation time (T2) prepared source images, each of the source images having distinct pulse timing between refocusing pulses (τ180);
deriving multiple T2-measurements from the T2 prepared source images, wherein each T2 measurement is derived from a combination of source images all having the same τ180 and different number of refocusing pulses (n);
using the multiple measurements of blood T2 (T2b) within a same blood pool, wherein each blood T2 measurement generated using source images with unique τ180, yields one Luz-Meiboom model derived equation of the form:
(
1
-
%
⁢
⁢
SbO
2
1
⁢
0
⁢
0
)
2
=
(
1
T
2
⁢
b
-
1
T
2
⁢
O
)
(
H
⁢
c
⁢
t
)
⁢
(
1
-
H
⁢
c
⁢
t
)
⁢
τ
e
⁢
x
⁡
(
α
⁢
ω
0
)
2
⁢
(
1
-
2
⁢
τ
e
⁢
x
τ
1
⁢
8
⁢
0
⁢
tanh
⁡
(
τ
1
⁢
8
⁢
0
2
⁢
τ
e
⁢
x
)
)
taking T2b together with independent measurements of patient hematocrit (Hct); and together with ω0, the proton resonance frequency that is fixed for the static magnetic field strength at which the measurements are acquired; and
applying a non-linear curve fitting to a system of equations generated using multiple T2-measurements of a blood pool, each with a unique τ180, to determine all of the unknown model parameters uniquely in each patient,
wherein the unknown model parameters include at least one of % SbO2, T2O, Tex and α,
wherein % SbO2 is the unknown blood O2sat in the system of equations that correspond to the blood pool whose O2sat needs to be determined,
T2O is the T2 value of fully oxygenated blood, and
τex is the water proton exchange time between erythrocytes and plasma, and
α is a dimensionless parameter dependent on the susceptibility difference of deoxy- and oxyhemoglobin in blood.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
receiving electrocardiogram data from a patient under examination; and
triggering the acquiring of the T2 prepared source images using the electrocardiogram data.

### Claim 3 (depends on 1)

The method of claim 1, wherein ranges or probabilistic priors are enforced, during curve fitting, on the unknown quantities in the L-M model, the method further comprising:
constraining allowed ranges of at least one of % SbO2, T2O, Tex, and α; and
estimating the unknown quantities in accordance with the constraining.

### Claim 4 (depends on 1)

The method of claim 1, the deriving multiple T2-maps further comprising generating single slice maps in which a pixel value reflects T2 of blood in chambers of a heart and blood vessels of a patient from whom the T2 prepared source images were acquired.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
using a non-invasive arterial oxygen saturation (O2 sat, % SbO2) measurement, measured using non-invasive pulse oximetry during the acquisition of MR data, that corresponds to blood T2 measurements in an arterial blood pool, to create an arterial reference system of equations that are jointly processed with the system of venous (or blood pool whose O2sat is unknown) blood T2 equations to aid in the non-linear curve fitting to determine the unknown quantities of the L-M model.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
estimating the unknown O2sat for a blood pool with the system of blood T2 equations corresponding to the blood pool alone without the need for an arterial reference system of equations.

### Claim 7 (depends on 1)

The method of claim 1, wherein to aid with each blood T2 measurement, an MR image, containing the blood pool, is collected without T2 preparation (0 refocusing pulses).

### Claim 8

A method for determining oxygen saturation (O2sat) of blood using magnetic resonance (MR) image data, comprising:
obtaining multiple transverse relaxation time (T2) prepared source images, each of the T2 source images having distinct pulse timing (τ180) and number of refocusing pulses (n);
directly using a measured signal, S, an MR signal for a given blood pool in the multiple T2 prepared source images to yield
a Luz-Meiboom (L-M) model derived equation of the form:
(
1
-
%
⁢
⁢
SbO
2
1
⁢
0
⁢
0
)
2
=
(
ln
⁡
(
S
S
0
)
-
n
⁢
τ
180
-
1
T
2
⁢
O
)
(
H
⁢
c
⁢
t
)
⁢
(
1
-
H
⁢
c
⁢
t
)
⁢
τ
e
⁢
x
⁡
(
α
⁢
ω
0
)
2
⁢
(
1
⁢
2
⁢
τ
e
⁢
x
τ
1
⁢
8
⁢
0
⁢
tanh
⁡
(
τ
1
⁢
8
⁢
0
2
⁢
τ
e
⁢
x
)
)
;
taking S together with independent measurements of patient hematocrit (Hct) and ω0, the proton resonance frequency that is fixed for the static magnetic field strength at which the measurements are acquired; and
applying a non-linear curve fitting to a system of equations generated from the multiple T2 prepared source images, each with a unique τ180, to determine all of the unknown model parameters uniquely in each patient,
wherein the unknown model parameters include at least one of % SbO2, T2O, τex, α, and S0,
wherein % SbO2 is the unknown blood O2sat in the set of equations that correspond to venous blood or blood pool whose O2sat needs to be determined,
T20 is the T2 value of fully oxygenated blood,
τex is the water proton exchange time between erythrocytes and plasma,
α is a dimensionless parameter dependent on the susceptibility difference of deoxy- and oxyhemoglobin, and
S0 is the steady-state MR signal.

### Claim 9 (depends on 8)

The method of claim 8, further comprising:
receiving electrocardiogram data from a patient under examination; and
triggering the acquiring of the T2 prepared source images using the electrocardiogram data.

### Claim 10 (depends on 8)

The method of claim 8, wherein ranges or probabilistic priors are enforced, during curve fitting, on the unknown quantities in the L-M model, the method further comprising:
constraining allowed ranges of at least one of % SbO2, T2O, Tex, α and S0; and
estimating the unknown quantities in accordance with the constraining.

### Claim 11 (depends on 8)

The method of claim 8, further comprising:
using a non-invasive arterial oxygen saturation (O2 sat, % SbO2) measurement, measured using non-invasive pulse oximetry during the acquisition of MR data, that corresponds to the arterial blood signal, S, in the T2 prepared source images, to create an arterial reference system of equations that are jointly processed with the system of venous (or blood pool whose O2sat is unknown) blood MR signal equations to aid in the non-linear curve fitting to determine the unknown quantities of the L-M model.

### Claim 12 (depends on 8)

The method of claim 8, further comprising:
estimating the unknown O2sat for a blood pool with the system of T2 prepared blood MR signal equations corresponding to the blood pool alone without the need for an arterial reference system of equations.

### Claim 13 (depends on 8)

The method of claim 8, wherein each T2 prepared source image is generated using a distinct combination of n and τ180, wherein n is the number of refocusing pulses.

### Claim 14 (depends on 8)

The method of claim 8, wherein constraints are imposed on n and τ180 to limit signal dephasing effects that may affect accuracy, wherein n is the number of refocusing pulses.

### Claim 15 (depends on 8)

The method of claim 8, wherein to aid with the curve fitting, an MR image, containing the blood pool, is collected without T2 preparation (0 refocusing pulses).

### Claim 16 (depends on 15)

The method of claim 15, wherein S0 is equated to the measured signal S of the blood pool within an MR image collected without any T2 preparation.

### Claim 17 (depends on 15)

The method of claim 15, wherein the measured signal S of the blood pool within an MR image collected without T2 preparation, is used to initialize the constraints for S0.