Systems and methods for determining oxygen consumption using magnetic resonance imaging

Abstract

Systems and methods for determining oxygen consumption using magnetic resonance imaging (MRI) are provided. In some implementations, a method includes steps for applying a pulse sequence using an MRI system to a region of interest (ROI) in a subject, and acquiring magnetic resonance (MR) signal data from the ROI using the MRI system. The method also includes steps for reconstructing the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image, and at least one blood-flow image; and analyzing the at least one motion-resolved, blood oxygenation-weighted image to determine oxygen consumption in the ROI.

Description

Attorney Docket: 065472-000999WOPTSYSTEMS AND METHODS FOR DETERMINING OXYGEN CONSUMPTION USING MAGNETIC RESONANCE IMAGINGCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of, and priority to, U. S. Provisional Patent Application No. 63 / 721,228 filed November 15, 2024, which is hereby incorporated by reference herein in its entirety.STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0002] This invention was made with government support under Grant No. HL165211 awarded by the National Institutes of Health. The government has certain rights in the invention.TECHNICAL FIELD

[0003] The present disclosure relates generally to systems and methods for diagnosing and treating disease, and more particularly, to systems and methods for determining oxygen consumption using magnetic resonance imaging (MRI).BACKGROUND

[0004] Heart failure (HF) accounts for nearly $40 billion in medical costs every year and is the most frequent cause of hospitalization. In HF pathophysiology, the depression of contractile force of the myocardium is not matched by a concomitant depression of energy consumption. This can result in uncoupling between mechanical contraction and energy expenditure of the heart, which drives systolic or diastolic dysfunction of the heart. Cardiac energy consumption provides an important indication of cardiac function, and detection of early alterations in cardiac energetics can identify patients that require early intervention to prevent HF progression and improve outcome. Yet conventional approaches looking at cardiac energy consumption often involve invasive procedures, ionizing radiation, or techniques providing indirect and / or unreliable data. Therefore, improved technologies for detecting and treating conditions responsible for HF are needed.SUMMARY

[0005] According to some implementations of the present disclosure, a method for determining oxygen consumption using magnetic resonance imaging (MRI) is provided. The method 14917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPTincludes applying a pulse sequence using an MRI system to a region of interest (ROI) in a subject, and acquiring magnetic resonance (MR) signal data from the ROI using the MRI system. The method also includes reconstructing the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image, and at least one blood-flow image, and analyzing the at least one motion-controlled, blood oxygenation-weighted image to determine oxygen consumption in the ROI.

[0006] According to some implementations of the present disclosure, a magnetic resonance imaging (MRI) system is provided. The MRI system includes a polarizing magnet configured to generate a polarizing magnetic field about a region of interest (ROI) in a subject, a plurality of gradient coils configured to apply a gradient field to the polarizing magnetic field, and a radio frequency (RF) system configured to generate an excitation field to the ROI and acquire MR image signal data therefrom. The MRI system also includes a computer system programmed to control operation of the plurality of gradient coils and RF system according to pulse sequence to apply the gradient field and excitation field, and acquire the MR signal data from the ROI, reconstruct the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image, and at least one blood-flow image, and analyze the at least one motion-controlled, blood oxygenation-weighted image to determine oxygen consumption in the ROI.

[0007] According to some implementations of the present disclosure, a system for determining oxygen consumption using magnetic resonance imaging (MRI) is provided. The system includes one or more processors configured to apply a pulse sequence using an MRI system to a region of interest (ROI) in a subject, and acquire magnetic resonance (MR) signal data from the ROI using the MRI system. The one or more processors are also configured to reconstruct the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image, analyze the at least one motion-controlled, blood oxygenation-weighted image to determine oxygen consumption in the ROI, analyze the at least one motion-controlled, blood oxygenation-weighted image to determine oxygen consumption in the ROI, and generate and provide an output indicative of oxygen consumption in the ROI.

[0008] According to some implementations of the present disclosure, a non-transitory computer-readable medium storing a set of instructions for determining oxygen consumption using magnetic resonance imaging is provided. The set of instructions includes one or more instructions that, when executed by one or more processors of a device or system, cause the device or system to apply a pulse sequence using a magnetic resonance imaging (MRI) to a24917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPTregion of interest (RO I) in a subject, and acquire magnetic resonance (MR) signal data from the ROI using the MRI system. The set of instructions also causes the device or system to reconstruct the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image and analyze the at least one motion-controlled, blood oxygenation-weighted image to determine oxygen consumption in the ROI.

[0009] The above summary is not intended to represent each implementation or every aspect of the present disclosure. Additional features and benefits of the present disclosure are apparent from the detailed description and figures set forth below.BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The disclosure, and its advantages and drawings, will be better understood from the following description of representative embodiments together with reference to the accompanying drawings. These drawings depict only representative embodiments and are therefore not to be considered as limitations on the scope of the various embodiments or claims.

[0011] FIG. 1 is a diagram of a magnetic resonance imaging (MRI) system, according to aspects of the present disclosure.

[0012] FIG. 2 is a flowchart setting forth steps of a process, according to aspects of the present disclosure.

[0013] FIG. 3 is an illustration showing a non-invasive, free-breathing imaging sequence, according to aspects of the present disclosure.

[0014] FIG. 4 is an illustration of another imaging sequence, according to aspects of the present disclosure.

[0015] FIG. 5 is an illustration showing an example data acquisition and processing pipeline, according to aspects of the present disclosure.

[0016] FIG. 6 A are examples of balanced steady-state free precession (bSSFP) images, according to aspects of the present disclosure.

[0017] FIG. 6B shows examples of reconstructed motion-controlled blood oxygenation images, according to aspects of the present disclosure.

[0018] FIG. 6C shows examples of oxygenation saturation (SbO2) maps fitted with constant blood hematocrit (HCT) and HCT calibration, according to aspects of the present disclosure.

[0019] FIG. 6D shows examples of HCT maps of short-axis mid and the cross section of coronary sinus (CS) under rest and adenosine stress condition, according to aspects of the present disclosure.34917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPT

[0020] FIG. 6E shows example graphs illustrating correlations between SbO2 and HCT, obtained in accordance with aspects of the present disclosure, and SbO2 and HCT obtained from invasive measurements.

[0021] FIG. 7 is an illustration showing left ventricle (LV) Mass and Function Analysis, CS blood flow analysis, and oxygen saturation in blood, according to aspects of the present disclosure.

[0022] FIG. 8 shows example graphs illustrating oxygen saturation difference, coronary sinus (CS) flow per gram myocardium, oxygen consumption (MvCh), and myocardial external efficiency (MEE) between a healthy group, myocardial infarction (MI) patients with ejection fraction <25%, and the MI patients with ejection fraction between 25-50%, according to aspects of the present disclosure.

[0023] FIG. 9 shows example graphs illustrating Spearman’s correlation of myocardial external efficiency and left ventricular ejection fraction (LVEF), longitudinal, radial, and circumferential strains, according to aspects of the present disclosure.DETAILED DESCRIPTION

[0024] Cardiac energy consumption provides an important indication of cardiac function. Since the heart relies almost exclusively on aerobic oxidation, conventional approaches staging any alterations in cardiac energetics rely on invasively measured myocardial oxygen consumption (MV02) using catheterization. However, invasive measurements can limit repeatable quantitative assessment of cardiac energetics, and prevent the ability to monitor disease and therapeutic efficacy over time.

[0025] Some non-invasive approaches look to surrogate markers to determine cardiac energetics (e.g., based on echocardiography or hemodynamic parameters, such as rate pressure product and tension-time index). Surrogate markers, however, do not provide direct information on tissue-specific alterations in energy metabolism, and can therefore misguide interventions. Other approaches utilize positron emission tomography (PET) and magnetic resonance spectroscopy (MRS) to measure MV02. Yet such conventional approaches have not made it into the clinical arena due to major technical / practical limitations.

[0026] Other approaches have attempted to determine cardiac energetics from magnetic resonance (MR) oximetry. For instance, transverse relaxation rate of an MR signal (T2* and T2) can provide a blood oxygenation level-dependent (BOLD) signal, which has served as a foundation of functional MRI in the brain and propelled studies in different organs in the body.44917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPTFor instance. MR-based MV02 can be determined by quantifying myocardial blood oxygenation and blood flow using Fick’s law (i.e., MV02 = myocardial oxygen extraction fraction (OEF) x myocardial blood flow (MBF)). While studies utilizing myocardial BOLD MRI have attempted to measure reginal OEF quantitatively in a preclinical setting, such methods have not translated clinically because of imaging confounders, particularly corresponding to heart-lung interfaces (e.g., susceptibility artifacts, rapidly changing cardiac motion, and difficulties of holding breaths during pharmacological stress). In addition, conventional MBF quantification approaches require gadolinium (Gd) contrast agents to quantify MBF, which is contraindicated for many HF patients due to presentation of concomitant impaired kidney function.

[0027] Also, reliable mapping has been difficult using current cardiac magnetic resonance (CMR) imaging techniques. For instance, the coronary sinus (CS) in the heart drains more than 95% of the venous blood in coronary circulation, accurate quantification of CS blood oxygenation level (SO2) and CS blood flow can serve as an access point to measure the OEF and MBF non-invasively, without ionizing radiation and exogenous contrast agent. While cardiac MR oximetry and flowmetry have advanced over the past 20 years, their application in the CS has been limited. Current technical challenges include insufficient spatial resolution and signal-to-noise from limited breath-hold time, compromised image quality from cardiac and respiratory motion, deficient accuracy and reliability due to confounders (e.g., partial volume, heart rate dependency, and inhomogeneous Bl and BO field particularly at fields greater than 3 Testa). Further, while CS is at least 3 times larger than coronary arteries in the heart, it is still a small conduit, which collapses by approximately 50 percent of its maximum size during diastole (where standard CMR acquisition are performed). In addition, the heart undergoes complex motion during respiration, which can complicate slight prescription leading to unreliable imaging positions between multiple breath holds.

[0028] Therefore, a new approach for determining oxygenation and oxygen consumption is introduced herein, which may provide single, fast, free-breathing, motion-insensitive image acquisition. In particular, systems and methods described herein provide a number of benefits and improvements over conventional approaches. For instance, accurate quantification of cardiac energy consumption can play a significant role in HF management by providing a low-risk and accurate approach for early HF detection and prediction, HF phenotype classification for targeted treatments, and enabling longitudinal monitoring of HF progression to guide the development of novel therapies. This can lead to improved outcomes, for example, through54917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPTreduced HF mortality and hospitalization. Further, non-invasive cardiac energetic quantification may allow for monitoring cardiac energetic changes during metabolism targeted therapies, and may pave the way for developing unique treatment plans using a personalized medicine approach.

[0029] In general, magnetic resonance imaging (MRI) uses nuclear magnetic resonance (NMR) phenomena to produce images. When a substance, such as tissue, is subjected to a polarizing magnetic field BO along a z direction, individual magnetic moments of protons in the substance begin to wobble (i.e., precess) about the polarizing magnetic field BO at a characteristic resonant frequency, and generate an equilibrium magnetic moment, Mz. If the substance is also subjected to an excitation magnetic field Bl in the x-y plane near the resonance frequency, the magnetic moment Mzmay be rotated into the x-y plane to produce a net transverse magnetic moment Mt. After the excitation magnetic field Bl is terminated, a magnetic resonance (MR) signal is emitted by excited protons, which may be captured and used to form an image. Typically, a region of interest (ROI) is scanned using an imaging sequence (e.g., a “scan”) that may contain a number of measurement cycles or pulse sequences in which various magnetic field gradients (Gx, Gy, and Gz) and excitation pulses may be applied to condition and capture MR signals. In particular, magnetic field gradients can distort the polarizing magnetic field locally, causing the resonant frequency of the protons to vary by position. Each measurement provides a "view," and a number of views determines the resolution of the image. The resulting set of received MR signals, or views, or k-space samples, are digitized and processed to reconstruct an image. The total scan time may be determined in part by the number of measurement cycles, or views, that are acquired for an image.

[0030] Turning to FIG. 1, a schematic diagram illustrating an example magnetic resonance imaging (MRI) system 100, in accordance with aspects of the present disclosure, is illustrated. In some embodiments, the MRI system 100 includes an imaging assembly 110 that may house at least a polarizing magnet 112 to provide a polarizing magnetic field B0, a gradient assembly 114 with gradient coils to provide magnetic field gradients (i.e., Gx, Gy, and Gz), and a radiofrequency (RF) assembly 116 with one or more RF coils to provide RF excitations. In some implementations, other RF coils (e.g., whole-body coil, coil array, and so forth) may be utilized to provide RF excitations. The imaging assembly 110 also includes an opening that provides access to a subject arranged on a table 118. The table 118 may be controlled by a positioning system 120, allowing the subject to be moved in and out of the imaging assembly 110 for imaging.64917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPT

[0031] As illustrated in FIG. 1, the MRI system 100 may include an operator workstation 122 controlling the MRI system 100, and various components therein. In some embodiments, the operator workstation 122 may include a display, one or more input devices (e.g., a keyboard, a mouse, and so forth), a processor, a processor-executable or process-accessible memory, and so forth. In some implementations, the operator workstation 122 may be used to manage operations of the MRI system 100, and can thus be configured to cause the imaging assembly 110 to perform MR imaging as described herein. For instance, the operator workstation 122 may provide an operator interface for operator to enter or select various imaging parameters and / or imaging sequences.

[0032] In some embodiments, the MRI system 100 may include a pulse sequence system 124, a data acquisition system 126, a data processing system 128, and a data storage system 130, as shown in FIG. 1. In particular, the pulse sequence system 124 may function in response to instructions provided by the operator workstation 122 to operate a magnet system 132 and an RF system 134. In some embodiments, the magnet system 132 may generate and provide gradient waveforms to the gradient coils in the gradient assembly 114, in accordance with a selected imaging sequence. The gradient waveforms excite the gradient coils to produce magnetic field gradients Gx, Gy, and Gzused for position encoding MR signals. The magnet system 132 may also control a polarizing magnetic field generated by the polarizing magnet 112.

[0033] RF waveforms may be generated and provided by the RF system 134 to the RF coil(s) in the RF assembly 116 to perform a selected imaging sequence that may include a number of pulse sequences. MR signals induced by the selected imaging sequence may be sensed by the RF coil(s), and processed by RF system 134. For instance, the RF system 134 may amplify, demodulate, filter, and digitize sensed MR signals, as directed by the pulse sequence system 124. In some embodiments, the RF system 134 may include an RF transmitter that includes various hardware for producing a wide variety of RF excitation pulses. By way of example, the RF transmitter may include a frequency synthesizer, a power amplifier, a mixer, a converter, and so forth. The RF transmitter may be responsive to a selected imaging sequence that is configured to produce various RF pulses of desired frequency, phase, and pulse amplitude waveform. The generated RF pulses may be applied to the RF coil(s) in the RF assembly 116. The RF system 134 may also include one or more RF receiver channels, each of which may include an RF preamplifier that amplifies a received MR signal, and a detector that detects and digitizes an in-phase (I) or quadrature (Q) component the received MR signal.74917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPT

[0034] In some implementations, the RF system 134 may receive physiological data acquired by a physiological system 136. For example, the physiological system 136 may receive physiological signals from various sensors connected to a subject, such as electrocardiogram (“ECG”) signals captured by electrodes, or respiratory signals from a respiratory bellows or other respiratory monitoring device. Such physiological signals may be used by the pulse sequence system 124 to synchronize, or “gate,” the performance of an imaging sequence with the subject's heartbeat or respiration.

[0035] The digitized MR signal data produced by the RF system 134 may be received by the data acquisition system 126. The data acquisition system 126 may function in response to instructions provided by the operator workstation 122 to receive real-time MR signal data, and in some implementations, provide buffer storage. In some embodiments, the data acquisition system 126 may transmit acquired MR signal data to the data processing system 128. However, some scans may require information derived from acquired MR signal data to control further performance of an imaging sequence. To this end, the data processing system 128 may be configured to produce and transmit such information to the pulse sequence system 124. For instance, in one example, “pre-scan” MR signal data may be acquired and used calibrate an imaging sequence performed by the pulse sequence system 124. In another example, “navigator” MR signal data may be acquired and used to adjust the operating parameters of the RF system 134 or the magnet system 132, or to control an order in which k-space is sampled.

[0036] The data processing system 128 may receive MR signal data from the data acquisition system 126, and process the data in accordance with instructions received from the operator workstation 122. Such processing may, for example, include one or more of the following: reconstructing one-dimensional, two-dimensional, three-dimensional images by performing a Fourier transformation of raw k-space data; performing other image reconstruction algorithms, such as iterative or backprojection reconstruction algorithms; applying filters to raw k-space data or to reconstructed images; generating functional magnetic resonance images; calculating motion or flow images; and so on. Such data processing may be carried out intermittently, periodically, or in substantially real-time (e.g., as MR signal data is acquired).

[0037] MR signal data and / or images reconstructed by the data processing system 128 may be transmitted to the operator workstation 122, where they may be stored, displayed, and / or further transmitted (e.g., to a network device 138). For example, MR signal data and / or images may be stored in a memory of the operator workstation 122. Stored MR signal data and / or images may be provided to an operator, for example, via a display. In some implementations,84917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPTMR signal data and / or images may be stored locally (e.g., in the data storage system 130), as well as remotely (e.g., a database, cloud, or other remote data storage location).

[0038] As appreciated from FIG. 1, various components of the MRI system 100 may be connected or connectable using a communication network 140 that includes various hardware providing signal and data exchange using various wired and / or wireless links. For example, the communication network 140 may include both proprietary and dedicated networks, as well as open networks, such as the internet.

[0039] While FIG. 1 illustrates a certain configuration of the MRI system 100, it may be appreciated that the MRI system 100 various modifications may be possible, including one or more additional processing devices such that various tasks corresponding to MR imaging can be performed by different processing devices. The MRI system 100 can also include one or more printers, one or more network interfaces, one or more other types of hardware, and so forth. In some implementations, a system to implement steps of methods described herein can include a control system, which may include one or more processors. In some implementations, a computer program product comprises instructions, which when executed by a computer (e.g., a control system, one or more processing devices and / or processors, etc.), carries out steps of methods described herein. The computer program product may be a non-transitory computer readable medium. In some implementations, a system to implement steps of methods described herein includes a memory device and a control system. The memory device has stored thereon machine-readable instructions. The control system includes one or more processors that are configured to execute the machine-readable instructions to carry out the steps of methods described herein.

[0040] Turning now to FIG. 2, a flowchart setting forth steps of a method 200, in accordance with aspects of the present disclosure, is illustrated. Steps of the method 200 may be carried out using any combination of suitable devices or systems, such as the MRI system 100, and / or various components therein, described with reference to FIG. 1. In some embodiments, steps of the method 200 may be implemented as instructions stored in non-transitory computer-readable media, such as a program, firmware or software, and executed by one or more general-purpose, programmed or programmable computer, processor or other computing device. In other embodiments, steps of the method 200 may be hardwired in one or more applicationspecific computer, processor, dedicated system, or module. Although the method 200 is illustrated and described as a sequence of steps, it is contemplated that the steps may be94917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPTperformed in any order or combination, need not include all illustrated steps, and may include additional steps.

[0041] The method 200 may begin at process block 202 with controlling an MRI system as described with reference to FIG. 1, for instance, to apply a pulse sequence to a region of interest in a subject. In one example, the ROI may include a portion or the entire heart of a subject. In another example, the ROI may include at least one vessel with arterial blood, or venous blood, or both, supporting an organ of the subject. As illustrated in FIG. 3, in some implementations, the pulse sequence applied by the MRI system may include one or more of (i) a T2-IR preparation sequence (e.g., an adiabatic T2-IR preparation sequence) that may be repeated at a fixed interval to introduce T1 and T2 weighting; (ii) interleaved velocity compensated / encoded-gradients during gradient echo (GRE) readouts to generate phasecontrast images; (iii) repeated acquisition of a set of central k-space lines every other repetition time (TR) to serve as a highly sampled temporal navigator for model construction; and (iv) one or more sets of interleaved Golden ratio radial GRE readout lines with fixed and / or variable flip angles, which may serve as low-rank tensor (LRT) training data for spatial sampling and correcting for Bl inhomogeneity. In some implementations, the pulse sequence may be repeated at a fixed interval. Also, in some implementations, the pulse sequence may extend over a number of cardiac and / or respiratory cycles of the subject.

[0042] Referring again to FIG.2, responsive to the pulse sequence applied by the MRI system, MR signal data from the ROI may be acquired at process block 204. In some implementations, scouting and shimming (e.g., whole-heart shimming) may be performed, along with obtaining cross-sectional images of various ROIs (e.g., aorta, pulmonary artery, and so forth). To this end, various imaging parameters may be utilized in the pulse sequence shown in FIG. 3. For example, a total scan time=4mins, delay between T2-IR preparation=2.5s; TE (T2prep time) = 0, 40, 80, 120, 160 ms; GRE readout (TE / TR=3 / 5 ms, velocity encoding (VENC)=120 cm / s, resolution: 2 ^ 2 x 6 mm3) may be utilized. In some implementations, conventional sequences may be prescribed with matching in-plane resolution (2 x 2 x 6 mm3) and balanced steadystate free precession (bSSFP) readouts. In some implementations, a MOLLI sequence with 8 inversion times (TI) with 2 Look-Locker cycles of 3 + 5 images and T2 matched preparation times (as TEs) = 0, 40, 80,120,160 ms (T2prep refocusing time(tl80)=10, 20, 30, and 40) may be applied to acquire Tl and T2 maps, respectively. Velocity-encoded images may be acquired with temporal resolution=40ms and VENC=150 cm / s. In some implementations, before a scan of the heart, venous and arterial blood may be collected from the cranial vena cava and carotid104917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPTartery. In some implementations, a T2-IR preparation sequence with 6 inter-echo spacings (tl 80), tl80-0,7.5, 15, 22.5, 27.5, 30 ms may be utilized, as illustrated in FIG. 4.

[0043] Referring again to FIG. 2, MR signal data acquired at process block 204 may be reconstructed using a reconstruction algorithm to derive one or more motion-controlled, blood oxygenation-weighted images, as indicated by process block 206. By way of example, k-space lines may be first be modeled as a partially separable low-rank tensor may include a core tensor C and different tensor dimensions: space (Ux), cardiac motion (Uc), respiratory motion (Ur), phase-contrast (Up), Tl recovery time (Uzl), and T2 decay time (Uzt). Each U may include a number of basic functions for a corresponding dimension.

[0044] A tensor-based model may be constructed, with <!>=C(Uc0UT0Ur0Up0Uzl0Uzt)Tbeing estimated from a subspace of LTR training data, as described, and being fit to a remainder of sparsely sampled MR signal data to recover Ux =argmin\\d — E(Ux<&) || \ + R(Ux), where d is measured MR signal data, E(-) describes multichannel MR encoding and sampling, and R(-) is a regularization term. Complete tensors of all sub-spaces may be recovered from a frequently sampled navigator signal using an LRT framework. Subsequently, various cardiac and respiratory phase-controlled, Tl, T2, and / or velocity images or maps may be generated from data corresponding to respective sub-spaces (as shown in the example of FIG. 5).

[0045] The image(s) or map(s) generated at process block 206 may then be analyzed to determine oxygen consumption in a ROI, as indicated by process block 208. To this end, in some implementations, delineations of various ROIs may be made with magnitude images, phase-contrast images, or a combination thereof. For example, one or more targeted vessel, as well as other ROIs, may be delineated manually and / or automatically using various software packages. In particular, phase-contrast images may be corrected for Maxwell and gradient field to a second order, and a slowly varying phase background may be fitted to remove linear as well as non-linear phase variations, for instance, induced by eddy currents. In some implementations, one or more images may be analyzed using a validated analysis package, such as CVI42. The flow in a targeted vessel may be obtained as a product between mean velocity in the targeted vessel at each cardiac phase and the area of the targeted vessel. The flow curve may be integrated over a respiration rate (RR) interval to provide an average. Flow in the target vessel may be multiplied by heart rate to derive flow per minute.

[0046] In some implementations, a blood oxygenation level may be derived using the image(s) or map(s) generated at process block 206. For instance, a Luz-Meiboom (L-M) model may be114917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPTutilized, where the model characterizes a relationship emulating proton exchange between red blood cell and free-water in the blood and describe blood T2 as:7TL2b = ^ + (P>l)(l - M)Ted(1- ^ 100% / )A“,» J}2 x( X1- T1802Tex / Eqn (1) where T2b is the apparent blood T2, PA is the portion of protons in the plasma (100%-Hct), AwO is the resonance frequency difference between deoxy- and oxyhemoglobin, Texis the proton exchange time, nso is the time between refocusing pulses, and T20 is the base T2 with fully oxygenated blood. In addition, self- calibration MR oximetry have further simplified the MR oximetry steps with multi-parametric (T1 and T2) acquisition and variable spin refocusing time ( so). In some aspects, simultaneously acquired T1 and T2 maps with variable nso may be used to calibrate underlying parameters in the L-M model and derive SO2 and Het.

[0047] For instance, in some implementations, a multiparametric fit based on the five T2 weighted images with different T2prep times and nso may be applied to determine the underlying parameters in the L-M model. Subsequentially, simultaneously acquired T1 maps may be to fit the Het according to:1 Het 1-Hct >z-x— = - - - - -I - Eqn. (2)T1 T\,ery+rli(l SO2) 1\p[aswhere Ti,eryand Ti,pias are the T1 relaxation time of erythrocytes at derived SO2 and plasma, rl’ is the longitudinal relativity of deoxygenated hemoglobin. The final SO2 and Het may then be calculated through an iterative approach.

[0048] In some implementations, a report may be generated and provided at process block 208. The report may be in any form (e.g., graphics, graph, table, image, listing, and so, forth) and include any signals, data, and information. In one example, the report may be indicative of oxygenation and oxygen consumption in a region of interest (e.g., heart).

[0049] In one example, healthy pigs (N = 10) were studied under rest and adenosine stress to validate an approach, in accordance with aspects of the present disclosure. In particular, a continuous Radial GRE T2Prep-IR sequence with flow compensation and water excitation was prescribed in a 3T clinical scanner to the mid left ventricle (LV) and the cross-section of the coronary sinus(CS) slices of the heart to obtain the corresponding SbCE and the HCT for both oxygenated and de-oxygenated blood (TE / TR = 3.3 / 5.8ms, FA = 5°, FOV = 270mm, Voxel size= 1.3*1.3*2.7mm3, BW = 1093Hz / pixel, T2Prep duration = 0,30,60,90,110,120ms). These parameters are examples, and not limiting. For instance, in some variation, increase flip angle numbers may be used in order to improve T1 estimate accuracy.124917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPT

[0050] The raw data were then reconstructed by a Luz-Meiboom (L-M) blood oxygenation formulated LRT model to generate motion-controlled HCT & SbCh weighted images. After reconstruction, HCT and SbCh maps are both generated through least squares fitting. The results then are compared to the invasive blood sampling.

[0051] Non-invasive CMR HCT and SbCh were estimated and compared to the invasive ground truth. Representative images of the self-calibration effect of HCT are displayed in FIGs. 6A-6E. In particular, FIG. 6A shows anatomical bSSFP images of LV, RV, and CS for a representative animal. FIG. 6B shows reconstructed motion-controlled blood oxygenation weighted images. FIG. 6C shows SbCh map fitted with constant HCT (HCT=40%) is compared to the SbCh map fitted with HCT calibration. Based on the invasive measurements, SbCh maps with HCT calibration show lesser error (2.41± 1.46%) than the one with constant HCT (9.63±7.45%), indicating that HCT calibration is necessary for accurate SbCh estimation. Representative SbO2 and Het maps during rest and stress are presented in FIG. 6D. SbCh differences between the LV, RV, and CS blood (LV = 96.02±2.09%, RV = 75.43±3.81%, CS = 52.22±8.23%) are apparent in all images while HCT stays homogeneous. Significant CS SbCh elevation is presented during adenosine infusion, showing the accurate measurement of CS SbO2 changes during vasodilation (CS at rest = 52.22±8.23%, CS at stress = 75.76±5.10%).

[0052] FIG. 6E. plots a comparison between CMR estimations and invasive measurements. Strong linear correlation is found for both SbCh (Spearman’s r = 0.983, p- value < 0.05) and HCT (Spearman’s r = 0.806, - value < 0.05). The accurate noninvasive estimation of SbO2 and HCT can not only facilitate reliable assessment of myocardial oxygenation but can also be used to boost the reliability of HCT-sensitive modalities (e.g., extracellular volume assessment) in clinical environments, where accurate invasive HCT during the scan section is hard to obtain.

[0053] As apparent from the above example, the present approach can quantify HCT and SbO2 simultaneously without need for invasive procedures. Such approach can provide an accessible, risk-free assessment of myocardial metabolism for patients with potential myocardial metabolic impairments.

[0054] In yet another example, a prospective human study in a single institution investigated 24 subjects (13 HF patients with ejection fraction <50% caused by MI and 15 healthy subjects without known cardiac diseases). A continuous Radial GRE Prep-IR research sequence with flow compensation and water excitation was prescribed at the heart’s mid-slice and the crosssection of the CS to acquire images of the arterial and venous blood (TE / TR=3.3 / 5.8ms, FA=5°, FOV=270mm, Voxel=1.3xl.3x2.7 mm3, BW=1093 Hz / pixel, T2 prep duration =134917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPT0,30,60,90,110,120ms). The images were then reconstructed using a blood oxygenation formulated LRT model to generate motion-controlled SbO2 maps. A clinical 2D phase contrast gradient echo sequence was prescribed for coronary sinus flow measurement (TE / TR=2.9 / 41.2ms, flip angle=20°, 20 cardiac phases per slice, slice thickness=6 mm, voxel=1.3xl.3x6 mm3, FOV=270 cm) and 2D bSSFP cine short-axis and long-axis images covering the whole LV for LV mass and function analysis (TR=~4.5 ms, TE=~1.5 ms, flip angle=45°, number of excitations=l, 20 cardiac phases per slice, FOV=270mm, matrix=256><128, and 8mm slice thickness without spacing). Two subgroups (EF <25% and EF 25-50%) were categorized within the HF patients. Oxygen saturation difference (arterial SbO2-venous SbO2), Mv02, CS blood flow, radial, circumferential, longitudinal direction strains, and MEE were measured and compared in healthy and HF subjects.

[0055] The illustration for the study flow chart is demonstrated in FIG. 7. The physiological data of the studied subjects is presented in Table 1 below.Table 1. The Physiological data of the studied subjectsControl HF with EF <25% HF with EF 25-50% P valueEF 25% vs EF <25% vs Control vs n= 11 n= 6 n= 7EF 25-50% Control EF 25-50% Age 63.55 (8.03) 48.14 (11 26) 54.57 (15.36) 0308 0031 0308 GenderMale 9 (81 82) 6 (85.71) 6 (85.71) 0956 0956 0956 Female 2 (18 18) 1 (14.29) 1 (14.29)SBP (minHg) 134.55 (1146) 121.00 (950) 131.43 (31.02) 0767 0325 0344 DBP (nniiHg) 81.45 (9.84) 80.43 (15 13) 84.14 (19.62) 1 000 1 000 1 000 MAP (mmHg) 99.39 (7.92) 93.95 (11 94) 100.67 (22.65) 0950 0672 0950 HR (m -1) 65.81 (6.87) 68.70 (11 33) 71.13 (12.06) 1 000 1 000 1 000 LV ass 91.96 (20.03) 171.93 (62.39) 136.80 (78.91) 0347 0035 0 177 LV mass index 51.84 (9.74) 81.42 (2837) 70.45 (28.72) 0555 0060 0 129 Stroke volume 77.91 (11.90) 50.97 (21 29) 136.80 (78.91) 0215 0009 0 104 Stroke volume44.39 (7.17) 23.69 (736) 33.90 (1.17) 0002 <001 <0.01 indexLVEF 62.97 (7.64) 16.81 (6 11) 35.03 (5.80) <0.01 <001 <0.01

[0056] The oxygen saturation difference was 51.85%±7.60% forEF<25%, 51.41%±4.05% for EF 25-50%, and 47.74%±8.84% for healthy subjects. CS blood flow was 86.64±35.36 mL / min, 56.98±28.88 mL / min, and 67.28±28.68 mL / min, respectively. Mv02 values were 8.53%±4.53%, 5.57%±2.17%, and 6.61%±3.35%, with no intergroup differences (all P>0.05, as shown in panels A-C in FIG. 8).4917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPT

[0057] Myocardial external efficiency (MEE) was significantly lower in the EF<25% group compared to the EF 25-50% (20.38%±8.93% vs 60.11%±22.94%, P=0.033) and the healthy group (20.38%±8.93% vs 70.26%±47.00%, P=0.004), with no difference between the EF 25-50% and healthy groups (P= 1.000, panel D in FIG. 8). MEE correlates with EF (r=0.608, panel A in FIG. 9), the longitudinal strain (r= -0.455, panel B in FIG. 9), the radial strain (r=0.495, panel C in FIG. 9), and the circumferential strain (r= -0.495, panel D in FIG. 9) (all P <0.05).

[0058] In this study, MEE was significantly reduced in patients with EF <25% compared to those with EF 25-50% and healthy controls. Hence HF with severely reduced systolic function may be associated with markedly impaired efficiency in converting metabolic energy into effective mechanical work, consistent with previous studies. While oxygen saturation differences, CS blood flow, and Mv02 did not show significant intergroup differences, there was a trend toward increased oxygen consumption and imbalanced oxygen utilization in HF patients, correlating with impaired heart function. The modest correlation between MEE and systolic function, as reflected by EF and the longitudinal, circumferential, and radial strains, further highlights the relationship between myocardial function and energetic efficiency, aligning with prior findings.

[0059] The present approach demonstrates clinical feasibility by enabling reliable measurement of myocardial oxygen consumption, which is conventionally only achievable through invasive blood sampling or positron emission tomography (PET) with radiotracer injection. Larger cohorts may further validate and explore the full potential of the present non-invasive technique across diverse populations and stages of heart disease.

[0060] One or more elements or aspects or steps, or any portion(s) thereof, from one or more of any of claims below can be combined with one or more elements or aspects or steps, or any portion(s) thereof, from one or more of any of the other claims or combinations thereof, to form one or more additional implementations and / or claims of the present disclosure.

[0061] One or more elements or aspects or steps, or any portion(s) thereof, from one or more of any of the claims or Alternative Implementations below can be combined with one or more elements or aspects or steps, or any portion(s) thereof, from one or more of any of the other claims or Alternative Implementations or combinations thereof, to form one or more additional implementations and / or claims of the present disclosure.

[0062] ALTERNATIVE IMPLEMENTATIONS

[0063] Alternative Implementation 1. A method for determining oxygen consumption using magnetic resonance imaging, the method comprising: applying a pulse sequence using an MRI154917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPTsystem to a region of interest (ROI) in a subject; acquiring magnetic resonance (MR) signal data from the ROI using the MRI system; reconstructing the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image, and at least one blood-flow image; and analyzing the at least one motion-controlled, blood oxygenation-weighted image to determine oxygen consumption in the ROI.

[0064] Alternative Implementation 2. The method of Alternative Implementation 1, wherein the method further comprises applying the pulse sequence to an ROI comprising at least one vessel with arterial blood, or venous blood, or both, supporting an organ.

[0065] Alternative Implementation 3. The method of Alternative Implementation 1 or Alternative Implementation 2, wherein the pulse sequence applied using the MRI system further comprises a T2-IR preparation sequence.

[0066] Alternative Implementation 4. The method of Alternative Implementation 3, further comprising repeating the pulse sequence at a fixed interval.

[0067] Alternative Implementation 5. The method of any one of Alternative Implementations 1 to 4, wherein the pulse sequence applied using the MRI system further comprises an interleaved velocity compensated and encoded-gradient sequence applied during at least one gradient echo (GRE) readout.

[0068] Alternative Implementation 6. The method of any one of Alternative Implementations 1 to 5, wherein the pulse sequence applied using the MRI system further comprises sampling a k-space at a Golden ratio.

[0069] Alternative Implementation 7. The method of any one of Alternative Implementations 1 to 6, wherein the method further comprises generating at least one phase-contrast image.

[0070] Alternative Implementation 8. The method of any one of Alternative Implementations 1 to 7, wherein the pulse sequence applied using the MRI system comprises a sequence to acquire a set of central k-space lines in a k-space every other repetition time.

[0071] Alternative Implementation 9. The method of Alternative Implementation 7, wherein the method further comprises utilizing the set of central k-space lines as a highly sampled temporal navigator to construct a low-rank tensor (LRT) model.

[0072]

[0073] 10. The method of Alternative Implementation 1, wherein the pulse sequence applied using the MRI system comprises a sequence to acquire one or more sets of interleaved Golden ratio radial GRE readout lines with fixed or variable flip angles.164917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPT

[0074] Alternative Implementation 11. The method of Alternative Implementation 10, wherein the method further comprises utilizing the one or more sets as LRT training data for spatial sampling and correcting Bl inhomogeneity and spin history.

[0075] Alternative Implementation 12. A magnetic resonance imaging (MRI) system comprising: a polarizing magnet configured to generate a polarizing magnetic field about a region of interest (ROI) in a subject; a plurality of gradient coils configured to apply a gradient field to the polarizing magnetic field; a radio frequency (RF) system configured to generate an excitation field to the ROI and acquire MR image signal data therefrom; and a computer system programmed to: control operation of the plurality of gradient coils and RF system according to pulse sequence to apply the gradient field and excitation field, and acquire the MR signal data from the ROI; reconstruct the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image, and at least one blood-flow image; and analyze the at least one motion-controlled, blood oxygenation-weighted image to determine oxygen consumption in the ROI.

[0076] Alternative Implementation 13. The MRI system of Alternative Implementation 12, wherein the pulse sequence further comprises aT2-IR preparation sequence.

[0077] Alternative Implementation 14. The MRI system of Alternative Implementation 13, wherein the computer system is further programmed to repeat the pulse sequence at a fixed interval.

[0078] Alternative Implementation 15. The MRI system of any one of Alternative Implementations 12 to 14, wherein the pulse sequence further comprises an interleaved velocity compensated and encoded-gradient sequence applied during at least one gradient echo (GRE) readout.

[0079] Alternative Implementation 16. The MRI system of any one of Alternative Implementations 12 to 15, wherein the pulse sequence further comprises a sequence sampling a k-space at a Golden ratio.

[0080] Alternative Implementation 17. The MRI system of any one of Alternative Implementations 12 to 16, wherein the computer system is further programmed to generate at least one phase-contrast image.

[0081] Alternative Implementation 18. The MRI system of any one of Alternative Implementations 12 to 17, wherein the pulse sequence further comprises a sequence to acquire a set of central k-space lines in a k-space every other repetition time.174917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPT

[0082] Alternative Implementation 19. The MRI system of Alternative Implementation 18, wherein the computer system is further programmed to generate utilize the set of central k-space lines as a highly sampled temporal navigator to construct an LRT model.

[0083] Alternative Implementation 20. The MRI system of any one of Alternative Implementations 12 to 19, wherein the pulse sequence further comprises a sequence to acquire one or more sets of interleaved Golden ratio radial GRE readout lines with fixed or variable flip angles.

[0084] Alternative Implementation 21. The MRI system of Alternative Implementation 20, wherein the computer system is further programmed to generate utilize the one or more sets as LRT training data for spatial sampling and correcting Bl inhomogeneity.

[0085] Alternative Implementation 22. A system for determining oxygen consumption using magnetic resonance imaging (MRI) comprising: one or more processors configured to: apply a pulse sequence using an MRI system to a region of interest (ROI) in a subject; acquire magnetic resonance (MR) signal data from the ROI using the MRI system; reconstruct the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image; analyze the at least one motion-controlled, blood oxygenation-weighted image to determine oxygen consumption in the ROI; and generate and provide an output indicative of oxygen consumption in the ROI.

[0086] Alternative Implementation 23. A non-transitory computer-readable medium storing a set of instructions for determining oxygen consumption using magnetic resonance imaging, the set of instructions comprising one or more instructions that, when executed by one or more processors of a device, cause the device to: apply a pulse sequence using a magnetic resonance imaging (MRI) to a region of interest (ROI) in a subject; acquire magnetic resonance (MR) signal data from the ROI using the MRI system; reconstruct the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image; and analyze the at least one motion-controlled, blood oxygenation- weighted image to determine oxygen consumption in the ROI.

[0087] While the present disclosure has been described with reference to one or more embodiments or implementations, those skilled in the art recognize that many changes may be made thereto without departing from the spirit and scope of the present disclosure. Each of these embodiments or implementations is contemplated as falling within the spirit and scope of the present disclosure. It is also contemplated that additional embodiments and184917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPTimplementations according to aspects of the present disclosure may combine any number of features from any of the embodiments or implementations described herein.194917-1090-9305 1065472-000999WOPT

Claims

Attorney Docket: 065472-000999WOPTCLAIMS WHAT IS CLAIMED IS:

1. A method for determining oxygen consumption using magnetic resonance imaging, the method comprising:applying a pulse sequence using an MRI system to a region of interest (ROI) in a subject;acquiring magnetic resonance (MR) signal data from the ROI using the MRI system; reconstructing the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image, and at least one blood-flow image; andanalyzing the at least one motion-controlled, blood oxygenation-weighted image to determine oxygen consumption in the ROI.

2. The method of claim 1, wherein the method further comprises applying the pulse sequence to an ROI comprising at least one vessel with arterial blood, or venous blood, or both, supporting an organ.

3. The method of claim 1, wherein the pulse sequence applied using the MRI system further comprises a T2-IR preparation sequence.

4. The method of claim 3, further comprising repeating the pulse sequence at a fixed interval.

5. The method of claim 1, wherein the pulse sequence applied using the MRI system further comprises an interleaved velocity compensated and encoded-gradient sequence applied during at least one gradient echo (GRE) readout.

6. The method of claim 1, wherein the pulse sequence applied using the MRI system further comprises sampling a k-space at a Golden ratio.

7. The method of claim 1, wherein the method further comprises generating at least one phase-contrast image.204917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPT8. The method of claim 1, wherein the pulse sequence applied using the MRI system comprises a sequence to acquire a set of central k-space lines in a k-space every other repetition time.

9. The method of claim 7, wherein the method further comprises utilizing the set of central k-space lines as a highly sampled temporal navigator to construct a low-rank tensor (LRT) model.

10. The method of claim 1, wherein the pulse sequence applied using the MRI system comprises a sequence to acquire one or more sets of interleaved Golden ratio radial GRE readout lines with fixed or variable flip angles.

11. The method of claim 9, wherein the method further comprises utilizing the one or more sets as LRT training data for spatial sampling and correcting Bl inhomogeneity and spin history.

12. A magnetic resonance imaging (MRI) system comprising:a polarizing magnet configured to generate a polarizing magnetic field about a region of interest (RO I) in a subject;a plurality of gradient coils configured to apply a gradient field to the polarizing magnetic field;a radio frequency (RF) system configured to generate an excitation field to the ROI and acquire MR image signal data therefrom; anda computer system programmed to:control operation of the plurality of gradient coils and RF system according to pulse sequence to apply the gradient field and excitation field, and acquire the MR signal data from the ROI;reconstruct the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image, and at least one blood-flow image; andanalyze the at least one motion-controlled, blood oxygenation-weighted image to determine oxygen consumption in the ROI.214917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPT13. The MRI system of claim 12, wherein the pulse sequence further comprises aT2-IR preparation sequence.

14. The MRI system of claim 13, wherein the computer system is further programmed to repeat the pulse sequence at a fixed interval.

15. The MRI system of claim 12, wherein the pulse sequence further comprises an interleaved velocity compensated and encoded-gradient sequence applied during at least one gradient echo (GRE) readout.

16. The MRI system of claim 12, wherein the pulse sequence further comprises a sequence sampling a k-space at a Golden ratio.

17. The MRI system of claim 12, wherein the computer system is further programmed to generate at least one phase-contrast image.

18. The MRI system of claim 12, wherein the pulse sequence further comprises a sequence to acquire a set of central k-space lines in a k-space every other repetition time.

19. The MRI system of claim 18, wherein the computer system is further programmed to generate utilize the set of central k-space lines as a highly sampled temporal navigator to construct an LRT model.

20. The MRI system of claim 12, wherein the pulse sequence further comprises a sequence to acquire one or more sets of interleaved Golden ratio radial GRE readout lines with fixed or variable flip angles.

21. The MRI system of claim 20, wherein the computer system is further programmed to generate utilize the one or more sets as LRT training data for spatial sampling and correcting Bl inhomogeneity.224917-1090-9305 1065472-000999WOPTAttorney Docket: 065472-000999WOPT22. A system for determining oxygen consumption using magnetic resonance imaging (MRI) comprising:one or more processors configured to:apply a pulse sequence using an MRI system to a region of interest (ROI) in a subject;acquire magnetic resonance (MR) signal data from the ROI using the MRI system;reconstruct the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image; analyze the at least one motion-controlled, blood oxygenation-weighted image to determine oxygen consumption in the ROI; and generate and provide an output indicative of oxygen consumption in the ROI.

23. A non-transitory computer-readable medium storing a set of instructions for determining oxygen consumption using magnetic resonance imaging, the set of instructions comprising one or more instructions that, when executed by one or more processors of a device, cause the device to:apply a pulse sequence using a magnetic resonance imaging (MRI) to a region of interest (ROI) in a subject;acquire magnetic resonance (MR) signal data from the ROI using the MRI system; reconstruct the MR signal data using a reconstruction algorithm to derive at least one motion-controlled, blood oxygenation-weighted image; andanalyze the at least one motion-controlled, blood oxygenation-weighted image to determine oxygen consumption in the ROI.234917-1090-9305 1065472-000999WOPT

Images