Krishna S. Nayak - US grants

[unreadable] DESCRIPTION (provided by applicant): Cardiac magnetic resonance imaging (MRI) shows great potential but requires substantial improvements in contrast, spatial resolution, and speed. Two recent advances of particular interest are the development of balanced steady state free precession (SSFP) imaging sequences (equivalently known as True-FISP, FIESTA, or Balanced-FFE), and the development and availability of 3 Telsa clinical scanners. Both provide substantial improvements in SNR and contrast but together, 3T and SSFP are relatively incompatible for rapid high-resolution cardiac imaging. Conventional SSFP is critically limited by banding artifacts that require the use of repetition times of 3 ms or less in the heart at 3T. This limits the achievable spatial resolution, and prevents the use of time-efficient acquisition schemes to reduce scan time. The aim of this proposal is to develop and validate a new imaging technique, which we call Wideband SSFP, that uses alternating repetition times to achieve a steady-state that provides up to double the bandwidth of conventional SSFP, with comparable SNR and contrast. This bandwidth improvement allows the use of longer imaging TRs which will supporting readouts on the order of 4 ms, to enable high spatial resolution (increased readout gradient area) and high temporal resolution (using time-efficient acquisition schemes). The goal of this proposal is to fully characterize and develop Wideband SSFP and apply it to rapid high- resolution cardiac MRI at 3 Tesla. Utilizing SNR improvements from the 3T platform and from SSFP sequences to achieve higher spatial and temporal resolution directly addresses the current limitations of clinical cardiac MRI. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Obesity is a growing epidemic in the United States, affecting multiple health outcomes in children, adolescents, and adults. Increasing data suggest that the link between obesity and poor health outcome is related to buildup of fat in specific areas (e.g. visceral fat inside the abdominal cavity) and/or infiltration of fat into the liver, pancreas, and other organs. Quantification of visceral fat and organ fat in vivo is generally performed by magnetic resonance imaging (MRI). MRI is well suited for this purpose because it is inherently three-dimensional, provides a sensitive mechanism for separating water and fat, and involves no ionizing radiation, leading to indefinite repeatability with ultra-low risk. Although MRI is becoming more frequently used in obesity research, its ability to directly quantify fat mass has not yet been developed and validated. Current measurements are either relative (e.g. fat signal fraction) or indirect (e.g. adipose tissue volume). Moreover, current MRI protocols are limited by the high-cost of magnet time, and the long and/or multiple breath-holds that are uncomfortable and reduce accuracy. The objective of this proposal is to overcome these limitations and develop and fully validate a novel MRI-based method for the rapid quantification of fat mass throughout the abdomen. Specifically, we will (1) develop signal calibration procedures that allow for the quantification of fat mass on a voxel-by-voxel basis, and (2) evaluate the accuracy of this quantification in swine, by comparing measurements of fat mass in adipose tissue, whole organs, and muscle, with post-mortem chemical analysis (the current gold standard). The SIGNIFICANCE of this proposal lies in the development of a new, promising, and potentially cost-effective tool for assessing fat mass and organ fat infiltration. The APPROACH utilizes recent advances in rapid MRI fat-water separation, calibration scans, and rigorous validation in an animal model. The INNOVATION lies in the application of novel signal models and calibration schemes for quantifying fat mass from signal intensity. The INVESTIGATORS include two MR physicists with expertise in the development of rapid MRI techniques for cardiovascular disease assessment and an established obesity researcher with expertise in the validation of body composition techniques. The ENVIRONMENT at USC provides generous access to research- dedicated imaging facilities, animal research support, and infrastructure for this translational research. PUBLIC HEALTH RELEVANCE: Abdominal obesity is a growing problem among Americans, and is linked with increased risks of fatty liver disease, heart disease, and type-2 diabetes, among other things. This proposal will develop a new non-invasive test based on magnetic resonance imaging that determines the mass and location of fat within the abdomen, including critical organs, and may potentially be a useful tool in obesity research.

PI Name: Krishna Nayak
Institution Name: University of Southern California
Proposal Title: BE-LA: Body Engineering Los Angeles
Proposal ID: 1045595

This project will establish a new GK-12 program at USC that will develop graduate fellows majoring in engineering and related disciplines into well-rounded STEM leaders of tomorrow, while introducing cutting-edge body engineering research into urban Los Angeles middle school science classrooms. Graduate fellows will be provided with training and practice in education, communication, leadership, collaboration, and cultural sensitivity that enhances their doctoral training. Each graduate fellow will be paired with a middle school science teacher, and will serve as an in-class science resource throughout the school year. Fellows will develop and deliver original lesson plans incorporating their doctoral research and the research of their faculty advisors. Topics will include: non-invasive sensing and imaging, speech articulation, hand articulation, neuromuscular control, vision and recognition, cardiovascular mechanics, nutrition and metabolism, biological and bio-compatible materials, and human-machine interaction. The intellectual merit of this project includes 1) the research theme of body engineering, which is a strength of USC, and provides an engaging vehicle for demonstrating scientific and engineering principles and introducing university research into K-12 classrooms; and 2) the development and use of powerful pedagogical structures that will help graduate fellows communicate their research to middle school students.

The broader significance of this project includes 1) development of a cadre of graduate fellows that possess the communication, leadership, and collaboration skills, and cultural sensitivity required to become STEM leaders; and 2) development of skills and interest in students at a critical stage for STEM learning.

Speech is a unique human capability. The vocal tract is the universal human instrument played with great dexterity and skill in the production of speech to convey rich linguistic and paralinguistic information. The project will enable fundamental understanding of how individuals differ in their speech articulation due to differences in shape and size of their physical vocal instrument. Knowledge of how people differ in their speech production can help create improved automatic speaker recognition, technologies important for national security. The project can inform design of technologies for robust speech-based access for all members of the population, including children, the elderly, and non-native speakers of a language. Results from the project can also assist in better understanding and treating disorders (e.g., cleft lip/palate), illness (e.g., head and neck cancer, apnea) or injury where human speech articulation is affected. The novel imaging data from 200 individuals, and associated tools, annotations and interpretations created by the interdisciplinary team will be shared broadly with the scientific community. The project will provide a unique research training opportunity for students in integrated speech science and technology.

The overarching goal of this project is to advance scientific understanding of how vocal tract morphology and speech articulation interact and explain the variant and invariant aspects of speech signal properties across talkers. Of particular scientific interest is the nature of articulatory strategies adopted by individuals in the presence of structural differences across them to achieve phonetic equivalence. Equally of interest are what aspects of, and how, vocal tract morphological differences are reflected in the acoustic speech signal, and if those differences can be estimated from speech acoustics. A crucial part of this goal is to create forward and inverse computational models that relate vocal tract details to speech acoustics toward shedding light on individual speaker differences and informing design of robust speaker recognition technologies. This project goes beyond state-of-the-art methods by focusing on direct investigation of the dynamic human vocal tract using novel imaging techniques and computational modeling to illuminate inter-speaker variability in vocal tract structure, as well as the strategies by which linguistic articulation is implemented. Using novel Magnetic Resonance Imaging with superior spatial resolution of the entire moving vocal tract that we helped develop (dynamic realtime 2D with excellent temporal resolution and accelerated volumetric 3D), the project will gather and quantify spatio-temporal details of speech production from 160 native American English covering the major dialectal regions of North America and 40 non-native speakers. The experimental, theoretical, and methodological approaches investigating the interplay between structure (shape and size) and function (dynamics of vocal-tract shaping and its acoustic consequences) can lead to new theoretical advances with improved phonetic characterizations of linguistic units that are general across speakers. It also offers the ability to explain individual specific speech patterns that can improve both understanding the scientific underpinning and creating robust automatic speaker recognition technology, enabling to determine not only that two talkers are different by the adoption of novel speaker dependent features, but also how and why they differ, by analyzing biologically-inspired details of structure and articulation.

? DESCRIPTION (provided by applicant): This project will develop a new ultra-safe technique for myocardial perfusion stress testing in humans. Rationale: A growing number of patients in the United States require frequent cardiovascular assessment, to determine the presence and significance of coronary artery disease (CAD). Non-invasive myocardial perfusion stress testing is a method of choice, but we are not able to use current tests as frequently as we would like in these patients due to ionizing radiation and/or risks associated with contrast agents. Our proposed technique will not require any ionizing radiation or any potentially toxic contrast agents, and could therefore be performed repeatedly with no incremental risk to the patient. Innovation: We will utilize arterial spin labeled magnetic resonance imaging (ASL-MRI), an established technology for measuring brain perfusion in humans and myocardial perfusion in small animals. To our knowledge, its application to clinical myocardial perfusion imaging and stress testing is novel. Current methods for human cardiac ASL are able to detect increases in blood flow due to vasodilation, but provide only single slice coverage and barely adequate measurement variability (per-segment variations of roughly 0.16 mL-blood / g-tissue / min). We will address these limitations by developing and optimizing innovative new labeling schemes that enable evaluation of all 17 left ventricular segments, while taking into account cardiac motion, pulsatile flow, and potential sources of artifact. We will also integrate several new MRI technologies, for the first time, into the cardiac ASL experiment: simultaneous multi-slice imaging, background suppression, and blood pool suppression. Approach: The new test will be developed and experimentally optimized in a porcine model of CAD. We will first develop and optimize a labeling method that is compatible with whole-heart spatial coverage and maximizes the strength of the ASL signal. We will then develop and optimize an imaging method that interrogates all 17 left ventricular segments and maximizes the ASL temporal signal-to-noise ratio. After this is completed, we will perform a clinical evaluation of the optimized ASL-MRI stress test in patients with suspected CAD, and will compare the new test with both single-photon emitted computed tomography (SPECT) and invasive coronary angiography with fractional flow reserve (FFR). We will determine test-retest reproducibility of the new test in a subset of the patients. This study will enable whole-heart ASL-MRI stress testing, optimize and validate its ability to measure myocardial perfusion reserve, and enhance its feasibility as a firs-line test for CAD screening. Broader Impact: The proposed myocardial perfusion stress test could have a broader role for screening of patients with suspected CAD, due to its safety and its potential for simplicity and low cost. The technical work, particularly the labeling and imaging schemes, are likely to have implications for kidney and liver ASL-MRI. Finally, this work will provide training projects for postdoctoral, graduate, and undergraduate engineering students and clinical fellows interested in advanced cardiovascular imaging.

AREA B: PRECISE DCE-MRI OF BRAIN TUMORS PROJECT SUMMARY This project will develop and validate an improved dynamic contrast-enhanced (DCE) MRI technique for assessing brain tumor response to therapy. Rationale: Historically, increases in tumor size or enhancement have signified tumor progression, and decreases in tumor size have signified treatment response. With the advent of novel chemotherapy agents, including immunotherapy, simple changes in size or enhancement are no longer sufficient to make treatment decisions. We believe that improved DCE-MRI methods can provide new and powerful biomarkers to image brain tumors and detect early response to therapy. This will enhance our ability to prolong survival in a higher proportion of brain tumor patients traditionally regarded as the most dire prognostic group (including recurrent high-grade glioma and metastatic melanoma), who are often left out of clinical trials due to very short life expectancy. Innovation: We propose Specially Tailored Acquisition and Reconstruction (STAR) DCE-MRI, in which acquisition and reconstruction are tailored from an estimation- theoretic point of view to create the most accurate and reproducible tracer kinetic (TK) parameter maps, unlike conventional approaches that optimize the quality of intermediate images. We will fully integrate TK models with DCE-MRI acquisition and reconstruction. Our preliminary data shows 36-fold improvement in spatial resolution and coverage compared to current techniques, with no loss of image quality in brain tumor patients; and we expect to only get better. Compared to current state-of-the-art DCE-MRI, this technique will provide three major advances: 1- exquisite (sub-1 mm isotropic) spatial resolution to quantitatively assess narrow tumor margins and small lesions, 2- whole-brain coverage including all lesions and all surrounding tissue thereby simplifying the exam, 3- robust measurement of patient specific arterial inputs which are required for accurate contrast agent kinetic modeling. Approach: We will optimize, technically validate, and clinically evaluate STAR DCE-MRI method that provides improved quantitative parametric brain maps including blood- brain barrier leakage and fractional plasma volume. Specifically, we will: 1- optimize and technically validate STAR DCE-MRI to produce accurate and reproducible TK parameter maps, 2- produce a robust clinical implementation of STAR DCE-MRI, and 3- clinically evaluate STAR DCE-MRI in patients with brain tumors, specifically those with recurrent high-grade glioma treated with an anti-angiogenic agent, and those with brain melanoma metastases treated with an immunotherapy agent. Broader Impact: Improved quantitative multi- parametric DCE-MRI has a potential role in the assessment of all neurologic diseases that have a microvascular component. This technical work, particularly leveraging specific nonlinear temporal models during image reconstruction, is likely to have implications for DCE-MRI of prostate, renal, breast, and liver tumors as well as outside of DCE-MRI. Importantly it addresses a critical unmet need in oncology in providing a robust, reproducible, high spatio-temporal resolution and high spatial coverage biomarker as a potential end point for clinical trials of novel therapeutic agents in cancer research.

This project features the development of a novel, high-performance, low-field Magnetic Resonance Imaging (MRI) system. The system is designed to enable groundbreaking research in imaging technology, as well as in the science and technology of human spoken communication and the science of sleep, that is not possible with any conventional MRI configuration. The work includes developing signal processing methods for MRI that enable rapid imaging in low signal-to-noise ratio environments. This will alleviate artifacts that occur at air-tissue interfaces, which will make it possible to track tissue boundaries with high precision during dynamic imaging studies, particularly those adjacent to air space such as the tongue, velum, and pharyngeal airway, which are not easily accessible by other instrumentation. The new MRI system will also have a much lower level of acoustic noise than conventional systems, making it possible to perform real-time conversational speech studies and studies during natural sleep. It is anticipated that the new system will advance the field of speech research, by enabling direct investigation of the dynamic human vocal tract during communication, and will support innovative research in speech technology.

Dynamic imaging research is deeply multidisciplinary and connects computational and informational sciences with physics, biology, linguistics, communication, and engineering. The deployment of a novel magnet dedicated to dynamic imaging on the host campus represents a contribution to infrastructure for research and teaching. Data from the new MRI system will be used in a course that teaches basic physical reasoning and quantitative thinking by using theoretical and empirical descriptions of the physical mechanisms of speech.
The course uses dynamic MRI data to allow students to understand speech production of vowels and consonants. The new system will allow for data of higher quality and, for the first time, acoustic recording of analyzable speech audio.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.