Project Areas (Charles Cain, Principal Investigator):

Back to top

Bubble-Based Acoustic Force Elasticity

     The main goal of this project is to develop a new technology to measure tissue elasticity by deformation with acoustic radiation force applied to femtosecond laser produced microbubbles.  First, we will study physical functional relationships by measuring bubble motion in gelatin phantoms with known elastic and viscoelastic properties. Bubbles will be placed in these specimens with photodisruption from a femtosecond pulsed laser so that bubble size and position can be controlled.  The bubble is displaced by acoustic radiation force from a high amplitude tone burst originating from one ultrasonic transducer and tracked by a broadband, low amplitude signal from a secondary higher frequency (7.5 MHz) transducer.  Relationships between physical parameters, specifically, bubble displacements and displacement time constants, can be used to determine elastic and viscoelastic properties in unknown specimens.

     After initial development, the technique will be validated on animal and then human lenses; mapping the spatial variation in these tissues. Preliminary experiments are limited to cadaverous tissues.  These results will be helpful for future research with this technique on manipulating lens elasticity for possible presbyopia correction by photodisrupting small areas of tissue in a loosely spaced grid. In another application, with further development of equipment, this technique could be modified to measure elasticity of structures within individual cells.

For more information, link to: http://bul.eecs.umich.edu/bubbles_RESEARCH/
Back to top

---

      Changes in soft tissue elasticity are usually related to pathological processes. Because of this, palpation is still widely used for diagnosis. Its efficacy, however, is limited to abnormalities located relatively close to the skin surface and it is very subjective. Ultrasound scan with mechanical compression expanded the palpation access depth, but diagnosis based on ultrasound echogenesity and geometry change is also relative and subjective. Using sensitive ultrasound speckle tracking procedures, controlled surface deformations, and quantitative reconstruction algorithms developed over the last decade, elasticity imaging has emerged as a potentially new diagnostic modality providing information about the absolute mechanical properties of internal organs. Promising techniques developed based on up-to-date results from controlled experiments with tissue phantom, excised tissue, and animal models have been applied into the clinical studies to transfer the engineering tools into the reliable clinical diagnostic tools. In parallel, attempts has been made to enhance the elasticity imaging techniques, including 3-D speckle tracking algorithm, and to cooperate with different imaging modalities such as Thermal Strain Imaging (TSI) and Pressure equalization Technique. The current researches in elasticity imaging include:

Functional assessment of Left Ventricular Mechanics with 3D Ultrasound

     The accurate quantification of  transmural left ventricular (LV) regional function is crucial for managing patients with ischemic heart disease and for delivering effective post-injury therapy. In this project, four partners from two academic institutions  and industry  work together to develop  an integrated imaging/ image analysis system that will accurately, robustly and reproducibly quantify transmural LV strain and strain rate from four-dimensional (3 spatial dimensions and time) echocardiographic (4DE) image sequences. Furthermore, these measurements are to be shown to be as accurate, robust and and reproducible as the same information derived from Magnetic Resonance Imaging (MRI) tagging data (felt by some to represent the state-of-the-art for in vivo strain analysis), but will be much more clinically accessible. The system combines estimates of i .) shape-tracked surface displacement information at the myocardial boundaries derived from B-mode images using a strategy being developed by a team lead by James Duncan, Ph.D. at Yale University  with ii.) intramural displacement information found using a phase-sensitive-correlation-based speckle tracking approach being developed by a team lead by Matthew O'Donnell at the University of Michigan. The jntramural displacements will be derived from RF signal data from an ultrasound array sensor ,  giving access to phase-sensitive, beamformed acoustic data  using an approach being developed on this BRP by a team lead by Jeff Powers, Ph.D. from Philips Medical Systems. Finally, using an integrated segmentation/deformation estimation approach   based on a biomechanical finite element model  ( also being developed at Yale), the two sets of displacement data will be combined to estimate 4D strains and strain rates.  The system will be validated, and evaluated in the context of LV transmural injury and remodeling, using both canine and human studies, in collaboration with a team  lead by Yale cardiologist Albert Sinusas , M.D.

DVT (Deep Venous Thrombosis)

     Deep venous thrombosis (DVT) remains a significant clinical problem today with over 250,000 patients per year affected. Unfortunately, the gold standard diagnostic technique, duplex venous ultrasound, can only diagnose but not characterize these clots. This inability to determine clot maturity has major implications regarding which anticoagulation therapy, with its associated morbidity and mortality, one would use in treatment. The sine qua non of clot maturity is increasing hardness of thrombi, and in this regard, there exists an ultrasound technique, known as reconstructive ultrasound elasticity imaging, that is a very sensitive and well-defined way to estimate tissue hardness. In addition, elasticity imaging has the very attractive property that it would require no change in the standard diagnostic ultrasound technique, i.e., pushing on the leg veins and simultaneously imaging. We are proposing to use elasticity imaging to determine the maturity/age of DVTs . We will attack this problem in three ways: 1) We will create theoretical models of clot in veins for optimizing speckle tracking algorithms, for modeling the vessel  boundary response to deformation when clots of varying hardness lie within a vein, and for estimating the non- linearities in Young’s modulus, the measure of hardness, as a function of strain. 2) We will study the ability of elasticity imaging to distinguish differences in clot maturity in a well-developed model of thrombosis in ligated rat inferior vena cavas . This model will be used to determine if elasticity imaging can detect day-to-day changes in thrombus hardness over a nine-day maturation period, where clots will change from acute, softer clots to subacute to chronic, hard thrombi. Further, we will correlate these hardness estimates with the clot fibrin concentration, the primary cause of hardening, over time. 3) We will validate elasticity imaging in two patient populations, one with known acute DVT in which the precise onset of thrombosis is known, and a second population with known long-standing, chronic DVT. We will accurately determine the ability of elasticity imaging to distinguish between the thrombi in these two groups. We believe that elasticity imaging is a natural solution to the clot characterization problem, and in this proposal, we will fully test the ability of the ultrasound elasticity imaging to address this important clinical issue.

Nonlinear Elasticity Imaging to Assess Vascular Compliance

     Previous attempts at non-invasive vascular elasticity measurements, including arterial wall motion estimation and pulse wave velocity (PWV) measurement, are limited in that they rely on imprecise motion estimation or indirect assessments of arterial wall motion. One factor limiting the success of previously used methods is that arteries normally distended under physiologic pressure produce only small strain. The normal arterial wall, however, is a highly non-linear elastic medium. We propose a new noninvasive arterial elasticity imaging technology to overcome these limitations by applying a lumen pressure equalization technique in combination with an ultrasound speckle tracking algorithm. The fundamental hypothesis of this research is that these technologies will establish a noninvasive arterial elastic modulus reconstruction procedure. Arterial elasticity can be accurately determined by measuring localized intramural strain with the help of sub-micron precision speckle tracking. The nonlinear characteristic will be determined over a large dynamic range of strain by applying a lumen pressure equalization technique. Using a simple least squares method, an optimized reconstruction procedure will be established to combine transverse scans for intramural strain measurements and longitudinal scans for PWV measurements. Optimized elastic modulus estimation will be correlated with pathology. Henseforth , specific aims of this research are to: 1. Characterize, over a large dynamic range of differential pressure across the wall, non-linear behavior of the peripheral arteries, including the carotid, which can be accessed with high frequency ultrasound and manipulated with force applied at the body surface. 2.Establish and validate the elastic modulus reconstruction procedure. Particular emphasis will be placed on optimizing reconstructed modulus from two different techniques by a least squares method.  3. Establish the accuracy and precision of non-invasive US elasticity imaging. The vascular compliance determined by dual (transverse scan for the strain and longitudinal scan for PWV measurement) US imaging will be compared with vascular pathology. The correlation between them will serve as an indication of the feasibility of this technique as a diagnostic tool. This dual US imaging technique with pressure equalization will provide a simple and accurate assessment of peripheral vessel compliance, including noninvasive measurements of the carotid artery.

Muscle/Nerve Elasticity

     We obtain noninvasive functional images of skeletal muscle with unprecedented temporal and spatial resolution using single-element or array-based ultrasound imaging systems.  The approach can measure internal muscle deformations to a spatial resolution of less than 50 mm and strains approaching 0.1%.  The imaging technique has been extended from small-structure ex-vivo imaging (isolated single fiber and whole muscle constructs) to large-structure in vivo imaging (muscles of the human forearm).  We have shown that fatigue is highly heterogeneous within a cross-section of skeletal muscle.  We expect to demonstrate that preferential injury to a subset of muscle fibers will lead to changes in axial strain.   We are also pursuing functional characterization of engineered muscle constructs using this technique.

     A similar approach is also applied to detect excited nerve tissue.  Other studies have demonstrated nerve motion during action potentials in invertebrates using integrated optical techniques for detection.  The scale of anticipated motion (<50 nm) tests the limits of the capability of our ultrasound system.  If proven viable for detecting action potentials in peripheral nerve preparations, the noninvasive imaging technique may be readily extended to the spinal cord and brain for detecting excited neurons with high spatial and temporal resolution.

For more information, link to: http://bul.eecs.umich.edu/elasticity_current_RESEARCH/
Back to top

---

Back to top

Integrated Forward/ Sidelooking IVUS catheter

        Intravascular ultrasound (IVUS) can guide coronary interventions in real time with the potential to significantly improve outcomes for many procedures. Coronary angiography remains the preferred guidance modality, however, because IVUS presentations are cross-sectional rather than fully three-dimensional, and lesion characterization with IVUS remains ambiguous. The central aim of this research is to explore new IVUS technologies to guide treatment of chronic total occlusion (CTO), a significant problem in interventional cardiology in which angiography is ineffective. Forward-looking, three-dimensional imaging with good lesion characterization is needed to guide these procedures. If ultrasound imaging technology can be developed to address this problem, then the number of coronary interventions using IVUS should double, leading to greatly improved outcomes for a large patient population.

      Current IVUS systems create cross-sectional images using either mechanical scanning of a single element transducer or synthetic aperture imaging with a 64 element circumferential array. We are investigating a fundamentally different approach combining arrays of capacitive micromachined ultrasound transducers (CMUT) for real-time 3-D imaging with new algorithms for plaque characterization taking advantage of the greatly improved image quality and signal to noise ratio (SNR) possible with CMUT devices, working in conjunction with Stanford University and Volcano Therapeutics, Inc.

     As part of this research we will design an IVUS catheter appropriate for coronary interventions using CMUT technology for simultaneous high resolution cross-sectional and forward-looking 3-D imaging. This catheter will be integrated into a real-time imaging system to present 3-D IVUS images and tissue composition maps derived from them. Using this system, we will test the hypothesis that arterial lesions can be characterized using 3-D IVUS images, as well as new algorithms for elasticity imaging, thermal strain imaging, and radio frequency (RF) signal analysis.

For more information, link to: http://bul.eecs.umich.edu/ivuscatheter_current_RESEARCH/
Back to top

---

Back to top

Molecular Imaging and Therapy using LIOB and Ultrasound

     The central aim of this project is to understand ultrafast light-DNC interactions in cells and tissue as monitored by high-frequency ultrasound. In particular, we use ultrasonic microscopy to monitor the photodisruption process transducing site-targeted nanoparticles into a detectable microbubble . Our short-term goal is to detect molecular agents targeted to squamous cell cancers and to monitor therapy applied to these cells. We propose to investigate two photodisruption regimes: one near threshold in which LIOB can be controlled to produce detectable microbubbles with little or no cellular injury (i.e., nondestructive); the second at a different set of optical parameters where LIOB can be highly destructive, killing labeled cells for therapy. If both regimes can be established, then ultrasonic detection of DNC-promoted LIOB can provide a sensitive tool for both site-targeted molecular imaging and molecular therapeutics. If these studies demonstrate that ultrasound can sensitively monitor LIOB operating either as a nondestructive sensor or a highly localized disruptor, we intend to pursue an integrated system for molecular imaging, molecular therapeutics, and treatment validation of squamous cell cancers, a rapidly growing and very important clinical problem.

For more information, link to: http://bul.eecs.umich.edu/liob_RESEARCH/
Back to top

---

Back to top

Optoacoustic Transduction for High Frequency Ultrasound

Over the last several years we have developed a technology to optically generate and detect ultrasound. This technology shows unique advantages over current state of the art ultrasound technology in forming high resolution and high frequency imaging arrays. These include:

1. Effective element size is determined by optical focusing therefore allowing micron scale single elements.

2. Optical imaging methods facilitate forming high element count 1-D and 2-D arrays.

3. Device Integration using optical fibers and flat optical chip technologies allows flexibility and robustness in the design of optical delivery systems from a console base to a remote transducer. These design concepts would play a significant role in the design of catheter based medical imaging devices such as intravascular ultrasound imaging (IVUS).

4. All polymer design optimizes acoustic bandwidth and reduces inter-element crosstalk.

Optical generation of ultrasound

     A short laser pulse absorbed in a highly absorptive material creates rapid heating followed by thermal expansion that results in the emission of short acoustic pulse. We have optimized this mechanism by using thin polymer films of extremely high thermal expansion coefficient. This approach produces ultrasonic intensity comparable to piezoelectric counterparts and is especially attractive for high-frequency applications.

Optical detection of ultrasound

     Optical detection of ultrasound also has been a subject of interest for several decades. Several techniques have been proposed and investigated. We have focused on the development of ultrasound detection using optical resonators as the active detector.

     Etalon ( Fabry -Perot resonator) : An etalon consists of a transparent layer coated by reflecting mirrors on both sides. Light incident from an external source undergoes multiple beam interference, producing a reflected signal intensity that depends on both the optical path length within the resonator and the optical wavelength. Acoustic displacement at the etalon surface changes the cavity length, which in turn changes the intensity of the optical signal reflected from the etalon. We have optimized the etalon structure for high frequency (up to 100MHz) ultrasound detection by using thin (<10micron) polymer films.

      Microring resonator : A microring resonator is typically designed as a closed-loop waveguide (the ring resonator) and a straight waveguide that couples light in and out of the resonator. The coupling is confined to a region where the light distributions at the two waveguides significantly overlap. Ultrasound waves stress the complex waveguides structure deforming both the ring waveguide and the coupling region. These deformations induce a wavelength shift in the resonance structure. Amplified modulation of the transmitted light at a fixed wavelength can be obtained by choosing a wavelength at the steep edge of the transmission spectrum.

For more information, link to: http://bul.eecs.umich.edu/optics_RESEARCH/
Back to top

---