Imaging Technology Development
Advanced Techniques for Magnetic Resonance Imaging
Stephen Riederer, Ph.D., has a long-standing research program focused on developing methods for high speed magnetic resonance imaging (MRI). His general area of investigation has been to develop acquisition techniques allowing image acquisition times in the range of sub-second to tens of seconds, depending on the application. In one implementation, continuous, sub-second imaging is done in conjunction with high speed image reconstruction and interactive control of image data acquisition, a technique called "MR fluoroscopy." The techniques can be extended to longer acquisition times in order to obtain improved spatial resolution. Currently these methods are being used to visualize the transit of contrast agents in dynamic contrast-enhanced MR angiography and to facilitate interventional procedures. Many fellow Program investigators (e.g. Richard Ehman, M.D.; Joel Felmlee, Ph.D.; Clifford Jack, Jr., M.D.; Matthew Bernstein, Ph.D.; Joel Fletcher, M.D.; and Bernard King, M.D.) have collaborated with Dr. Riederer on these projects. His current project focuses on vascular imaging of contrast agents, particularly in several anatomic areas (renal artery, carotid artery, aorta), as well as the further technical study of physics-imposed limits. Future applications include using high resolution 3D imaging of contrast agents to study differential perfusion between malignant and non-malignant regions in various organs, including breast, liver, and prostate. In each case, fluoroscopic imaging can be used to monitor contrast arrival, and the operator will then be able to trigger the acquisition of high resolution 3D data sets at the specific phases during the contrast bolus uptake. The technique of partial image updating can potentially be used to maintain the necessary high spatial resolution while still providing adequate temporal resolution. In yet another future direction it is expected that real-time interactive techniques will be further developed to enable "on-the-spot" interactive MR diagnostic imaging by the radiologist, providing a method similar to the highly interactive modality of ultrasound but with exploitation of the highly flexible and variable contrasts of MRI.
Another ongoing project of Dr. Riederer's is "Long Field of View MRA Using Continuous Table Motion." The purpose of this work is to develop means which permit MR data acquisition during continuous motion of the patient table. The specific targeted application is contrast-enhanced MR angiography of the peripheral vasculature. Because of the speed of the advancing contrast bolus, the typical maximum time that any one longitudinal level lies within the actively sampled field of view is about 20 seconds. One of his major efforts in this project is to develop means which allow acquisition of images with adequate spatial resolution within this time period. He is working on developing techniques to allow the table velocity to be altered over the course of the procedure to match the variable speed of the contrast bolus on a patient-specific basis.
Matthew Bernstein, Ph.D., a clinical medical physicist specializing in MRI, is researching and developing the Shells trajectory, a new acquisition method for MRI. The acquisition method has particular application to contrast enhanced methods and motion correction. The goal of this work is to introduce Shells acquisitions into clinical practice over the next 2 to 3 years. Two recent papers on this topic are:
- Shu Y, Riederer SJ, Bernstein MA. Three-dimensional MRI with an under-sampled spherical shells trajectory. Magnetic Resonance in Medicine. 2006 Sep;56(3):553-62.
- Shu Y, Elliott AM, Riederer SJ, Bernstein MA. Motion correction properties of the shells k-space trajectory. Magnetic Resonance Imaging. 2006 Jul; 24(6):739-49.
- Akira Kawashima, M.D., Ph.D., is Mayo Clinic's site Principal Investigator of ACRIN protocol 6659, "MR imaging and MR spectroscopic imaging of prostate cancer prior to radical prostatectomy: a prospective multi-institutional clinicopathological study." His co-investigators are Bernard King, M.D.; and C. Daniel Johnson, M.D.; Kiaran McGee, Ph.D.; Robert Myers, M.D.; and Thomas Sebo, M.D.
Motion-Compensated MR Imaging
Armando Manduca, Ph.D., is investigating and developing automatic methods for retrospective correction of motion (and other) artifacts in MRI. These methods deduce and correct for patient motion during image acquisition from the raw data alone, with no need for special pulse sequences, motion tracking, patient preparation, or specialized hardware. The improvements in image quality are applicable to cancer imaging for many tumor sites. The techniques also apply to other sources of artifacts, such as Nyquist ghost artifacts in echo-planar imaging (EPI). Dr. Manduca hopes to increase the speed and robustness of the techniques, improve the treatment of rotational and non-rigid motion, extend the techniques to a wider variety of acquisitions, develop versions optimized to specific application areas, and integrate them into the clinical environment.
Clifford Jack, M.D., has pioneered clinical functional neuroimaging as a tool for pre-surgical estimation of functional deficit. One of the major barriers to clinical application of functional magnetic resonance imaging (fMRI) is corruption of the intrinsically small signal by global head motion. For several years, Dr. Jack has been adapting and extending the adaptive motion correction techniques originally developed by Drs. Ehman and Felmlee to functional MRI. While his research currently is relating to patients with Alzheimer's-related cognitive impairment, it also has direct application for fMRI in the context of neurooncology. His research project addresses a fundamental problem common to all fMRI studies. In pursuit of improved artifact reduction in clinical fMRI, two new types of navigator echoes have been developed, 1) a spherical motion detection navigator, which can encode all six degrees of rigid body motion in a near-instantaneous snapshot; and 2) a shim navigator echo, which will be used to correct in real-time, motion-induced first order changes in the B0 field.
Magnetic Resonance Elastography
Dr. Ehman leads Mayo's magnetic resonance elastography (MRE) research efforts. By applying and imaging propagating shear waves, it is possible to generate images depicting the elastic properties of tissues. His overall goal is to develop a method for quantitative "palpation by imaging" which may be capable of detecting a variety of disease processes, including tumors before they are large enough to be detected by touch.
It has long been known that malignant tumors are often characterized by substantially different mechanical properties than surrounding normal tissue. This accounts for the efficacy of palpation as a clinical technique to detect cancer in accessible regions of the body. Indeed, most tumors of the thyroid, breast, and prostate are still first detected by this centuries-old diagnostic technique. Unfortunately, small or inaccessible lesions cannot be detected by touch, and conventional diagnostic imaging methods such as ultrasound, computed tomography (CT), and magnetic resonance imaging (MRI) do not provide information that is in any way analogous. The goal of this work is to continue to develop and validate a diagnostic imaging technique for quantitatively delineating mechanical properties of tissues. The technique applies mechanical waves to tissue and measures regional elasticity by analyzing the pattern of wave propagation. A critical component of this approach is sensitive MRI method for directly observing propagating acoustic waves in tissue, using an MRI sequence with synchronous motion-sensitizing gradients. The central hypothesis of Dr. Ehman's work is that MRE can be successfully implemented as a practical scientific and clinical tool and that it will be useful for detecting and characterizing focal and diffuse disease processes that may be difficult to investigate by other methods.
Dr. Manduca is working with Dr. Ehman, contributing in development of processing and inversion algorithms to yield images or measures of tissue elasticity. Tasks include the extension of inversion algorithms to the calculation of other material parameters such as viscosity, dispersion, and anisotropic measures of elasticity, and the development of algorithms for novel acquisitions such as transient excitation, intra-voxel phase dispersion, and beam-mode methods. The technology development aspects of this work are also being pursued in collaboration with Drs. Felmlee and Riederer.
High Field MRI
Dr. Felmlee, Matthew Bernstein, Ph.D.; John Huston, III, M.D.; Kiaran McGee, Ph.D.; Bradley Erickson, M.D., Ph.D.; and Kimberly Amrami, M.D.; are collaborating to develop and test methods of whole-body and brain MRI at three Tesla (T), which is twice the field strength of commonly-used MR scanners. The challenges have involved re-engineering pulse sequences to account for changes in tissue relaxation times, finding solutions for increased radiofrequency energy deposition, and developing radiofrequency coil designs suitable for the higher field.
Dr. Erickson is currently pursuing several projects exploring the opportunities offered by high field MRI in neurooncology. One of these studies investigates an advanced perfusion imaging method to guide stereotactic biopsy of brain tumors and seeks to determine the best technical approach for using imaging to identify the highest grade section of the tumor for biopsy.
John Huston, III, M.D., is also investigating 3T tumor neuroimaging with methods including diffusion tensor imaging, perfusion imaging, spectroscopy and magnetic resonance angiography. Two recent publications are highlighted below:
Molecular Imaging
Val Lowe, M.D., is leading a group of investigations in the use of positron emission tomography PET for tumor detection and therapy evaluation. He is the Principal Investigator in "Comparison of Novel PET/CT Imaging Agents and MR/MRS in Metastatic and High Risk Prostate Cancer" as part of the Mayo Prostate SPORE and in collaboration with Johns Hopkins. This trial will obtain pilot data relative to new imaging methods for prostate cancer. PET C11 Choline, C11 Acetate, MRS, and In111 B12 imaging will be compared. Co-investigators include Kiaran McGee, M.D.; Douglas Collins, M.D.; Brian Davis, M.D., Ph.D.; Akira Kawashima, M.D.; Bradley Kemp, Ph.D.; Michael Lieber, M.D.; and Donald Tindall, Ph.D.
Val Lowe, M.D.; Gregory Wiseman, M.D.; David Kallmes, M.D.; and the Principal Investigator from the Gene and Virus Therapy Program Angela Dispenzieri, M.D.; are engaged in "Phase I Trial of Systemic Administration of Edmonston Strain of Measles Virus, Genetically Engineered to Express NIS, with or without Cyclophosphamide." MV-NIS is an attenuated measles virus, engineered to express the human thyroidal sodium-iodide symporter (NIS). The virus is selectively oncolytic, targeting and destroying tumor cells through CD46, a membrane regulator of complement activation that is known to be overexpressed on many human malignancies including myeloma plasma cells. NIS expression in MV-NIS infected cells permits noninvasive monitoring of virus spread by serial gamma camera imaging of radioiodine uptake. In addition, the anti-neoplastic activity of the virus can be amplified by administering I-131, a potently ionizing beta emitting isotope of radioiodine. This Phase I clinical trial is evaluating the safety of MV-NIS administered intravenously to patients with advanced treatment, refractory multiple myeloma.
Dr. Lowe and Jann Sarkaria, M.D., are working using flourothymidine (FLT) in evaluation of cancer therapy through a clinical trial in early therapy assessment of epidermal growth factor receptor (EGFR) inhibitor therapy of head and neck cancer. 3-deoxy-3-[18F]-fluorothymidine (18F-FLT) is being developed as a proliferation-specific positron emission tomography (PET) radiotracer, and Mayo Clinic and others have demonstrated that inhibitors of the EGFR markedly suppress 18F-FLT tumor uptake within 48 to 72 hours of starting treatment in animal models. These data suggest that FLT PET imaging could be used to rapidly identify tumors that are responding to EGFR inhibitor treatment. Although tumor proliferation correlates more closely with FLT uptake than 2-deoxy-2-[18F]-fluoro-D-glucose (18F-FDG) uptake, 18F-FDG PET is routinely used for head and neck cancer staging. Drs. Lowe and Sarkaria seek to compare FLT and FDG PET for monitoring treatment response. If this study demonstrates the proof-of-concept that FLT PET can be used to predict response early during a course of therapy, then this approach could integrated into an individualized approach to cancer therapy.
Vitamin B12 for Nuclear Imaging
Douglas Collins, M.D., is investigating the potential use of paramagnetic and radiolabeled cobalamin-DTPA analogs as tumor-avid diagnostic and therapeutic agents. These novel agents, based on vitamin B12 (Cobalamin) may offer useful imaging properties and low toxicity. Preliminary studies have achieved in vivo imaging of cobalamin metabolism in numerous transplanted murine and human neoplasms, as well spontaneous animal carcinomas, via the endocytosis of the analogs by transcobalamin receptors (TC-R) overexpressed on the cell membrane of tumors. The concept has been validated in Mayo Clinic patients (n=52) being evaluated for radiographic- or clinically-diagnosed tumors. In-111-DTPA-Adenoylcobalamin images multiple tumor types as small as 7 millimeters via clinical scintigraphy. Overall, 94.1 percent of breast cancers (16 of17) and 90 percent of lung cancers (9 of 10) were imaged successfully with the vitamin B-12 analog. Dr. Collins is optimistic about the clinical applicability of In-111-DTPA-Adenosylcobalamin in cancer imaging with probable improved performance over existing cancer imaging methods.
Computed Tomography
Cynthia McCollough, Ph.D., is the director of the CT Clinical Innovation Center and the CT Imaging Resource in Mayo's new Center for Advanced Imaging Research. Her research program focuses on the development of new technology and clinical applications of X-ray computed tomography (CT). She is researching the implementation of dose reduction techniques in clinical practice and developing more accurate ways to quantitate the doses delivered to patients.
Dr. McCollough is a co-principal investigator on a multi-center research effort focusing on the extension of a previously validated Monte Carlo model to take into account different CT scanner geometries, beam widths and tube current modulation and to estimate dose to critical organs from various imaging protocols for voxelized patient models. Concern over the radiation doses from diagnostic imaging procedures have been ever more concerning as imaging procedures are more frequent. Their role in cancer genesis is a concern especially in cancer patients who undergo many serial imaging to monitor disease. Reducing and understanding radiation dose from diagnostic imaging is therefore of great importance in cancer care.
Some of her other collaborations in this area include working on the National Institutes of Health (NIH)-sponsored projects of Brian Bartholmai, M.D.; C. Daniel Johnson, M.D.; and Sundeep Khosla, M.D. Dr. Johnson's work is focused on extending CT colonography, for the purpose of cancer detection, to patients not having to undergo pre-procedure bowel cleansing. Dr. Bartholmai's Lung Tissue Research Consortium multi-center project and Dr. Khosla's Epidemiology of Bone Loss and Fracture projects both use volumetric high spatial-resolution CT data to quantitate and characterize pathology as part of more global efforts to explore disease mechanisms.
The work of the CT Clinical Innovation Center is highly interdisciplinary, with co-investigators from across the institution. Lifeng Yu, Ph.D., is an expert in CT reconstruction algorithms and is co-developing novel dual-energy material decomposition algorithms to perform in vivo characterization of tissue composition and density. Dr. Manduca is collaborating on advanced noise reduction algorithms. The results may be used to either increase the quality of the information obtained at current radiation dose levels or to maintain the current level of image quality while using lower radiation dose levels. Other investigators' projects cover the gamut of phantom measurements, animal experiments and human studies. Examples include Dr. McCollough's work with Erik Ritman, M.D., Ph.D., in the use of advanced CT technology to study microvascular disease. She has assisted Dr. Ritman in translating his research techniques to the dual-source CT scanner in the Center for Advanced Imaging Research. Oncological studies underway with clinical collaborators include the use of CT perfusion imaging to study tumor blood flow characteristics and the impact of anti-angiogenesis therapy. Dr. McCollough is a co-developer with Stephanie Carlson, M.D., of advanced interventional CT visualization techniques to improve the speed, accuracy, and comfort of procedures such as CT-guided tumor biopsy or ablation.
With Dr. Kawashima and colleagues, dual-energy CT imaging is being used to assist in the differentiation of renal masses from renal cysts. With Jose Pulido, M.D., a feasibility study of CT perfusion to characterize the vascularity of ocular melanomas is ongoing. With Joel Fletcher, M.D., new image processing algorithms to automatically segment and quantitate the dimensions of solid organ tumors is ongoing.
Advanced Ultrasound Imaging
James Greenleaf, Ph.D., is involved in developing ultrasound-based techniques for measuring material properties of tissue such as stiffness and elasticity and viscosity. These developments are clearly relevant to may areas of cancer imaging.
Dr. Fatemi is a pioneer in developing novel imaging and tissue characterization techniques. He is currently the PI on three NIH-funded projects dealing with cancer imaging. One line of research is to develop a clinical vibro-acoustography system for general abdominal imaging. This system is designed as a stand-alone system that can be used by clinicians to acquire 3D images of tissue. Another avenue he is pursuing is to develop a vibro-acoustography system for detection of breast masses and micro-calcifications. Dr. Fatemi is also leading a project to develop a new prostate imaging modality and testing the system on human subjects. The new system may be used as a diagnostic tool to detect prostate masses and calcifications, as well as to guide minimally-invasive procedures, such as prostate brachytherapy and cryotherapy.
Azra Alizad, M.D., is interested in breast vibro-acoustography of human subjects. In one project, she is testing vibro-acoustography with an array probe on patients for detection of breast masses. In another, Dr. Alizad seeks to differentiate breast masses. Dr. Alizad is also involved in prostate imaging and more recently, in thyroid imaging using vibro-acoustography.
Gina Hesley, M.D., is evaluating a new technology for ultrasound-based strain imaging of the breast, to determine whether this qualitative elastography technique shows promise for differentiating benign from malignant solid breast masses. If successful, this technique might decrease the number of benign breast biopsies being performed. Co-investigators include Marilyn Morton, D.O. and Nicholas Hangiandreou, Ph.D.
