Multi-Modal Volume Rendering and Segmentation

Dr. Ron Kikinis/SPL

This project is a collaboration between Center for Integrative Biomedical Computing and Dr. Ron Kikinis, Director of the Surgical Planning Lab (SPL) at the Brigham and Women's Hospital, Harvard Medical School. Dr. Kikins and members of his lab actively develop, evalute, and refine segmentation and visualization algorithms to increase their practical utility for medical applications. His research interest in this technology collaboration is to incorporate advanced, interactive three-dimensional direct volume rendering techniques into the 3D Slicer (or simply 'Slicer') program. The addition of direct volume rendering into Slicer will complement the current display of discrete polygonal structures, by adding support for the simultaneous visualization of diffuse quantities, such as perfusion and PET data.

Dr. Ron Kikinis has done pioneering research in the integration of image processing and visualization into medical practice, including surgical planning, neurosurgury, endovascular intervention, as well as the study of psychiatric and neurologic diseases. This has been accomplished by bridging algorithms and technology from computer science with the needs of working medical systems. Areas of research have included segmentation, registration, and visualization.

Dr. Kikins has served as a member of several editorial boards, including Medical Image Analysis, Neuroimage, and Computer Aided Surgery, and has received numerous awards for his contributions in computational medicine.

Dr. Kikins founded the SPL in 1990, and has continuously fostered its growth into a premier, world-class institution for advanced computational medicine.

Diffusion Tensor Analysis, Tracking, and Visualization

Dr. Carl-Fredrik Westin

This project is a collaboration between Center for Integrative Biomedical Computing and Dr. Carl-Fredrik Westin, Director of the Laboratory of Mathematics in Imaging (LMI) at the Brigham and Women's Hospital. Dr. Westin's research interests are focused on medical applications of image analysis. He is currently working on analysis of Diffusion Tensor MRI data, and automated segmentation and registration of data from MRI, CT, and Ultrasound, using multidimensional signal processing techniques.

The purpose of this collaboration is to advance a standard method of diffusion tensor image analysis, fiber tracking, by performing clustering of fiber tracts and computing statistics on fibers within and between scans. The biomedical focus of this project will be to better understand and quantify the impact of schizophrenia on white matter architecture.

Dr. Carl-Fredrik Westin received the MSc degree in Applied Physics and Electrical Engineering in 1988 from Linkoping University. He joined the Computer Vision Laboratory the same year where he did research on color, information representation, image flow, frequency estimation, filtering of uncertain and irregularly sampled data and tensor operators in image analysis. In 1991, Dr. Westin was awarded the SAAB-SCANIA prize for his work in field of Computer Vision. He received the Lic.Techn. degree on the topic feature extraction from a tensor image descriptions, in 1991. In 1994 Westin, graduated as a PhD in computer vision, also from Linkoping University. His thesis "A tensor framework for multidimensional signal processing" presents a novel method for filtering uncertain and irregularly sampled data termed normalized convolution. He joined Brigham and Women's Hospital and Harvard Medical School in 1996. He is presently an Assistant Professor of Radiology, in the Radiology Department at Harvard Medical School, and Director of the Laboratory of Mathematics in Imaging (LMI) at the Brigham and Women's Hospital. Additionally, he has a joint appointment with the MIT Artificial Intelligence Laboratory, Cambridge, MA. Dr. Westin is the Core PI responsible for the DT-MRI core of the Neuroimaging Analysis Center (NAC).

Dr. Gordon Kindlmann is a post-doctoral Research Fellow in the LMI, having received his PhD from the University of Utah, under Dr. Chris Johnson of the Scientific Computing and Imaging Institute. His methods for tensor visualization have been adopted in the DT-MRI community, and include deflection-based tractography, barycentric description of tensor shape, and superquadric tensor glyphs. He is currently working under Dr. Carl-Fredrik Westin on developing novel methods for quantifying and visualization clinically relevant parameters of diffusion-weighted images.

Diffusion Tensor Image Analysis in the Study of Age-Related Neurodegeneration

Professor Stephen Wong

This collaboration addresses the problem of visualizing and analyzing large numbers of diffusion weighted and diffusion tensor (DTI) images in order to study the effects of aging on neurodegeneration. Aging and aging related neurodegeneration conditions affect tens of millions of Americans. Typical conditions affecting central nervous system include Alzheimer's disease, dementia, stroke, and Parkinson's disease. By one account, more than four million people in the United States suffer from Alzheimer's disease. In recent years, there has been increasing interest in research on aging. The study of aging has been underpinned by rapid technology development in medical imaging, such as MRI, CT, and PET. It is well accepted that during aging process the adult nervous system changes with deteriorating cognitive abilities, worsening reflexes, and much higher incidence of neurodegenerative disease. However, despite our current understanding of the etiology of neurodegeneration, the underpinnings of normal aging remain a complete mystery. Therefore, the age-related mechanisms of increased susceptibility to pathological alterations are also enigmatic. In order to address the scientific questions surrounding these issues, researchers in Professor Wong's group at Brigham and Women's Hospital are acquiring databases of DTI images, in order to quantify the affects of aging. The tools they need to do this will entail user interaction, and the corresponding visualization of DTI volumes.

The Harvard Center for Neurodegeneration and Repair (HCNR) is a Harvard-wide initiative of Harvard Medical School (HMS) and eight affiliated teaching hospitals, including Brigham and Women's Hospital and Massachusetts General Hospital. The HCNR is pursuing an interdisciplinary approach in neuroscience research. The HCNR Center for Bioinformatics, in particular, had the charter to investigate advanced magnetic resonance imaging and bioinformatics techniques to evaluate and compare normal brains and aging brains. They expect to derive longitudinal association and correlation among time-lapse, in-vivo, high-resolution MRI. They use diffusion tensor imaging (DTI) to examine age-related changes in genu, splenium, and centrum semiovale white matter diffusivity and calculate the relationship between diffusivity (trace) and fractional anisotropy (FA) across and within individual subjects. Researchers associated with this work perform DTI scans on both human and animal subjects (e.g., rat and mice of aging models). They are also studying the role of macrostructural and microstructural partial volume effects on the DTI metrics. They hypothesize that ellipsoid (diffusion tensor) shape analysis will provide insights into the relationship between morphometric erosion in the splenium and centrum and ages of subjects.

Collaborative Project on Visualization of Brain Variability

Dr. Paul Thompson

Dr. Thompson is an Associate Professor of Neurology at the UCLA School of Medicine. His research focuses on the neuroscience, mathematics, software engineering and clinical aspects of neuroimaging and brain mapping. He is a member of the UCLA Brain Research Institute, and is Principal Investigator on the Image Analysis Core of an NCRR-funded National Resource at UCLA.

He has organized national workshops and courses on brain imaging in dementia (e.g., at the Human Brain Mapping meeting in New York, 2003), and has advised the National Academy of Sciences, Institute of Medicine, and IPSEN Foundation (France) on implications of brain research for understanding dementia and childhood brain development. Working closely with Dr. Arthur Toga at the UCLA Laboratory of Neuro Imaging for the last 10 years, Dr. Thompson's 300+ collaborative publications (including 100 peer-reviewed journal papers) describe novel mathematical and computational strategies for mapping brain structure and function in health and disease. These tools are being applied to map disease progression in Alzheimer's disease and those at risk. His recent algorithms for brain image analysis are currently being tested and used in over 20 national and international collaborations, by research teams in both pharmaceutical companies and other universities.