Magnetic resonance images of the brain of an elderly subject who has vascular lesions in the white matter.  The images show(upper left across to lower left) proton density, T2, T1 and FLAIR weighting. The final image shows a segmented image using an automated algorithm

Automated Image Analysis Example

James R. MacFall, Ph.D.

Professor of Radiology and Biomedical Engineering

Research Interest:

Magnetic Resonance Imaging Characterization of Tissue Structure and Function


Research Summary:

Magnetic Resonance (MR) has a wide variety of sensitivities to the structure and function of tissue. The most common form of MR imaging is sensitive to the water proton density in the tissue and the MR relaxation times: T1 and T2. In addition, MR is sensitive to the bulk motion of blood in larger vessels (flow) and to its slow percolation in capillaries (perfusion/microcirculation). The slowest motion of water, that of its convection and self-diffusion, can also be observed with MR imaging by the use of large magnetic field gradients. MR imaging can also be performed with other biologically important nuclei such as phosphorus and carbon or even for nuclei such as deuterium, fluorine, helium and xenon. My research interests are centered around the exploitation of this extraordinary richness of sensitivities to detect or understand disease in human subjects and in animal models of disease.


Current Projects

1.Dynamic Contrast Enhanced Magnetic Resonance Imaging.  In collaboration with colleagues in Radiology (Dr. Barboriak's Lab) and Radiation Oncology, we are developing MR data acquisition (pulse sequences), analytic tools and contrast agents to use for dynamic contrast imaging. Such methods will be valuable for detecting cancer and for monitoring therapy.

2.Diffusion Correlates of Neuronal Activity.  When neurons are activated by some stimulus there are alterations in the blood flow and the metabolic activities of the nearby supporting tissue.  We are investigating in humans and animals whether there are also water diffusion changes in the tissue which could potentially serve as an additional way to localize brain activity.  

3.Brain Structural Changes in Psychiatric Disorders.  In collaboration with the Department of Psychiatry at Duke we are studying whether brain regions show changes in such disorders as Late Life Depression and Bipolar Disorder. An example of the automated image segmentation we are doing is shown above. Of particular interest is whether the vascular lesions that are often seen in elderly subjects may be contributing to an imbalance in mood regulation by damaging the white matter connections between brain regions.

4.Potential Link Between Childhood Febrile Siezures and Later Development of Temporal Lobe Epilepsy: The FEBSTAT study. Collaborating with investigators in this multi-site study of this possible link between febrile siezures and epilepsy, developing the neuroimaging protocols and participating in the analysis.


Cooperating Centers

1. The Duke Department of Radiology's Center for Advanced Magnetic Resonance Development (CAMRD).  Dr. MacFall and Dr. H. Cecil Charles are the technical directors of this center.. It contains a 3 Tesla whole body commercial MR imaging and spectroscopy system (Siemens Trio) as well as a variety of computer systems for image processing

2. The Center for In Vivo Microscopy directed by Dr. G. Alan Johnson, in the Duke Department of Radiology which has 2 Tesla, 7 Tesla and 9.4 Tesla small animal MR imaging instruments.

3. The Duke Brain Imaging and Analysis Center , directed by Dr. G. McCarthy. This laboratory has resources to develop stimuli for fMRI and computers and programs for image analysis. Planned enhancements for this Center are a 1.5T MRI system and a 4.0T MRI system.


Department of Radiology
Department of Biomedical Engineering
Duke University Medical Center, Box 3808
1812C Duke Hospital
Phone (919) 684-7808 (office)
Fax: (919) 684-7126

© Center for Advanced Magnetic Resonance Development, Department of Radiology, Duke University Medical Center, 2005