Faculty Bibliography

Validation and method of comparison for segmentation of magnetic resonance images (MRI) presenting pathology is a challenging task due to the lack of reliable ground truth. We propose a new method for generating synthetic multi-modal 3D brain MRI with tumor and edema, along with the ground truth. Tumor mass effect is modeled using a biomechanical model, while tumor and edema infiltration is modeled as a reaction-diffusion process that is guided by a modified diffusion tensor MRI. We propose the use of warping and geodesic interpolation on the diffusion tensors to simulate the displacement and the destruction of the white matter fibers. We also model the process where the contrast agent tends to accumulate in cortical csf regions and active tumor regions to obtain contrast enhanced T1w MR image that appear realistic. The result is simulated multi-modal MRI with ground truth available as sets of probability maps. The system will be able to generate large sets of simulation images with tumors of varying size, shape and location, and will additionally generate infiltrated and deformed healthy tissue probabilities.

Active contour segmentation and its robust implementation using level sets have been studied thoroughly in the medical image analysis literature. Despite the availability of these powerful methods, clinical research still largely relies on manual slice-by-slice outlining for anatomical structure segmentation. To bridge the gap between methodological advances and clinical routine, we developed ITK-SNAP: an open source application intended to make level set segmentation easily accessible to a wide range of users with various levels of mathematical expertise. We briefly describe this new tool and report the results of a validation study in which ITK-SNAP was compared to manual segmentation of the caudate in the context of an ongoing child neuroimaging autism study

Despite multiple advances in medical imaging, noninvasive monitoring of therapeutic efficacy for malignant gliomas remains problematic. An underutilized observation is that malignancy induces characteristic abnormalities of vessel shape. These characteristic shape abnormalities affect both capillaries and much larger vessels in the tumor vicinity, involve larger vessels prior to sprout formation, and are generally not present in hypervascular benign tumors. Vessel shape abnormalities associated with malignancy thus may appear independently of increase in vessel density. We hypothesize that an automated, computerized analysis of vessel shape as defined from high-resolution MRA can provide valuable information about tumor activity during the treatment of malignant gliomas. This report describes vessel shape properties in 10 malignant gliomas prior to treatment, in 2 patients in remission during treatment, and in 2 patients with recurrent disease. One subject was scanned multiple times. The method involves an automated, statistical analysis of vessel shape within a region of interest for each tumor, normalized by the values obtained from the vessels within the same region of interest of 34 healthy subjects. Results indicate that untreated tumors display statistically significant vessel tortuosity abnormalities. These abnormalities involve vessels not only within the tumor margins as defined from MR but also vessels in the surrounding tissue. The abnormalities resolve during effective treatment and recur with tumor recurrence. We conclude that vessel shape analysis could provide an important means of assessing tumor activity.

While it has been hypothesized that brain development is abnormal in schizophrenia and other neurodevelopmental disorders, there have been few attempts to study very early brain development in children. Twenty unsedated healthy newborns underwent 3 Tesla magnetic resonance imaging (MRI), including diffusion tensor imaging (DTI). The left ventricle was significantly larger than the right; females had significantly larger ventricles than males. Fractional anisotropy (FA) increased significantly with gestational age in the genu and splenium of the corpus callosum. It is feasible to study brain development in unsedated newborns using 3 T MRI.

This paper describes a framework for automatic brain tumor segmentation from MR images. The detection of edema is done simultaneously with tumor segmentation, as the knowledge of the extent of edema is important for diagnosis, planning, and treatment. Whereas many other tumor segmentation methods rely on the intensity enhancement produced by the gadolinium contrast agent in the T1-weighted image, the method proposed here does not require contrast enhanced image channels. The only required input for the segmentation procedure is the T2 MR image channel, but it can make use of any additional non-enhanced image channels for improved tissue segmentation. The segmentation framework is composed of three stages. First, we detect abnormal regions using a registered brain atlas as a model for healthy brains. We then make use of the robust estimates of the location and dispersion of the normal brain tissue intensity clusters to determine the intensity properties of the different tissue types. In the second stage, we determine from the T2 image intensities whether edema appears together with tumor in the abnormal regions. Finally, we apply geometric and spatial constraints to the detected tumor and edema regions. The segmentation procedure has been applied to three real datasets, representing different tumor shapes, locations, sizes, image intensities, and enhancement.

Statistical shape analysis has become of increasing interest to the neuroimaging community due to its potential to precisely locate morphological changes and thus potentially discriminate between healthy and pathological structures. This paper describes a combined boundary and medial shape analysis based on two different shape descriptions applied to a study of the hippocampus shape abnormalities in schizophrenia. The first shape description is the sampled boundary implied by the spherical harmonic SPHARM description. The second one is the medial shape description called M-rep. Both descriptions are sampled descriptions with inherent point correspondence. Their shape analysis is based on computing differences from an average template structure analyzed using standard group mean difference tests. The results of the global and local shape analysis in the presented hippocampus study exhibit the same patterns for the boundary and the medial analysis. The results strongly suggest that the normalized hippocampal shape of the schizophrenic group is different from the control group, most significantly as a deformation difference in the tail region.