Alignment by Maximization of Mutual Information

A new information-theoretic approach is presented for finding the pose of an object in an image. The technique does not require information about the surface properties of the object, besides its shape, and is robust with respect to variations of illumination. In our derivation, few assumptions are made about the nature of the imaging process. As a result the algorithms are quite general and can foreseeably be used in a wide variety of imaging situations. Experiments are presented that demonstrate the approach registering magnetic resonance (MR) images with computed tomography (CT) images, aligning a complex 3D object model to real scenes including clutter and occlusion, tracking a human head in a video sequence and aligning a view-based 2D object model to real images. The method is based on a formulation of the mutual information between the model and the image called EMMA. As applied here the technique is intensity-based, rather than feature-based. It works well in domains where edge or gradient-magnitude based methods have difficulty, yet it is more robust than traditional correlation. Additionally, it has an efficient implementation that is based on stochastic approximation. Finally, we will describe a number of additional real-world applications that can be solved efficiently and reliably using EMMA. EMMA can be used in machine learning to find maximally informative projections of high-dimensional data. EMMA can also be used to detect and correct corruption in magnetic resonance images (MRI).

Segmentation of Brain Tissue from Magnetic Resonance Images

Segmentation of medical imagery is a challenging problem due to the complexity of the images, as well as to the absence of models of the anatomy that fully capture the possible deformations in each structure. Brain tissue is a particularly complex structure, and its segmentation is an important step for studies in temporal change detection of morphology, as well as for 3D visualization in surgical planning. In this paper, we present a method for segmentation of brain tissue from magnetic resonance images that is a combination of three existing techniques from the Computer Vision literature: EM segmentation, binary morphology, and active contour models. Each of these techniques has been customized for the problem of brain tissue segmentation in a way that the resultant method is more robust than its components. Finally, we present the results of a parallel implementation of this method on IBM's supercomputer Power Visualization System for a database of 20 brain scans each with 256x256x124 voxels and validate those against segmentations generated by neuroanatomy experts.

Segmentation and Alignment of Speech and Sketching in a Design Environment

Sketches are commonly used in the early stages of design. Our previous system allows users to sketch mechanical systems that the computer interprets. However, some parts of the mechanical system might be too hard or too complicated to express in the sketch. Adding speech recognition to create a multimodal system would move us toward our goal of creating a more natural user interface. This thesis examines the relationship between the verbal and sketch input, particularly how to segment and align the two inputs. Toward this end, subjects were recorded while they sketched and talked. These recordings were transcribed, and a set of rules to perform segmentation and alignment was created. These rules represent the knowledge that the computer needs to perform segmentation and alignment. The rules successfully interpreted the 24 data sets that they were given.

Surface Functionalization of Monodisperse Magnetic Nanoparticles

We present a systematic methodology to functionalize magnetic nanoparticles through surface-initiated atom-transfer radical polymerization (ATRP). The magnetite nanoparticles are prepared according to the method proposed by Sun et al. (2004), which leads to a monodisperse population of ~ 6 nm particles stabilized by oleic acid. The functionalization of the nanoparticles has been performed by transforming particles into macro-initiators for the ATRP, and to achieve this two different routes have been explored. The first one is the ligand-exchange method, which consists of replacing some oleic acid molecules adsorbed on the particle surface with molecules that act as an initiator for ATRP. The second method consists in using the addition reaction of bromine to the oleic acid double bond, which turns the oleic acid itself into an initiator for the ATRP. We have then grown polymer brushes of a variety of acrylic polymers on the particles, including polyisopropylacrylamide and polyacrylic acid. The nanoparticles so functionalized are water soluble and show responsive behavior: either temperature responsive behavior when polyisopropylacrylamide is grown from the surface or PH responsive in the case of polyacrylic acid. This methodology has potential applications in the control of clustering of magnetic nanoparticles.

Synthesis and Characterization of Magnetic Nanoparticles with High Magnetization and Good Oxidation Resistibility

Magnetic nanoparticles attract increasing attention because of their current and potential biomedical applications, such as, magnetically targeted and controlled drug delivery, magnetic hyperthermia and magnetic extraction. Increased magnetization can lead to improved performance in targeting and retention in drug delivery and a higher efficiency in biomaterials extraction. We reported an approach to synthesize iron contained magnetic nanoparticles with high magnetization and good oxidation resistibility by pyrolysis of iron pentacarbonyl (Fe(CO)[subscript 5]) in methane (CH[subscript 4]). Using the high reactivity of Fe nanoparticles, decomposition of CH[subscript 4] on the Fe nanoparticles leads to the formation of nanocrystalline iron carbides at a temperature below 260°C. Structural investigation indicated that the as-synthesized nanoparticles contained crystalline bcc Fe, iron carbides and spinel iron oxide. The Mössbauer and DSC results testified that the as-synthesized nanoparticle contained three crystalline iron carbide phases, which converted to Fe[subscript 3]C after a heat treatment. Surface analysis suggested that the as-synthesized and subsequently heated iron-iron carbide particles were coated by iron oxide, which originated from oxidization of surface Fe atoms. The heat-treated nanoparticles exhibited a magnetization of 160 emu/g, which is two times of that of currently used spinel iron oxide nanoparticles. After heating in an acidic solution with a pH value of 5 at 60°C for 20 h, the nanoparticles retained 90 percentage of the magnetization.