PARAMETERS AFFECTING THE IMAGING AND DETECTION OF MICROBUBBLES IN BLOOD

Stabilized micron-sized bubbles, known as contrast agents, are often injected into the body to enhance ultrasound imaging of blood flow. The ability to detect such bubbles in blood depends on the relative magnitude of the acoustic power backscattered from the microbubbles (‘signal’) to the power backscattered from the red blood cells (‘noise’). Erythrocytes are acoustically small (Rayleigh regime), weak scatterers, and therefore the backscatter coefficient (BSC) of blood increases as the fourth power of frequency throughout the diagnostic frequency range. Microbubbles, on the other hand, are either resonant or super-resonant in the range 5-30 MHz. Above resonance, their total scattering cross-section remains constant with increasing frequency. In the present thesis, a theoretical model of the BSC of a suspension of red blood cells is presented and compared to the BSC of Optison® contrast agent microbubbles. It is predicted that, as the frequency increases, the BSC of red blood cell suspensions eventually exceeds the BSC of the strong scattering microbubbles, leading to a dramatic reduction in signal-to-noise ratio (SNR). This decrease in SNR with increasing frequency was also confirmed experimentally by use of an active cavitation detector for different concentrations of Optison® microbubbles in erythrocyte suspensions of different hematocrits. The magnitude of the observed decrease in SNR correlated well with theoretical predictions in most cases, except for very dense suspensions of red blood cells, where it is hypothesized that the close proximity of erythrocytes inhibits the acoustic response of the microbubbles.

Specialized Mappings Architecture with Applications to Vision-based Estimation of Articulated Body Pose

A fundamental task of vision systems is to infer the state of the world given some form of visual observations. From a computational perspective, this often involves facing an ill-posed problem; e.g., information is lost via projection of the 3D world into a 2D image. Solution of an ill-posed problem requires additional information, usually provided as a model of the underlying process. It is important that the model be both computationally feasible as well as theoretically well-founded. In this thesis, a probabilistic, nonlinear supervised computational learning model is proposed: the Specialized Mappings Architecture (SMA). The SMA framework is demonstrated in a computer vision system that can estimate the articulated pose parameters of a human body or human hands, given images obtained via one or more uncalibrated cameras. The SMA consists of several specialized forward mapping functions that are estimated automatically from training data, and a possibly known feedback function. Each specialized function maps certain domains of the input space (e.g., image features) onto the output space (e.g., articulated body parameters). A probabilistic model for the architecture is first formalized. Solutions to key algorithmic problems are then derived: simultaneous learning of the specialized domains along with the mapping functions, as well as performing inference given inputs and a feedback function. The SMA employs a variant of the Expectation-Maximization algorithm and approximate inference. The approach allows the use of alternative conditional independence assumptions for learning and inference, which are derived from a forward model and a feedback model. Experimental validation of the proposed approach is conducted in the task of estimating articulated body pose from image silhouettes. Accuracy and stability of the SMA framework is tested using artificial data sets, as well as synthetic and real video sequences of human bodies and hands.

Estimating Human Body Pose from a Single Image via the Specialized Mappings Architecture

A non-linear supervised learning architecture, the Specialized Mapping Architecture (SMA) and its application to articulated body pose reconstruction from single monocular images is described. The architecture is formed by a number of specialized mapping functions, each of them with the purpose of mapping certain portions (connected or not) of the input space, and a feedback matching process. A probabilistic model for the architecture is described along with a mechanism for learning its parameters. The learning problem is approached using a maximum likelihood estimation framework; we present Expectation Maximization (EM) algorithms for two different instances of the likelihood probability. Performance is characterized by estimating human body postures from low level visual features, showing promising results.