The study of leukocyte rolling, arrest, and adhesion is a fundamental part of the investigation of inflammatory disease. Inflammatory disease is a major cause of morbidity (arthritis, multiple sclerosis) and mortality (sepsis). Several salient features of leukocyte rolling are relevant to inflammation including rolling leukocyte flux, rolling leukocyte flux fraction, leukocyte rolling velocity distribution, mean leukocyte rolling velocity, mean leukocyte rolling acceleration and rolling leukocyte volume fraction. In this project, the information technology needed for automated tracking and automated data collection within the study of rolling leukocytes in vivo is developed. Previously, limited success has been achieved with automated cell tracking for in vitro experiments. The state-of-the-art in tracking is extended here to track rolling leukocytes in vivo, which has not been possible with existing technology. Advanced information processing techniques are used to reliably track the leukocytes and to collect meaningful data in an automated manner. The tracking system includes tools for edge detection-based background registration, background removal, image restoration using nonlinear image models, video enhancement by morphological filters and diffusion techniques, adaptive cell templates that can accommodate cell deformation, cell position estimation and prediction using Kalman filter techniques, and track coasting for the conditions of occlusion and severe clutter. In the R21 phase, the ability to track rolling leukocytes in vivo and to automatically collect meaningful data will be demonstrated. Furthermore, the improvement in efficiency for in vivo experiments will be quantified. The milestones will validate the proof of concept using video recordings of rolling leukocytes observed by intravital microscopy in the mouse cremaster muscle, the rat mesentery, and the mouse carotid artery with and without fluorescent labeling. Video recordings from in vitro microscopy and from synthetic examples will be used in the validation process for comparative purposes. In the R33 phase, the cell tracking system is extended to fully digital, real-time implementation that allows for automated acquisition as well as tracking and data collection. A robust, user-friendly tracking system that is applicable to several cell analysis problems will be demonstrated and made available to the research community. For the specific study of rolling leukocytes in vivo, the automated tracking system will allow the computation of the velocity and acceleration of many leukocytes at many points in time and space, which is likely to result in new discoveries regarding the molecular mechanism underlying leukocyte recruitment. The automated tracking system will improve the database dramatically, enabling time averaging, spatial averaging, various derivations and studies of modulation of the activation process in rolling leukocytes. The new cell tracking system will not only accelerate and enhance analysis, but will also enable the generation of new knowledge not available through existing technology.
