Automatic tracking of protein vesicles
This work addresses a key need in cell biology for analyzing vesicle dynamics, but it is incremental as it builds on existing tracking methods with specific improvements for biological data.
The paper tackles the problem of automatically tracking protein vesicles in fluorescence imaging by formulating it as a video object tracking task, using dynamic programming for single objects and extending it with Kalman filters for multiple objects to handle track crossings, achieving high accuracy and robustness to noise in simulations.
With the advance of fluorescence imaging technologies, recently cell biologists are able to record the movement of protein vesicles within a living cell. Automatic tracking of the movements of these vesicles become key for qualitative analysis of dynamics of theses vesicles. In this thesis, we formulate such tracking problem as video object tracking problem, and design a dynamic programming method for tracking single object. Our experiments on simulation data show that the method can identify a track with high accuracy which is robust to the choose of tracking parameters and presence of high level noise. We then extend this method to the tracking multiple objects using the track elimination strategy. In multiple object tracking, the above approach often fails to correctly identify a track when two tracks cross. We solve this problem by incorporating the Kalman filter into the dynamic programming framework. Our experiments on simulated data show that the tracking accuracy is significantly improved.