The method of microtubule tracking and dynamics analysis, presented here, improves upon the current means of manual and automated quantification of microtubule behavior. Key contributions are increasing accuracy and data volume, eliminating user bias and providing advanced analysis tools for the discovery of temporal patterns in cellular processes. By tracking the entire length of each resolvable microtubule, as opposed to only the tip, it is possible to boost dynamics studies with positional information that is virtually impossible to collect manually. We demonstrate the method on the analysis of a microtubule dataset, which was manually tracked and analyzed in the study of βIII-tubulin isoform. Our results show that automated recognition of temporal patterns in cellular processes offers a highly promising potential.

It has been recently shown that local feature approaches to face verification are considerably more robust than holistic approaches, in terms of translations (caused by automatic face localization) and pose variations. In this paper we first investigate whether features based on local Principal Component Analysis (LPCA) are more discriminative than features based on the 2D Discrete Cosine Transform (2D DCT). We also investigate several methods for modifying the two feature extraction techniques in order to counteract the effects of linear and non-linear illumination changes, without losing discriminative information. Results on the XM2VTS database show that when using a Bayesian classifier based on Gaussian Mixture Models (GMMs), the performances of 2D DCT and LPCA techniques are quite similar, suggesting that the 2D DCT technique is preferable due to its lower computational complexity. When using 8×8 blocks, modifying the 2D DCT and LPCA techniques by removing the first coefficient, which is the most affected by illumination changes, enhances robustness with little change in discrimination ability; removing further coefficients causes a noticeable reduction in performance on clean images and provides little gain in robustness. When using the 2D DCT with 16×16 blocks, the first three coefficients need to be removed in order to achieve good robustness. It is further shown that contrary to previously published results, the use of deltas of low-order coefficients (to alleviate performance losses caused by removing coefficients) can adversely affect robustness.

No abstract

We propose in this paper an interactive segmentation algorithm based on curve evolution techniques. The task of automated segmentation has proven to be highly complex and application dependent. User's knowledge can be used to alleviate the problem. In this paper, we propose the use of a recently developed curve evolution technique [1], augmented with a relevance feedback phase through user interaction. After the initial automatic segmentation is computed, the user presents his positive/negative feedback via a simple user interface. Segmentation parameters are then adapted locally to reflect user's requirements. Experimental results show the usefulness of the proposed approach in interactive segmentation tasks.