Time Series Analysis: Forecasting and Control
Time Series Analysis: Forecasting and Control
Object segmentation by long term analysis of point trajectories
ECCV'10 Proceedings of the 11th European conference on Computer vision: Part V
On analyzing video with very small motions
CVPR '11 Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition
Hi-index | 0.00 |
Stem-cell derived cardiomyocytes are increasingly being studied for pre-clinical drug testing for cardiotoxity. Traditional analysis using plates that measure motion patterns through physical contact are not suitable for high-throughput analysis, and may cause undesirable tissue response. We recently developed a method for automated cardiomyocytes monitoring by analyzing the apparent motion captured by video microscopy. However, this method is limited to producing overall signals on the whole image, which makes it unsuitable for images containing multiple motion patterns in different regions. Here we introduce a motion-based segmentation method that can robustly segment regions with different beating rhythms without knowing beforehand the number of regions. The regions can then be characterized separately for more robust cardiomyocytes analysis. We demonstrate the accuracy and effectiveness of our approach on a number of synthetic and real datasets.