Time-series analysis of cellular shapes using transported velocity fields
Abstract
This paper presents a generative statistical model for analyzing time series of planar shapes. Using elastic shape analysis, we separate object kinematics (rigid motions and speed variability) from morphological evolution, representing the latter through transported velocity fields (TVFs). A principal component analysis (PCA) based dimensionality reduction of the TVF representation provides a finite-dimensional Euclidean framework, enabling traditional time-series analysis. We then fit a vector auto-regressive (VAR) model to the TVF-PCA time series, capturing the statistical dynamics of shape evolution. To characterize morphological changes, we use VAR model parameters for model comparison, synthesis, and sequence classification. Leveraging these parameters, along with machine learning classifiers, we achieve high classification accuracy. Extensive experiments on cell motility data validate our approach, demonstrating its effectiveness in modeling and classifying migrating cells based on morphological evolution—marking a novel contribution to the field.
Related
- Variation in enhancer activity across brain regions defines neurological disease risk and shapes cellular pathology
- Variabilities in the shapes of the receptive fields of simple cells in mice
- FlowMap: Geometry Dynamics Consistent Embedding of RNA Velocity for Interpretable Cellular Trajectories
- STAVelo: deciphering spatiotemporal cellular dynamics via spatial RNA velocity inference from spatial transcriptomics data
- Cellular context shapes the transcriptomic impact of somatic variants
BibTeX
@article{deng2026timeseries,
title = {{Time-series analysis of cellular shapes using transported velocity fields}},
author = {Ximu Deng and Rituparna Sarkar and Elisabeth Labruyere and Jean-Christophe Olivo-Marin and Anuj Srivastava},
journal = {Pattern Recognition},
year = {2026},
month = nov,
doi = {10.1016/j.patcog.2025.112056},
url = {https://hal.science/hal-05852521v1},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Time-series analysis of cellular shapes using transported velocity fields
AU - Ximu Deng
AU - Rituparna Sarkar
AU - Elisabeth Labruyere
AU - Jean-Christophe Olivo-Marin
AU - Anuj Srivastava
PY - 2026
DA - 2026/11/04
JO - Pattern Recognition
DO - 10.1016/j.patcog.2025.112056
UR - https://hal.science/hal-05852521v1
AB - This paper presents a generative statistical model for analyzing time series of planar shapes. Using elastic shape analysis, we separate object kinematics (rigid motions and speed variability) from morphological evolution, representing the latter through transported velocity fields (TVFs). A principal component analysis (PCA) based dimensionality reduction of the TVF representation provides a finite-dimensional Euclidean framework, enabling traditional time-series analysis. We then fit a vector auto-regressive (VAR) model to the TVF-PCA time series, capturing the statistical dynamics of shape evolution. To characterize morphological changes, we use VAR model parameters for model comparison, synthesis, and sequence classification. Leveraging these parameters, along with machine learning classifiers, we achieve high classification accuracy. Extensive experiments on cell motility data validate our approach, demonstrating its effectiveness in modeling and classifying migrating cells based on morphological evolution—marking a novel contribution to the field.
SN - 0031-3203
SN - 1873-5142
ER -