Search for temporal cell segmentation robustness in phase-contrast microscopy videos

TL;DR

This paper introduces a deep learning workflow utilizing transfer learning and recurrent units for robust, temporally consistent segmentation of cancer cells in phase-contrast microscopy videos, aiding in cell morphology studies.

Contribution

It presents a novel deep learning approach with temporal modeling for cell segmentation, along with a new annotated dataset and open-source tools.

Findings

Stable segmentation results over time

Robust to different initializations and training data sampling

Effective in analyzing cell morphology changes

Abstract

Studying cell morphology changes in time is critical to understanding cell migration mechanisms. In this work, we present a deep learning-based workflow to segment cancer cells embedded in 3D collagen matrices and imaged with phase-contrast microscopy. Our approach uses transfer learning and recurrent convolutional long-short term memory units to exploit the temporal information from the past and provide a consistent segmentation result. Lastly, we propose a geometrical-characterization approach to studying cancer cell morphology. Our approach provides stable results in time, and it is robust to the different weight initialization or training data sampling. We introduce a new annotated dataset for 2D cell segmentation and tracking, and an open-source implementation to replicate the experiments or adapt them to new image processing problems.