A probabilistic approach to joint cell tracking and segmentation in high-throughput microscopy videos

• Published in: Medical image analysis Vol. 47; pp. 140 - 152
• Main Authors: Arbelle, Assaf, Reyes, Jose, Chen, Jia-Yun, Lahav, Galit, Riklin Raviv, Tammy
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01-07-2018; Elsevier BV
• Subjects: Algorithms; Bayes Theorem; Bayesian analysis; Cell; Cell lineage; Cell size; Cell Tracking - methods; Cells; Computer applications; Data processing; Dynamic models; Fast marching; Humans; Image processing; Image segmentation; Joint; Mathematical models; Microscopy; Microscopy, Video; Multiple object; Probability; Segmentation; Software; Statistical inference; Time series; Tracking; Multiple object; Tracking; Segmentation; Microscopy; Joint; Cell; Fast marching
• ISSN: 1361-8415; 1361-8423
• DOI: 10.1016/j.media.2018.04.006

•We propose an unsupervised, automatic tracking and segmentation framework for high-throughput microscopy image sequences.•Cell segmentation and tracking are tied together via Bayesian inference of dynamic models.•The Kalman inference problem is exploited to estimate the time-wise cell shape uncertainty in addition to cell trajectory. The inferred cell properties are integrated with the observed image, using a fast marching algorithm, to obtain the image likelihood for cell segmentation and association.•We present highly accurate results, surpassing the state of the art, for a variety of microscopy data sets with high dynamics, including long sequences (hundreds of frames).
[Display omitted]
We present a novel computational framework for the analysis of high-throughput microscopy videos of living cells. The proposed framework is generally useful and can be applied to different datasets acquired in a variety of laboratory settings. This is accomplished by tying together two fundamental aspects of cell lineage construction, namely cell segmentation and tracking, via a Bayesian inference of dynamic models. In contrast to most existing approaches, which aim to be general, no assumption of cell shape is made. Spatial, temporal, and cross-sectional variation of the analysed data are accommodated by two key contributions. First, time series analysis is exploited to estimate the temporal cell shape uncertainty in addition to cell trajectory. Second, a fast marching (FM) algorithm is used to integrate the inferred cell properties with the observed image measurements in order to obtain image likelihood for cell segmentation, and association. The proposed approach has been tested on eight different time-lapse microscopy data sets, some of which are high-throughput, demonstrating promising results for the detection, segmentation and association of planar cells. Our results surpass the state of the art for the Fluo-C2DL-MSC data set of the Cell Tracking Challenge (Maška et al., 2014).