Format

Send to

Choose Destination
Med Image Anal. 2018 Aug;48:147-161. doi: 10.1016/j.media.2018.05.009. Epub 2018 Jun 5.

Tracing cell lineages in videos of lens-free microscopy.

Author information

1
Department of Informatics, Technical University of Munich, Germany; Institute for Advanced Study, Technical University of Munich, Germany. Electronic address: markus.rempfler@tum.de.
2
Faculty of Physics, Ludwig-Maximilians University of Munich, Germany.
3
Department of Informatics, Technical University of Munich, Germany.
4
Max Planck Institute for Informatics, Saarbrücken, Germany.
5
Department of Informatics, Technical University of Munich, Germany; Institute for Advanced Study, Technical University of Munich, Germany.

Abstract

In vitro experiments with cultured cells are essential for studying their growth and migration pattern and thus, for gaining a better understanding of cancer progression and its treatment. Recent progress in lens-free microscopy (LFM) has rendered it an inexpensive tool for label-free, continuous live cell imaging, yet there is only little work on analysing such time-lapse image sequences. We propose (1) a cell detector for LFM images based on fully convolutional networks and residual learning, and (2) a probabilistic model based on moral lineage tracing that explicitly handles multiple detections and temporal successor hypotheses by clustering and tracking simultaneously. (3) We benchmark our method in terms of detection and tracking scores on a dataset of three annotated sequences of several hours of LFM, where we demonstrate our method to produce high quality lineages. (4) We evaluate its performance on a somewhat more challenging problem: estimating cell lineages from the LFM sequence as would be possible from a corresponding fluorescence microscopy sequence. We present experiments on 16 LFM sequences for which we acquired fluorescence microscopy in parallel and generated annotations from them. Finally, (5) we showcase our methods effectiveness for quantifying cell dynamics in an experiment with skin cancer cells.

KEYWORDS:

Cell detection; Cell lineage tracing; Fully convolutional neural networks; Lens-free microscopy

PMID:
29933115
DOI:
10.1016/j.media.2018.05.009
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center