Automated detection and tracking of many cells by using 4D live-cell imaging data

Abstract

Motivation: Automated fluorescence microscopes produce massive amounts of images observing cells, often in four dimensions of space and time. This study addresses two tasks of time-lapse imaging analyses; detection and tracking of the many imaged cells, and it is especially intended for 4D live-cell imaging of neuronal nuclei of Caenorhabditis elegans. The cells of interest appear as slightly deformed ellipsoidal forms. They are densely distributed, and move rapidly in a series of 3D images. Thus, existing tracking methods often fail because more than one tracker will follow the same target or a tracker transits from one to other of different targets during rapid moves.Results: The present method begins by performing the kernel density estimation in order to convert each 3D image into a smooth, continuous function. The cell bodies in the image are assumed to lie in the regions near the multiple local maxima of the density function. The tasks of detecting and tracking the cells are then addressed with two hill-climbing algorithms. The positions of the trackers are initialized by applying the cell-detection method to an image in the first frame. The tracking method keeps attacking them to near the local maxima in each subsequent image. To prevent the tracker from following multiple cells, we use a Markov random field (MRF) to model the spatial and temporal covariation of the cells and to maximize the image forces and the MRF-induced constraint on the trackers. The tracking procedure is demonstrated with dynamic 3D images that each contain >100 neurons of C.elegans.Availability: http://daweb.ism.ac.jp/yoshidalab/crest/ismb2014Supplementary information: Supplementary data are available at http://daweb.ism.ac.jp/yoshidalab/crest/ismb2014Contact: yoshidar@ism.ac.jp

SPF-CellTracker: Tracking Multiple Cells with Strongly-Correlated Moves Using a Spatial Particle Filter

“…We evaluate tracking performance of SPF and the state-of-the-art method reported by Tokunaga et al [23].…”

“…The top panel of the figure shows the starting positions of trackers. Five trackers that do not track Figure 3 show positions of trackers in the final frame for [23] and SPF, respectively. The figure suggests that at least large inconsistencies between cell centroids and positions of trackers for the both methods do not exist.…”

Neuron ID dataset facilitates neuronal annotation for whole-brain activity imaging of C. elegans

“…Such detailed knowledge opens up unique opportunities in neuroscience at both single-cell and network levels. Recent advances in microscopy techniques also enable whole-brain activity imaging of the worm [4][5][6][7][8][9][10][11], even for the free-moving worms [12][13][14]. The neural activities were obtained at single-cell resolution, and the identities of limited numbers of neurons were annotated manually in some of the studies [4-8, 10, 14].…”

An ensemble learning approach to auto-annotation for whole-brain C. elegans imaging

“…To achieve the ultimate goal of fully automated data processing pipeline for whole-brain images of C. elegans, it is important to have reliable algorithms for detecting neurons from a series of 3D images, segmenting the voxels for neuron size estimation, tracking the identified neurons along time and annotating neurons for functional analyses. Researchers have reached certain success for detection, segmentation and tracking [25,26]. However, for annotation, existing literatures mainly focused on the body cells [12,20,1,6].…”