An objective comparison of cell-tracking algorithms

Abstract

We present a combined report on the results of three editions of the Cell Tracking Challenge, an ongoing initiative aimed at promoting the development and objective evaluation of cell tracking algorithms. With twenty-one participating algorithms and a data repository consisting of thirteen datasets of various microscopy modalities, the challenge displays today’s state of the art in the field. We analyze the results using performance measures for segmentation and tracking that rank all participating methods. We also analyze the performance of all algorithms in terms of biological measures and their practical usability. Even though some methods score high in all technical aspects, not a single one obtains fully correct solutions. We show that methods that either take prior information into account using learning strategies or analyze cells in a global spatio-temporal video context perform better than other methods under the segmentation and tracking scenarios included in the challenge.

Cell Segmentation for Image Cytometry: Advances, Insufficiencies, and Challenges

“…It has taken the community many decades and great endeavor to segment, identify, and analyze cells for explaining the cellular and molecular processes of health and disease. Studies have showed that cell segmentation can contribute to the better cell recognition in various fields of cell biology (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) and great advances have been achieved. However, it still has a long way to go before meeting the ultimate goal that the developed method can segment and identify different types of cells autonomously and the biologists can trust the segmentation and identification results blindly.…”

Analysis of the Interactions of Migrating Macrophages

“…Cell segmentation, the correct identification of a cell from other cells and background, and tracking, linking detected objects in one time frame to the objects in the following time frames, has been widely studied [6,7,8,9]. Cellular events, such as the interaction between cells, are also of importance.…”

“…Cellular events, such as the interaction between cells, are also of importance. In [6], several tracking algorithms were evaluated with a series migratory cells with very different conditions, not only in their ability to track detected objects, but also to identify cellular events, like mitosis. In [10], the analysis of the trajectories of crustaceans observed in microfluidic settings produced by the tracking algorithm was used to provide insights on an external state of the segmented objects, i.e.…”

Evaluation of Deep Learning Strategies for Nucleus Segmentation in Fluorescence Images

“…The most commonly-used metric for nucleus/cell segmentation evaluation is the Jaccard index (17)(18)(19), which measures pixel-wise overlap between ground truth and segmentations estimated by an algorithm. The most commonly-used metric for nucleus/cell segmentation evaluation is the Jaccard index (17)(18)(19), which measures pixel-wise overlap between ground truth and segmentations estimated by an algorithm.…”