Tracking leukocytes in intravital time lapse images using 3D cell association learning network

作者：

Highlights：

• We proposed a novel deep-learning-based object linkage method to tackle the common challenges involved in tracking highly amorphous and migrating leukocytes in large-scale dataset.

• We constructed a cell tracking dataset by cropping and labelling 3D images of single fluorescent zebrafish's neutrophils from time laps images.

• A comparison of our tracking accuracy with other available tracking algorithms shows that our approach performs well (about 95% accuracy rate) in relation to addressing cell tracking problems.

摘要

•We proposed a novel deep-learning-based object linkage method to tackle the common challenges involved in tracking highly amorphous and migrating leukocytes in large-scale dataset.•We constructed a cell tracking dataset by cropping and labelling 3D images of single fluorescent zebrafish's neutrophils from time laps images.•A comparison of our tracking accuracy with other available tracking algorithms shows that our approach performs well (about 95% accuracy rate) in relation to addressing cell tracking problems.

论文关键词：Cell tracking,Leukocyte,Cellular microscopic imaging,Zebrafish model,3D-cell association learning network,Deep learning

论文评审过程：Received 30 March 2020, Revised 30 May 2021, Accepted 22 June 2021, Available online 30 June 2021, Version of Record 6 July 2021.

论文官网地址：https://doi.org/10.1016/j.artmed.2021.102129