C. Elegans Detection, Segmentation, and Counting

Shu Kong, Edmund Florendo, Zhongxuan Mou, Olivier Cinquin, Charless Fowlkes

Page is still under construction, last update Oct. 26, 2016

Caenorhabditis elegans is a microscopic nematode that is used in a wide variety of biological studies and provides one of the single best model systems to study lifespan development. The small size of C. elegans makes it possible to grow them in large number, however this also provides a challenge as manual observation can be tedious and time-consuming. To fully realize the potential for high-throughput studies with strong statistical power, it would be very advantageous to automate the acquisition of important information about worm lifespan and the timing of reproduction under conditions of high population density. This project aims to develop a robust imaging system for automatically analyzing images collected in order to detect, classify and catalog the biological specimens such as C. elegans. Vision-wise, a unified model is involved to segment individual worms to automatically count the number of larvae and adult worms. This model contains keypoint detection, semantic segmentation, patch-based counting, instance segmentation, etc.

Updated Visualization

Reference

  • N Stroustrup, et al., "The Caenorhabditis elegans lifespan machine", Nature methods, 2013.