Cell Lineage
We used visualization techniques, human-computer interaction methods, and machine learning to help biologists to construct cell lineage effectively.
Backgrounds: Cell lineage is the development of a (plant) embryo from a single ovum cell. Based on a confocal microscopy dataset, traditionally biologists manually constructed the cell lineage, starting from this observation and reasoning backward in time to establish their inheritance.
Challenges:
- 1. Cells in the embryo are densely packed together.
- 2. The traditional lineage construction process is tedious and time-consuming.
Solutions: We first explored selection interaction techniques in 3D environments. Then, We developed a specialized tool for biologists to construct cell lineages. This tool features a hierarchical tree built using both top-down and bottom-up approaches. The top-down approach follows the original temporal development, while the bottom-up approach aligns with the traditional cell lineage-building method. We also made use of machine learning (ML) models trained on a database of manually established cell lineages to assist the biologist in cell assignments. Most biologists, however, are not familiar with ML, nor is it clear to them which model best predicts the embryo's development. We thus enhanced the system to support biologists in exploring and comparing ML models, checking the model predictions, detecting possible ML model mistakes, and deciding on the most likely embryo development.
References
|
TVCG 2024 Full Paper
Visualizing and Comparing Machine Learning Predictions to Improve Human-AI Teaming on the Example of Cell Lineage.
IEEE Transactions on Visualization and Computer Graphics, 30(4):1956–1969, April 2024.
10.1109/TVCG.2023.3302308
|
|
EuroVis'22 / CGF 2022 Full Paper
LineageD: An Interactive Visual System for Plant Cell Lineage Assignments based on Correctable Machine Learning.
Computer Graphics Forum, 41(3):195–207, June 2022.
10.1111/cgf.14533
|
|
GI'21 Full Paper
Design and Evaluation of Three Selection Techniques for Tightly Packed 3D Objects in Cell Lineage Specification in Botany.
In Proceedings of Graphics Interface (GI, May 27–28, virtually in Vancouver, BC, Canada), pages 213–223, Canadian Human-Computer Communications Society, Mississauga, ON, Canada, 2021.
10.20380/GI2021.33
|
