Work

Tui: A Multigenerational and Expert-Correctable Tracker for Cellular Dynamics

Year: 2026

Type: article

Abstract: Highlights • Tui tracker utilizes an integer linear programming framework that jointly models cost for cell migration, mitosis, fusion, appearance, and disappearance, ensuring globally optimal lineage construction. Especially, fusion detection is capability absent in most existing state-of-the-art trackers. • A graphical user interface that allows manual inspection and correction of lineage trees, integrating biological expertise with automation for reliable cell-lineage analysis. • Broad applicability with annotation of the cell-tracking challenge format for lineage-association analysis and migration of 2D objects. Validated against synthetic data, the CTC DIC-C2DH-HeLa and Fluo-N2DH-SIM+ datasets, and the glioblastoma T98G electrotaxis dataset, achieving near-perfect tracking accuracy (TRA = 0.984, 0.999, and 0.999, respectively) and allowing high-content biological discovery.

Source: Computational and Structural Biotechnology Journal

DOI: 10.34133/csbj.0054

Authors: I-Ming Chang, Hsieh-Fu Tsai

Institutions: Chang Gung University, Chang Gung Memorial Hospital, Keelung Chang Gung Memorial Hospital

Language: English

Topic: Cell Image Analysis Techniques

Subfield: Biophysics

Field: Biochemistry, Genetics and Molecular Biology

Domain: Life Sciences

Open Access status: gold

Estimated APC paid: $3,230

Funders: National Science and Technology Council, National Science and Technology Council