Cellular cell tracking and pedigree reconstruction method fusing local topology and coordinate decoupling
By integrating local topology and coordinate decoupling, a method for cellular cell tracking and lineage reconstruction was constructed, which solved the problem of cell tracking failure caused by field of view drift and topological mutation in long-term microscopy observation, and achieved high-fidelity and automated cell lineage reconstruction.
Patent Information
- Application Number
- CN202610545056.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-23
- Publication Date
- 2026-07-24
AI Technical Summary
Existing technologies cannot effectively address cell tracking failures caused by field-of-view drift and topological mutations during long-term microscopic observations. This is especially true in biological tissues with dense honeycomb structures, such as Porphyra yezoensis, where existing methods suffer from poor robustness, low tracking accuracy, and insufficient depth of biological logic mining.
A method combining local topology and coordinate decoupling is adopted. By constructing a local topological neighborhood graph based on surround sorting, a coordinate decoupling feature matrix composed of vector angle, distance ratio and area ratio is extracted. Combined with a global optimization algorithm and a reverse virtual fusion mechanism, cross-frame correlation and split detection are achieved.
High-fidelity and automated cell tracking and lineage reconstruction were achieved in long-term, unaligned microscopic image sequences, ensuring the robustness and biological rationality of cell identification and overcoming the problems of topological mutation and feature confusion.
