Automahout: an autonomous image analysis system for cell tracking

Automahout, an autonomous agent for the Mastodon/ELEPHANT platform, produced human-free cell lineages on three time-lapse datasets and achieved TRA scores of 0.95–0.96.

• Why it matters: Cell tracking across new time-lapse datasets often needs manual tuning for detection, linking, and proofreading, creating a bottleneck for scalable analysis.
• What they did: The system iteratively tests existing models, annotates reference cells when needed, trains a detector from sparse annotations, reviews outputs against references, and feeds corrections back until stopping criteria are met.
• The result: Across a 2D fluorescence culture, a 2D phase-contrast culture, and a 3D embryo dataset, it generated lineages without human input and reached TRA scores of 0.95–0.96.

AI summary of the abstract · may contain errors · not medical advice · read the paper.