EmbSAM: cell boundary localization and Segment Anything Model for fast images of developing embryos
Bibliographic record
Abstract
Cellular shape dynamics are critical for understanding cell fate determination and organogenesis during development. However, fluorescence live-cell images of cell membranes frequently suffer from a low signal-to-noise ratio, especially during long-duration imaging with high spatiotemporal resolutions. The low ratio is caused by a tradeoff between embryo viability and phototoxicity and photobleaching of fluorescent markers, which hinders effective cell shape reconstruction, particularly in rapidly developing embryos. Here, we devise an integrative computational framework, EmbSAM, that incorporates a deep-learning-based cell boundary localization algorithm and the Segment Anything Model. EmbSAM enables accurate segmentation of three-dimensional cell membrane images for roundworm Caenorhabditis elegans embryos imaged with exceptional temporal resolution, i.e., every 10 seconds per stack. The resolved cell shapes prior to gastrulation quantitatively characterize a series of cell-division-coupled morphodynamics associated with cell position, cell division phase duration, cell division axis reorientation, cell identity, lineage, fate, among others, which can be accessed locally and online. By integrating a deep-learning-based cell boundary localization algorithm with the Segment Anything Model, EmbSAM enables accurate 3D cell membrane segmentation across C. elegans embryos and quantitative analysis of cellular-to-multicellular morphodynamics at 10-second resolution.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".