{"entity": "publication", "iuid": "7175e117d1f344b3a30632f37a354124", "timestamp": "2026-11-15T13:04:48.947Z", "links": {"self": {"href": "https://publications-affiliated.scilifelab.se/publication/7175e117d1f344b3a30632f37a354124.json"}, "display": {"href": "https://publications-affiliated.scilifelab.se/publication/7175e117d1f344b3a30632f37a354124"}}, "title": "Scellseg: A style-aware deep learning tool for adaptive cell instance segmentation by contrastive fine-tuning.", "authors": [{"family": "Xun", "given": "Dejin", "initials": "D"}, {"family": "Chen", "given": "Deheng", "initials": "D"}, {"family": "Zhou", "given": "Yitian", "initials": "Y"}, {"family": "Lauschke", "given": "Volker M", "initials": "VM"}, {"family": "Wang", "given": "Rui", "initials": "R"}, {"family": "Wang", "given": "Yi", "initials": "Y"}], "type": "journal article", "published": "2022-12-22", "journal": {"title": "iScience", "issn": "2589-0042", "volume": "25", "issue": "12", "pages": "105506", "issn-l": null}, "abstract": "Deep learning-based cell segmentation is increasingly utilized in cell biology due to the massive accumulation of large-scale datasets and excellent progress in model architecture and instance representation. However, the development of specialist algorithms has long been hampered by a paucity of annotated training data, whereas the performance of generalist algorithms is limited without experiment-specific calibration. Here, we present Scellseg, an adaptive pipeline that utilizes a style-aware pre-trained model coupled to a contrastive fine-tuning strategy that also learns from unlabeled data. Scellseg achieves state-of-the-art transferability in average precision and Aggregated Jaccard Index on disparate datasets containing microscopy images at three biological levels, from organelle, cell to organism. Interestingly, when fine-tuning Scellseg, we show that performance plateaued after approximately eight images, implying that a specialist model can be obtained with few manual efforts. For convenient dissemination, we develop a graphical user interface that allows biologists to easily specialize their self-adaptive segmentation model.", "doi": "10.1016/j.isci.2022.105506", "pmid": "36425762", "labels": {"DDLS Fellow": null, "Yitian Zhou": null}, "xrefs": [{"db": "pmc", "key": "PMC9678729"}, {"db": "pii", "key": "S2589-0042(22)01778-3"}], "notes": [], "created": "2026-11-06T09:32:14.717Z", "modified": "2026-11-06T09:32:14.724Z"}