Engineering Papers⌕ Search
Home/Archive/DOE OSTI

DOE OSTI · 2572801

SwinCell: a 3D transformer and flow-based framework for improved cell segmentation

Zhang, Xiao [Brookhaven National Laboratory (BNL), Upton, NY (United States)] (ORCID:0000000275840712)·Lin, Zihan [Brookhaven National Laboratory (BNL), Upton, NY (United States)]·Wang, Liguo [Brookhaven National Laboratory (BNL), Upton, NY (United States)]·Chu, Yong S. [Brookhaven National Laboratory (BNL), Upton, NY (United States). National Synchrotron Light Source II (NSLS-II)]·Yang, Yang [Brookhaven National Laboratory (BNL), Upton, NY (United States). National Synchrotron Light Source II (NSLS-II)] (ORCID:0000000153185385)·Xiao, Xianghui [Brookhaven National Laboratory (BNL), Upton, NY (United States). National Synchrotron Light Source II (NSLS-II)] (ORCID:0000000271423452)·Lin, Yuewei [Brookhaven National Laboratory (BNL), Upton, NY (United States)] (ORCID:0000000214294543)·Liu, Qun [Brookhaven National Laboratory (BNL), Upton, NY (United States). National Synchrotron Light Source II (NSLS-II)] (ORCID:000000021179290X)
View original source ↗Read source document ↗Export metadata ↓

Abstract

Segmentation of three-dimensional (3D) cellular images is fundamental for studying and understanding cell structure and function. However, 3D cellular segmentation is challenging, particularly for dense cells and tissues. This challenge arises mainly from the complex contextual information within 3D images, anisotropic properties, and the sensitivity to internal cellular structures, which often lead to incorrect segmentation. In this work, we introduce SwinCell, a 3D transformer-based framework that leverages Swin-transformer to predict flow and differentiate individual cell instances. We demonstrate SwinCell’s utility in the segmentation of nuclei, colon tissue cells, and densely cultured cells. SwinCell strikes a balance between maintaining detailed local feature recognition and understanding broader contextual information. Through extensive testing with both public and in-house 3D cell imaging datasets, SwinCell shows utility in segmenting dense cells, making it a valuable tool for 3D segmentation in cellular analysis that could expedite research in cell biology and tissue engineering.

Explore related subjects

59 BASIC BIOLOGICAL SCIENCEScellular imagingimage processingmachine learning

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Zhang, Xiao [Brookhaven National Laboratory (BNL), Upton, NY (United States)] (ORCID:0000000275840712), Lin, Zihan [Brookhaven National Laboratory (BNL), Upton, NY (United States)], Wang, Liguo [Brookhaven National Laboratory (BNL), Upton, NY (United States)], Chu, Yong S. [Brookhaven National Laboratory (BNL), Upton, NY (United States). National Synchrotron Light Source II (NSLS-II)], Yang, Yang [Brookhaven National Laboratory (BNL), Upton, NY (United States). National Synchrotron Light Source II (NSLS-II)] (ORCID:0000000153185385), Xiao, Xianghui [Brookhaven National Laboratory (BNL), Upton, NY (United States). National Synchrotron Light Source II (NSLS-II)] (ORCID:0000000271423452), Lin, Yuewei [Brookhaven National Laboratory (BNL), Upton, NY (United States)] (ORCID:0000000214294543), Liu, Qun [Brookhaven National Laboratory (BNL), Upton, NY (United States). National Synchrotron Light Source II (NSLS-II)] (ORCID:000000021179290X). 2025-07-01. SwinCell: a 3D transformer and flow-based framework for improved cell segmentation. https://doi.org/10.1038/s42003-025-08397-x

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related reports

DOE OSTI2026

Soil metagenomics umbrella narrative

Implementing accessible, authentic research experiences in introductory courses is challenging, particularly at institutions serving diverse student populations. To address this gap, we developed and deployed a Course-based Undergraduate Research Experience (CURE) focused on plant-microbe interactions in General Biology II at Northeastern Illinois University (NEIU), a minority-serving institution with a diverse student body. Students grew sugar beets (Beta vulgaris), extracted DNA from the rhizoplane, and used the Department of Energy Systems Biology Knowledgebase (KBase) for bioinformatic analysis to compare microbial relative abundance in fertilized versus unfertilized soil. Over five semesters, the CURE engaged 103 students and leveraged the intuitive KBase platform to make complex sequencing data accessible. Pre/post-course survey data revealed significant increases in student self-assessed research skills, including the ability to explain results and determine the types of data to collect. Furthermore, students reported significant gains in confidence related to experimental design and hypothesis development, alongside a strong increase in familiarity with KBase. Informal faculty feedback indicated high student engagement and appreciation for the real-world connections (e.g. food systems, agriculture, and health). This scalable, low-cost model effectively integrates data science tools into the foundational curriculum, demonstrating a potent strategy for boosting research skills and broadening participation in authentic scientific inquiry among diverse undergraduate students.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Genome-resolved insights into microbial diversity and elemental cycling in Winogradsky columns

We retained 18 MAGs with ≥50% completion and <10% contamination (i.e., at least medium quality). Of these, 10 had >90% completion and <5% contamination; however, only one (Paceibacteria Bin.003_MG) can be described as high-quality, as the others lacked a full suite of 5S, 16S, and 23S rRNA genes. To maximize the diversity of our recovered MAGs, we also retained one MAG (Chromatiaceae Bin.008_AM) with >40% (but less than 50%) completion and <5% contamination, as well as one (Rhodopseudomonas Bin.015_MK) with >90% completion and <20% (but>10%) contamination. Interestingly, significant chimerism was not detected in this MAG (40) , suggesting that the elevated contamination (20%) may instead reflect two closely related strains collapsing into a single bin. Consistent with this, contig coverage was bimodal, with roughly 17% of the assembly at ~115x and the remaining 83% at ~282x, while GC content remained uniform across both groups (~64%), arguing against contamination from a taxonomically distinct source.

59 BASIC BIOLOGICAL SCIENCES↗