EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

SHEST: single-cell-level artificial intelligence from haematoxylin and eosin morphology for cell-type prediction and spatial transcriptomics reconstruction

Hoyeon Jeong iD, Junghan Oh iD, Donggeon Lee iD, Jae Hwan Kang iD, Yoon-La Choi iD

Briefings in Bioinformatics · 2026

DOI: 10.1093/bib/bbag037 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2026. Verfasst von Hoyeon Jeong und 4 weiteren Autor:innen. Beteiligte Einrichtung: Medical Research Institute, Sungkyunkwan University, 81 Irwon-Ro, 06351 Seoul, Republic of Korea. Hauptthemen laut MEDLINE: Artificial Intelligence, Single-Cell Analysis, Eosine Yellowish-(YS), Lung Neoplasms. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National R&D Program for Cancer Control, Ministry of Health & Welfare, Republic of Korea, Samsung Medical Center.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

A comprehensive understanding of cancer progression requires integrating tissue morphology with spatial molecular profiles. We present SHEST, a multi-task profiling framework that leverages haematoxylin and eosin morphology to predict cellular composition and reconstruct spatial gene expression at single-cell resolution. SHEST employs a quadruple-tile input capturing nuclear and contextual information, combined with a neighbourhood-informed clustering algorithm to filter ambiguous cellular signals. It comprises a shared morphological encoder with two task-specific heads: a classifier for cell-type prediction and a reconstructor for gene expression. Multi-task optimization uses cross-entropy and zero-inflated negative binomial losses, specifically addressing the sparsity of spatial transcriptomic data. Evaluation on human lung adenocarcinoma datasets demonstrated high accuracy for the principal reciprocal constituents of the tumour–immune axis ($F_{1}$: 0.97 for tumour cells and 0.91 for lymphocytes). External validation confirmed its generalizability, revealing alveolar cells and their early neoplastic transitions. Reconstructed gene expression reproduced spatially resolved, cell-type-specific marker patterns—EPCAM in tumour cells, LTBP2 in fibroblasts, and CD3E in lymphocytes—recovering biologically coherent transcriptional architecture. SHEST also preserved distance-dependent spatial relationships and gene-level autocorrelation, reflecting the multicellular niche structure of the tumour microenvironment. By unifying cell-type identification, gene expression reconstruction, and spatial mapping within a single interpretable framework, SHEST provides a synergistic and cost-efficient bridge between histopathology and spatial transcriptomics. This approach facilitates comprehensive tissue characterization and forms a foundation for precision oncology through spatially informed, cell-level insights into tumour–immune ecosystems.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Artificial IntelligenceSingle-Cell AnalysisEosine Yellowish-(YS)Lung NeoplasmsHumansSpatial TranscriptomicsAlgorithmsSingle-Cell Gene Expression AnalysisGene Expression ProfilingTumor Microenvironment

Stichwörter der Autor:innen

cell-type predictioncomputational pathologydigital pathologygene expression reconstructionhaematoxylin and eosinlung adenocarcinomaspatial transcriptomicstumour microenvironment

Substanzen

Eosine Yellowish-(YS)

Autor:innen und Institutionen

  • Medical Research Institute, Sungkyunkwan University, 81 Irwon-Ro, 06351 Seoul, Republic of Korea.
  • Research Institute for Future Medicine, Samsung Medical Center, 81 Irwon-Ro, 06351 Seoul, Republic of Korea.
  • Department of Health Sciences and Technology, Samsung Advanced Institute for Health Sciences and Technology, 81 Irwon-Ro, 06351 Seoul, Republic of Korea.
  • Department of Digital Health, Samsung Advanced Institute for Health Sciences and Technology, 81 Irwon-Ro, 06351 Seoul, Republic of Korea.
  • Department of Pathology and Translational Genomics, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-Ro, 06351 Seoul, Republic of Korea.

ORCID-Profile: Hoyeon Jeong, Junghan Oh, Donggeon Lee, Jae Hwan Kang, Yoon-La Choi

Förderung

  • National R&D Program for Cancer Control
  • Ministry of Health & Welfare, Republic of Korea
  • Samsung Medical Center

Zitieren

APA

Jeong, H., Oh, J., Lee, D., Kang, J. H., & Choi, Y. L. (2026). SHEST: single-cell-level artificial intelligence from haematoxylin and eosin morphology for cell-type prediction and spatial transcriptomics reconstruction. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag037

Vancouver

Jeong H, Oh J, Lee D, Kang JH, Choi YL. SHEST: single-cell-level artificial intelligence from haematoxylin and eosin morphology for cell-type prediction and spatial transcriptomics reconstruction. Briefings in Bioinformatics. 2026. doi:10.1093/bib/bbag037

BibTeX
@article{Jeong2026,
  author = {Jeong, Hoyeon and Oh, Junghan and Lee, Donggeon and Kang, Jae Hwan and Choi, Yoon-La},
  title = {SHEST: single-cell-level artificial intelligence from haematoxylin and eosin morphology for cell-type prediction and spatial transcriptomics reconstruction},
  journal = {Briefings in Bioinformatics},
  year = {2026},
  doi = {10.1093/bib/bbag037}
}
Kostenlos starten →