HELIOS: A spatial omics-informed multi-organ cell atlas enables AI-based tumour microenvironment profiling and biomarker discovery from routine histopathology
Source check: Source checked. Not independent scientific confirmation.
Finding: Spatial-omics cell identities supervise a YOLOv10-m detector for routine H&E; the study reports cross-dataset cell-classification benchmarks and cancer-cohort associations. Prognostic analyses are retrospective and do not establish clinical utility.
Technology: YOLOv10-m detector trained with spatial-omics-derived cell labels
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Spatial transcriptomics labels transferred to matched H&E train a multiclass YOLOv10-m cell detector; the learned operation supplies cell identity and density features for downstream analysis.
- Test and result
- The paper reports 12.9 million cells from 60 specimens across 14 tissue types and eight cell classes; the held-out evaluation covers about 1.23 million cells, with performance varying by class. It reports cross-dataset F1 values of 0.66 (Lizard), 0.78 (MoNuSAC), and 0.59 (NuCLS), plus retrospective cancer-specific survival associations.
- Limitation / still unknown
- Methods specify five-fold train/validation/test evaluation and external benchmark comparisons. Prognostic models use retrospective TCGA-type cohorts and adjustment; authors state clinical utility remains unproven and subgroup analyses are exploratory.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://www.medrxiv.org/content/10.64898/2026.10.26.26366087v1.full.pdf · https://www.medrxiv.org/content/10.64898/2026.10.26.26366087v1
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.