Towards Cellular-Scale Interpretability in Pathology Foundation Models for Biomarker Assessment
arXiv:2511.05150v2 Announce Type: replace-cross Abstract: Molecular biomarker testing in pathology is often costly and tissue-consuming, limiting scalable clinical deployment. Artificial intelligence applied to hematoxylin and eosin (HE)-stained histology could enable rapid biomarker screening, but clinical translation requires models that are both accurate and interpretable. Here we introduce Hireca, a biomarker-focused pathology foundation model pretrained on more than 80,000 whole-slide images spanning 38 organ types from three medical centers, together with CytoMap, an interpretability mod
Advances in AI, particularly foundation models and interpretability methods, are enabling new applications in complex fields like medical diagnostics, addressing previous barriers to clinical translation.
This development represents a significant step towards scalable, cost-effective, and interpretable AI for medical diagnostics, potentially transforming biomarker assessment and personalized medicine.
The ability to assess molecular biomarkers cheaply and efficiently through AI on standard histology slides greatly reduces reliance on expensive and tissue-consuming traditional methods.
- · Medical AI developers
- · Healthcare providers
- · Patients
- · Pharmaceutical research
- · Traditional molecular biomarker testing companies
- · Pathologists with high dependence on manual biomarker review
Reduced cost and increased accessibility of molecular biomarker testing for various diseases.
Accelerated drug discovery and development due to more efficient patient stratification and biomarker analysis.
Potential for AI-driven disease screening programs at a population scale, leading to earlier detection and better health outcomes.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI