TileBio

What TileBio does

TileBio is a University of Glasgow spin-out developing self-supervised AI foundation models for digital pathology, headquartered in Glasgow, Scotland. Its core technology converts histopathology whole-slide images into an AI-learned visual language of tissue, representing each slide as a structured 'document' that a proprietary large language model can interpret. The platform is built around three tightly integrated components: a self-supervised learning backbone (e.g., Barlow Twins) trained on unlabelled images, Histomorphological Phenotype Clusters (HPCs) that produce interpretable morphology groupings, and a Tissue-Language Foundation Model that links these phenotypes to transcriptomic signatures, cell-type enrichments, and survival outcomes.

The company has no disclosed revenue and operates pre-commercially, with monetization modeled around three streams: licensing of clinical diagnostic AI to hospitals and pathology labs, research partnerships and milestone payments with pharmaceutical and biotech firms, and biomarker discovery services. Distribution is enterprise-led, combining direct outreach to clinical institutions, NHS data partnerships (NHS Greater Glasgow and Clyde, NHS West of Scotland Innovation Hub) supplying training data, and a flagship co-development agreement with Terrain Life Science to integrate spatial proteomics with TileBio's H&E and multiplex immunofluorescence algorithms. TileBio completed a £1.6M seed round in February 2026 led by Twin Path Ventures, with Scottish Enterprise and GU Holdings participating, and is actively recruiting across engineering, regulatory, and commercial functions to convert these scientific and data foundations into a regulated, revenue-generating product.

Name

TileBio

Legal name

TILEBIO LTD

Website

https://tilebio.com

Company type

Private

Founded year

2026

Operating status

Operating

Headcount range

1–10 employees

Short description

TileBio is a University of Glasgow spin-out building self-supervised AI foundation models that convert histopathology images into a visual language of tissue. It serves clinical diagnostics, pharmaceutical research, and precision-medicine customers seeking interpretable AI for cancer detection and biomarker discovery.

Ownership category

akta.pro rank

Product category

Digital Pathology AI

NAICS

Research and Development in Biotechnology (except Nanobiotechnology) (541714), Research and Development in the Physical, Engineering, and Life Sciences (except Nanotechnology and Biotechnology) (541715)

SIC

Laboratory Analytical Instruments (3826)

akta.pro primary industry

Medical Imaging AI (HDAAAEAH)

akta.pro secondary industries

SaMD – Imaging & Signal Analysis (AI/ML) (HLACANAG), Digital & AI-Enabled Diagnostics (algorithmic interpretation, decision support tied to assays) (HLAAALAO), Anatomic Pathology & Histology/Cytology IVD (Tissue Diagnostics) (HLAHAAAE), SaMD – Digital Biomarkers & Algorithmic Endpoints (HLACANAF)

Keywords

• Digital pathology AI
• Pathology foundation models
• Biomarker discovery
• Cancer diagnostics AI
• Computational pathology

Headquarters

HQ city

Glasgow

HQ country

United Kingdom

HQ region

Europe

Offices1 record

Markets served

GTM type

B2B

Offering type

Software

Cost components

Personnel, Technology or R&D, Infrastructure, Marketing or Sales, Operations

Revenue model

1. Clinical Diagnostics AI: Licensing of foundation models and AI analysis tools to clinical institutions for cancer detection, triage, and diagnostic support. Revenue generated through platform licensing and API access fees.
2. Pharmaceutical Research Partnerships: Collaboration with pharmaceutical and biotechnology companies for biomarker discovery, drug development support, and clinical trial optimization. Revenue through research partnerships, milestone payments, and ongoing collaboration fees.
3. Biomarker Discovery Services: AI-powered analysis services for revealing hidden morphological signals linked to patient outcomes and therapeutic response, enabling novel biomarker identification for research and clinical applications.

Go-to-market motion1 record

Distribution channels2 records

Marketing channels4 records

Core offering

TileBio develops self-learning AI foundation models for digital pathology that convert histopathology images into an AI-learned visual language of tissue without requiring human-annotated data. The platform delivers pathology foundation models, tissue-language foundation models, histomorphological phenotype clusters (HPCs), and H&E/multiplex immunofluorescence analysis algorithms for clinical diagnostics, biomarker discovery, and pharmaceutical research applications.

Product overview

TileBio offers a unified pathology AI platform centered on its foundation model that learns a visual language of tissue directly from unlabeled histopathology images. The core Tissue-Language Foundation Model converts whole slide images into structured 'documents' represented by learned tokens, enabling interpretation through a proprietary LLM. The Pathology Foundation Model serves as the primary engine, while Histomorphological Phenotype Clusters (HPCs) provide interpretable phenotype outputs. H&E and Multiplex Immunofluorescence Algorithms extend the platform's analytical capabilities. Together, these products support clinical triage, biomarker discovery, and pharmaceutical research applications.

Differentiator

Problem solved

Functional benefit

Products and services

• Pathology Foundation Model Self-learning AI model that learns a universal visual language of tissue directly from routine H&E slides without human annotations. Converts histopathology images into structured representations, enabling population-scale cancer detection, triage, and biomarker discovery. Targeted at clinical institutions, pharmaceutical researchers, and biotech companies.
• Tissue-Language Foundation Model Proprietary visual language model that maps tissue samples into reproducible histomorphological phenotype clusters (HPCs) that correlate with tumor biology and clinical outcomes. Uses state-of-the-art LLM techniques to interpret context and relationships within slides. Targeted at clinical diagnostics and translational research.
• Histomorphological Phenotype Clusters (HPCs) Nearest-neighbor graph clustering over tile embeddings that yields distinct, interpretable phenotype communities. Captures tumour, stromal, immune and other recognisable morphologies across all pathology with demonstrated links to transcriptomic signatures and cell-type enrichments. Targeted at clinical and pharmaceutical biomarker applications.
• H&E and Multiplex Immunofluorescence Algorithms Algorithms that analyze both H&E-stained and multiplex immunofluorescence slides, creating a proprietary visual language of cancer. Used in collaboration with Terrain Life Science for biomarker discovery and diagnostic development. Targeted at pharmaceutical and precision medicine partners.

Quantifiable outcome

• Training on 1M+ whole slide images with plans to scale to 2M+
• +2 more outcomes

Named customers3 records

Segments3 records

Ideal customer profiles3 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

AI capability8 records

Feature4 records

Partnerships

Three partnerships are on record, tiered core and flagship.

• NHS West of Scotland Innovation HubcoreTechnology or Integration · 16 February 2026Central data partnership for training and validating TileBio's foundation model on millions of digitized pathology images. The partnership uses no patient-identifiable data in model training, enabling ethical, population-scale research with strict governance and privacy controls.
• NHS Greater Glasgow and ClydecoreTechnology or Integration · 16 February 2026One of the UK's largest health boards providing clinical pathology data for AI model training. Partnership enables TileBio to access millions of digitized pathology images representing diverse, real-world clinical data for foundation model development.
• Terrain Life ScienceflagshipStrategic or Co-development Partner · 16 September 2025Strategic collaboration combining Terrain's hyperplex immunofluorescence spatial proteomics with TileBio's H&E and multiplex immunofluorescence algorithms. The partnership aims to create a powerful offering for biomarker discovery and diagnostic development, specifically addressing the challenge of delivering the right novel class of drug to the right patient. Both companies share a vision to revolutionize cancer care through precision medicine.

Scale indicators4 records

Recent moves6 records

Expansion highlights5 records

Direct peers

• Owkin: AI biotech combining pathology, multi-omic, and federated learning for biomarker discovery and drug development; overlaps directly with TileBio's pharma GTM and AI-on-histopathology approach.
• PathAI: AI-powered pathology platform serving both biopharma (biomarker discovery, clinical trial support) and clinical diagnostics; closest direct competitor in business model and customer overlap.
• Proscia: Digital pathology platform combining enterprise pathology workflow software with AI applications, including foundation-model-style approaches; comparable in both clinical and pharma life sciences customer base.
• Ibex Medical Analytics: AI pathology company with deployed clinical decision support products for cancer detection across multiple tumor types; directly comparable clinical GTM and disease coverage.
• Paige.AI: AI foundation model company for digital pathology with FDA-cleared products for prostate cancer detection; directly comparable to TileBio as a pathology foundation model play spanning clinical diagnostics and pharma research.

Emerging players

• Mindpeak: AI pathology startup focused on automated image analysis for routine diagnostic workflows in oncology; comparable technology stack and clinical customer base, though narrower in scope than a foundation model play.
• Aiosyn: AI pathology company developing computational pathology biomarkers for clinical and research use; early-stage European peer with overlapping target customers in pharma and clinical labs.
• Deep Bio: AI pathology startup with FDA-cleared prostate cancer diagnostic and ongoing expansion into other cancer types; comparable scope of clinical AI applications, with overlapping technical approach.

Broad incumbents

• Visiopharm: Established digital pathology and AI image analysis vendor with broad life sciences and clinical diagnostics portfolio; an incumbent TileBio would compete against for hospital and pharma contracts.
• Indica Labs: Long-standing digital pathology software provider with HALO image analysis platform widely used in pharma research; an incumbent infrastructure player in the same end markets TileBio targets.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat4 records

Key risks7 records

Key highlights7 records

Customer concentration

• LinkedInlinkedin.com/company/tilebio

Revenue estimate

Valuation estimate

Number of profiles

Profiles4 records

Funding overview

Funding rounds1 record

Investors3 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about TileBio