Digital pathology and AI-driven computational biomarkers in 2026: overview
Digital pathology turns tissue slides into high-resolution digital images that can be reviewed and analyzed computationally. AI models use these images to identify tissue patterns, classify disease and estimate biomarkers that might otherwise require separate tests. This creates opportunities to support diagnosis and treatment decisions, but clinical utility depends on reliable performance across slides, scanners, patient groups and laboratory workflows.
Recent work is moving from fixed image classification toward biological inference and more flexible analysis. In September 2026, a St Jude Children’s Research Hospital study described learned image resizing to preserve tissue architecture during model training, while a Sun Yat Sen University study explored a vision-language-guided model for predicting Helicobacter pylori antibiotic resistance from routine histology. Also in September 2026, Leica Biosystems added Tempus pathology products to its Aperio AI Store, while a news report described physics-guided super-resolution for cancer screening.
- 1From morphology to biology: Molecular phenotype inference and treatment-prediction models are pushing pathology AI beyond visual classification toward clinically relevant signals in routine tissue images.
- 2Spatial context matters: Analysis of tissue microenvironments is gaining momentum because cell arrangement and neighboring tissue features can add information that isolated image patches miss.
- 3Validation remains central: Slide-selection uncertainty, image artifacts and differences in acquisition conditions make robust quality control and external validation essential for clinical use.
Top organizations working on digital pathology and AI-driven computational biomarkers
Ranked by Linknovate activity score · last 24 monthsThe activity score adds up the relevance of every patent, publication, grant and news item linking an organization to digital pathology and AI-driven computational biomarkers. Showing 7 of 3,000 organizations.
| # | Organization | Approach | Signals (24 mo) | Momentum | Why it matters |
|---|---|---|---|---|---|
| 1 | 
U.S. Food and Drug Administration
Other · United States |
General | PUB 6 · NEWS 664 | -69% | Recently, the FDA cleared Hamamatsu NanoZoomer slide-scanner systems through its 510(k) pathway, applying regulatory review to equipment used in digital pathology. |
| 2 | 
MayoClinic
Research · United States |
Spatial microenvironment analysis | PAT 2 · PUB 159 NEWS 98 |
-4% | Mayo Clinic researchers apply AI to whole-slide images, including interpretable methods for identifying spatial tissue patterns and discovering biomarkers relevant to precision oncology. |
| 3 | 
Fudan University
Research · China |
Histologic classification and grading | PAT 1 · PUB 181 NEWS 1 |
+28% | Fudan University research includes AI methods for labeling and analyzing whole-slide images, alongside work on computational pathology systems that extend beyond fixed classification tasks. |
| 4 | 
Southern Medical University
Research · China |
Prognostic and predictive modeling | PUB 161 | +56% | Southern Medical University researchers use digital pathology to model cancer prognosis and immunotherapy response, and develop image-analysis methods for assessing tissue and immune features. |
| 5 | 
Sun Yat Sen University
Research · China |
Molecular phenotype inference | PUB 170 · NEWS 6 | +45% | Sun Yat Sen University researchers investigate vision-language and foundation-model methods that infer clinically relevant phenotypes, including antibiotic resistance, from routine histology. |
| 6 | 
Hoffmann-La Roche
Corporate · Switzerland |
Molecular phenotype inference | PAT 14 · PUB 12 NEWS 281 |
-33% | Roche's diagnostics and pharmaceutical businesses work across pathology, biomarker testing, and personalized medicine, creating potential applications for computational analysis of tissue images. |
| 7 | 
PathAI Inc
Startup · United States |
Histologic classification and grading | PAT 4 · PUB 10 NEWS 144 |
-5% | PathAI develops AI tools for cancer diagnosis and pathology workflows, applying image analysis to support consistent interpretation of histologic features and clinically relevant biomarkers. |
| 8 | ··Hidden organizationCompany | Approach | PAT 00 · PUB 00 GRANTS 0 · NEWS 00 | +00% | Available to registered users. |
| 9 | ··Hidden organizationCompany | Approach | PAT 00 · PUB 00 GRANTS 0 · NEWS 00 | +00% | Available to registered users. |
| 10 | ··Hidden organizationCompany | Approach | PAT 00 · PUB 00 GRANTS 0 · NEWS 00 | +00% | Available to registered users. |
Startups to watch and leading research centers
Startups to watch 43 total
- PathAI IncUnited StatesPathAI develops AI tools for cancer diagnosis and pathology workflows, applying image analysis to support consistent interpretation of histologic features and clinically relevant biomarkers.
Proscia Inc.United StatesRecently, Proscia's Concentriq platform was adopted by Unilabs for digital pathology, and the company expanded integrations that help partners deploy AI tools in pathology workflows.
Lunit Inc.South KoreaLunit develops AI for cancer diagnostics and precision oncology, including pathology-based assessment of biomarkers and prediction of treatment response.- ··Hidden startupCountryAvailable to registered users, with funding and founding year.
- ··Hidden startupCountryAvailable to registered users, with funding and founding year.
Leading research institutions 2,621 total
- Southern Medical UniversityChinaSouthern Medical University researchers use digital pathology to model cancer prognosis and immunotherapy response, and develop image-analysis methods for assessing tissue and immune features.
- Sun Yat Sen UniversityChinaSun Yat Sen University researchers investigate vision-language and foundation-model methods that infer clinically relevant phenotypes, including antibiotic resistance, from routine histology.
- Fudan UniversityChinaFudan University research includes AI methods for labeling and analyzing whole-slide images, alongside work on computational pathology systems that extend beyond fixed classification tasks.
- ··Hidden research institutionCountryAvailable to registered users, with their key researchers.
- ··Hidden research institutionCountryAvailable to registered users, with their key researchers.
Where digital pathology and AI-driven computational biomarkers innovation happens
Share of documents with at least one organization from each country.
Open the interactive map with all 3,000 organizationsMain technical approaches
Last 24 months| Approach | What it is | Documents | Leading organizations |
|---|---|---|---|
| Morphometric quantification | Measures tissue structures and cell features; it produces interpretable metrics but can be sensitive to staining, scanning and segmentation variability. | 380 |
Medical University of Vienna, ZhejiangUniversity School ofMedicine, Medical College of Wisconsin |
| Histologic classification and grading | Classifies tissue patterns or disease grade; it supports routine assessment but may miss molecular differences and struggle across institutions. | 1,592 |
Pathology, Hoffmann-La Roche, U.S. Food and Drug Administration |
| Molecular phenotype inference | Predicts molecular features from tissue morphology; it may reduce testing burdens but requires validation against reference assays and diverse cohorts. | 1,586 |
National Cancer Institute, University of Chicago, Google |
| Prognostic and predictive modeling | Estimates outcomes or treatment response from pathology images; it can inform care choices but needs evidence that predictions generalize to clinical populations. | 1,895 |
MayoClinic, Stanford University, Southern Medical University |
| Spatial microenvironment analysis | Maps relationships among cells and tissue regions; it captures contextual biology but can demand substantial computation and careful interpretation. | 1,141 |
MayoClinic, Lunit Inc., Google |
Emerging trends
Last 12 months vs previous 36 monthsTerms whose frequency in the last 12 months grew the most compared with the previous 36 months.
Funding and investment
Recent funding rounds
Since 2020, the field has received $150,645,980 across 76 grants. These figures summarize recorded grants and do not establish that every award was dedicated exclusively to computational pathology.
Patent landscape
Top patent holders · last 24 monthsRecent developments
Latest news, duplicates removed- Sep 2026VISTA, Calif., SEPTEMBER 24, 2026 – Leica Biosystems, a Danaher company and a global leader in end-to-end anatomic and digital pathology solutions, announces the addition of two Paige digital pathology AI-enabled…The Medical News
- Sep 2026Digital pathology relies on high-resolution imaging for accurate cancer diagnosis, yet high-magnification systems remain slow and costly for large-scale screening. Meanwhile, standard AI super-resolution often generates…The Medical News
- Aug 2026Digital pathology succeeds when laboratories treat it as a clinical transformation programme — not simply a scanner purchase. This roadmap explains how to move from initial planning to validated, sustainable diagnostic…Medicine on Medium
- Jun 2026Dublin, June 30, 2026 (GLOBE NEWSWIRE) -- The "AI In Pathology - Market Share Analysis, Industry Trends & Statistics, Growth Forecasts (2026-2031)" has been added to ResearchAndMarkets.com's offering. The Artificial…GlobeNewswire - Research Analysis And Reports
- Month 20XXHidden developmentAvailable to registered users with alerts enabled.
Frequently asked questions about digital pathology and AI-driven computational biomarkers
What companies are working on digital pathology and AI-driven computational biomarkers?
PathAI, Proscia and Lunit are named companies working in digital pathology and AI-driven analysis. PathAI develops tools for cancer diagnosis and pathology workflows, Proscia provides a digital pathology platform, and Lunit works on cancer diagnostics and precision oncology. Larger organizations including Hoffmann-La Roche and Leica Biosystems also have activity in pathology products and AI integration.
Which digital pathology and AI-driven computational biomarker startups are active in 2026?
PathAI, Proscia and Lunit are among the active startups identified in the field. PathAI focuses on diagnostic and workflow tools, Proscia on digital pathology infrastructure and integrations, and Lunit on cancer diagnostics and precision oncology. The dataset identifies 43 startups overall, but it does not rank them by commercial prospects or clinical performance.
Which universities and research centers lead digital pathology and AI-driven computational biomarker research?
Southern Medical University, Sun Yat Sen University and Fudan University are prominent research organizations in the field. Their work includes prognosis and treatment-response modeling, molecular phenotype inference, and whole-slide image analysis. Mayo Clinic, Stanford University and Memorial Sloan Kettering Cancer Center are also active in spatial analysis and pathology-based prediction.
How much funding has digital pathology and AI-driven computational biomarkers received?
The field has recorded $150,645,980 across 76 grants since 2020. This grant total is distinct from the separate funding-round figures in the data, which may include activity not specific to computational pathology. The available information does not identify a complete, field-specific total for private investment.
Who holds the most digital pathology and AI-driven computational biomarker patents?
Paige.Ai leads the listed patent holders with 30 patents, followed by Ventana Medical Systems with 25. Genentech holds 18, Hoffmann-La Roche 14 and Foundation Medicine 13. These figures identify leading holders in the dataset; they do not indicate patent quality, enforceability or commercial use.
Is digital pathology and AI-driven computational biomarker patent activity growing?
No: patent activity declined 52% in the latest 12-month comparison with the previous 12 months. The dataset records 262 patents in its defined 24-month window. This contrasts with publication activity, which grew 19.6% in the latest 12-month comparison, suggesting research output and patenting are moving at different rates.
What are the latest developments in digital pathology and AI-driven computational biomarkers?
Recent developments include a September 2026 St Jude Children’s Research Hospital study on learned image resizing for histopathology models and a September 2026 Sun Yat Sen University study using a vision-language-guided model to predict antibiotic resistance from routine histology. In September 2026, Leica Biosystems also added Tempus pathology products to its Aperio AI Store.
What are the main approaches in digital pathology and AI-driven computational biomarkers, and how do they differ?
The main approaches are morphometric quantification, histologic classification and grading, molecular phenotype inference, prognostic and predictive modeling, and spatial microenvironment analysis. They range from measuring visible tissue features to inferring biology, outcomes or cell relationships. Recent activity grew fastest in prognostic and predictive modeling, while classification and grading activity declined in the latest 12-month comparison.
Which countries lead digital pathology and AI-driven computational biomarkers?
The United States leads with 14,890 recorded documents, followed by China with 7,795 and the United Kingdom with 5,049. Germany is also a leading contributor, with 4,460 documents. These country figures reflect recorded activity and do not by themselves measure clinical adoption, technology quality or market share.
What are the applications of digital pathology and AI-driven computational biomarkers?
The main applications include cancer diagnosis and grading, precision oncology, prognosis and treatment assessment, drug development and clinical trials, and non-oncology tissue diagnostics. Cancer diagnosis and grading has the largest recorded activity, while precision oncology and prognosis and treatment assessment are also substantial areas. Models may support biomarker prediction, but they do not replace clinical validation or pathologist review.