DiaDeep

Name

DiaDeep

Legal name

DIADEEP SAS

Website

https://diadeep.com

Company type

Private

Founded year

2022

Operating status

Operating

Headcount range

11–50 employees

Ownership category

akta.pro rank

Product category

Digital Pathology AI

NAICS

Medical Laboratories (621511), Software Publishers (5132), Research and Development in Biotechnology (except Nanobiotechnology) (541714)

SIC

In Vitro & In Vivo Diagnostic Substances (2835), Services-Medical Laboratories (8071)

akta.pro primary industry

Molecular Pathology & Tissue Biomarkers (IHC/ISH, digital pathology-linked markers) (HLAAAMAI)

akta.pro secondary industries

Cytology & Histopathology Diagnostics (IHC, ISH, tissue-based assays) (HLAAALAI), Pathology & Histopathology Services (HLAGAFAI), Multiplex & High-Throughput Diagnostics Platforms (microarrays, digital PCR, lab automation-enabled assays) (HLAAALAL)

Keywords

• Digital pathology AI
• Cancer biomarker quantification
• Whole slide imaging
• Computational pathology software
• AI prognostic assays

Headquarters

HQ city

Lyon

HQ country

France

HQ region

Europe

Offices2 records

Markets served

GTM type

B2B

Offering type

Software

Cost components

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

Revenue model

1. Software licensing (CE-IVDR certified medical devices): Commercial licensing of CE-IVDR certified AI algorithms (DiaKwant™ and DiaSurv™) to pathology laboratories and hospitals for research use and clinical support.
2. Distribution partnership revenue: Revenue sharing with Roche through navify® Digital Pathology platform distribution in US, Middle East, and Europe.
3. Research collaborations and consortium partnerships: Partnerships with academic institutions and hospitals for validation studies and clinical research.

Pricing tiers

Go-to-market motion3 records

Distribution channels3 records

Marketing channels5 records

Core offering

DiaDeep develops AI-powered software for cancer diagnostics applied to digitized whole-slide histopathology images. The company sells two core product lines: DiaKwant™, a CE-IVDR certified biomarker quantification platform that automatically scores PD-L1 (CPS/TPS), HER2, Ki-67, ER/PR expression and detects mitoses; and DiaSurv™, a prognostic assay that predicts survival outcomes and stratifies cancer patients into risk groups from standard H&E histology slides across melanoma, colon, breast, and ovarian cancer indications.

Product overview

DiaDeep offers two core AI product lines: DiaKwant™ for biomarker quantification (PD-L1 CPS/TPS, HER2, Ki-67, ER/PR, mitosis) and DiaSurv™ for prognostic risk stratification (melanoma, colorectal, breast, ovarian cancer). Both products analyze digitized whole-slide histology images. DiaKwant™ is CE-IVDR Class C certified and available through Roche's navify Digital Pathology platform. DiaSurv™ is open for research collaboration. Supporting tools include the PD-L1 CPS Trainer educational platform and a demo/upload service for the DiaKwant™ product. All clinical products are for research use only unless CE-IVDR certified.

Differentiator

Problem solved

Functional benefit

Brands

• DiaKwant™: AI-based biomarker quantification tool providing automated scoring for PD-L1 (CPS/TPS), HER2, Ki-67, ER/PR, and mitosis detection across immunohistochemistry and H&E whole slide images. CE-IVDR certified.
• DiaSurv™
• PD-L1 CPS Trainer

Products and services

• DiaKwant™ AI-powered biomarker quantification platform that automatically scores PD-L1 (CPS/TPS), HER2, Ki-67, ER/PR expression and mitosis counts from immunohistochemistry and H&E-stained whole-slide images. Delivers cell-by-cell analysis with heatmap visualization for pathologist review. CE-IVDR Class C certified for clinical use and distributed through Roche navify® Digital Pathology platform.
• DiaSurv™ AI-driven prognostic assay that analyzes initial H&E histology slides to predict survival outcomes (5-year OS, RFS) and stratify patients into high- and low-risk categories. Validated across melanoma (c-index 0.78), colon cancer, breast cancer, and ovarian cancer (HRD/PARP inhibitor response) indications without requiring additional tissue or molecular testing. For research use.
• PD-L1 CPS Trainer Gamified web-based and mobile training platform that enables pathologists to practice PD-L1 Combined Positive Score (CPS) annotation on tumor tiles, with performance tracking, leaderboard ranking, and skill development. Hosted at pdl1-trainer.diadeep.com. Free to use.

Quantifiable outcome

• AI-assisted PD-L1 scoring improved accuracy from 75% to 88%, sensitivity from 78% to 96%
• +7 more outcomes

Named customers5 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 capability4 records

Feature9 records

Partnerships

Eight partnerships are on record, tiered core and minor.

• Aix-Marseille UniversitécoreStrategic or Co-development Partner · 1 February 2025Academic partner in France 2030 SmartProg consortium with DiaDeep and IHP Group for AI-based cancer prediction research. Jointly developing and validating melanoma prognosis algorithms.
• RochecoreChannel Partner/ Reseller/ Distributor · 9 September 2024International collaboration for commercialization of AI tools through Roche's navify® Digital Pathology platform. Roche markets DiaDeep's algorithms for breast cancer biomarkers (Ki67, ER, PR, HER2) to laboratories in US, Middle East, and Europe. Revenue sharing enables DiaDeep to fund R&D investments in cancer progression prediction.
• IHP GroupcoreStrategic or Co-development PartnerFrance's leading independent pathology laboratory serving as primary development partner and first user of DiaDeep's technology. Contributed to clinical validation studies and research publications. One of the founding pathologist-investor institutions.
• Gustave RoussycoreStrategic or Co-development PartnerOne of Europe's leading cancer centers, collaborating on international validation studies for melanoma prognosis AI. Contributed to clinical validation and research publications.
• CHUM (Centre Hospitalier Universitaire de Montréal)minorStrategic or Co-development PartnerCanadian university hospital contributing external validation cohort of 124 colon cancer patients for DiaSurv™ Colon algorithm validation.
• Foch HospitalminorStrategic or Co-development PartnerFrench hospital institution participating in clinical research collaborations with DiaDeep.
• CLB (Centre Léon Bérard)minorStrategic or Co-development PartnerLyon-based comprehensive cancer center participating in DiaDeep's research studies and clinical validation.
• CHU CaenminorStrategic or Co-development PartnerFrench university hospital contributing to clinical validation studies and research collaborations.

Scale indicators10 records

Recent moves5 records

Expansion highlights6 records

Direct peers

• Owkin: France-based AI pathology and precision-oncology company developing deep-learning models from histology and multimodal clinical data. Owkin directly competes with DiaDeep in AI-based biomarker quantification and prognostic prediction for cancer.
• PathAI: US-based AI pathology company offering deep-learning models for biomarker quantification and disease diagnosis from whole-slide images. Direct overlap with DiaKwant across oncology biomarkers and with DiaSurv-style prognostic applications.
• Paige: AI pathology company with FDA-authorized cancer-detection products and partnership with Roche for diagnostic deployment. Paige is the most directly comparable rival combining CE/FDA clearances with Roche channel access for digital pathology.
• Ibex Medical Analytics: Israel-based AI pathology company with CE-marked products for cancer detection and biomarker quantification across multiple tissue types. Directly comparable in deploying deep-learning biomarker scoring in clinical pathology workflows.
• Aignostics: Berlin-based AI pathology spinout from Charité developing foundation models for biomarker quantification and prognostic prediction from H&E and IHC slides. Comparable in European AI pathology focus, multi-biomarker assays, and pharma/biopharma partnerships.
• Proscia: Digital pathology platform company offering Concentriq, with AI-based diagnostic applications including biomarker quantification. Comparable as a distribution/operating platform for AI algorithms in pathology laboratories.

Emerging players

• Artera: AI precision-oncology company developing prognostic tests (initially for prostate cancer) from pathology images. Comparable as an emerging player in AI-derived prognostic assays with regulatory pathways overlapping DiaSurv's vision.

Broad incumbents

• Roche Diagnostics (Digital Pathology): Owner of the navify® Digital Pathology platform that distributes DiaDeep's algorithms; also markets its own image-analysis AI. Important context as both DiaDeep's go-to-market partner and a potential broader incumbent in the same workflow.
• Tempus: Large precision-medicine platform combining molecular profiling, imaging AI and clinical data. Overlaps with DiaDeep in AI-driven cancer decision support although its primary focus is molecular rather than histology-based.

Others

• Indica Labs: Provider of digital pathology image-analysis software (HALO) widely used by research pathology labs. Comparable as an adjacent infrastructure provider whose platform competes for the same biomarker-quantification workflows.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat5 records

Key risks6 records

Key highlights6 records

Customer concentration

Compliance1 record

Revenue estimate

Valuation estimate

Number of profiles

Profiles7 records

Funding overview

Funding rounds

Investors

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 DiaDeep