Digital Pathology AI
L

Lunit SCOPE

Lunit SCOPE is the pathology AI line of Lunit, the Seoul medical AI company. Drug companies, contract research organizations and digital pathology platforms use it to measure cancer biomarkers on tissue slides for research and trials. Lunit's radiology products have a separate record.

SCOPE IO reads a routine stained slide and classifies how immune cells are arranged around a tumor. In a 2022 study of 518 lung cancer patients in the Journal of Clinical Oncology, that pattern tracked response and survival on immunotherapy. The IHC suite scores PD-L1, HER2 and other stained markers, with cut offs set for a specific therapy. SCOPE GP predicts gene changes from a routine slide, to help pick which patients to send for molecular testing, and Lunit is codeveloping a lung cancer version with AstraZeneca.

Every SCOPE product page labels it for research use only, and Lunit states that SCOPE PD-L1 holds a CE mark. No price is published.

AI Health Index verifiedNovember 8, 2026
Compare Lunit SCOPE with other vendors
Founded
2013
Headquarters
Seoul, South Korea
Categories
pathology-ai, clinical-trials-ai
Indexed Products
Lunit SCOPE IO, Lunit SCOPE PD-L1, Lunit SCOPE HER2, Lunit SCOPE uIHC, Lunit SCOPE GP (Genotype Predictor)
Buyer Segments
Biopharma translational and companion diagnostic teams, Contract research organizations, Digital pathology platform providers, Pathologists (research use)
Assessment

Capability Axes

An AI Health Index grade measures what a buyer can verify from public sources on the date shown. It is not a rating of how good the product is. A vendor can build an excellent system and grade low on an axis because it publishes nothing an outsider can check. How grades read

AI Capability
AA on AI CentralityThe artificial intelligence is the product. Remove the model and there is nothing left to sell.
Vendor Published

The models are the product. Every Lunit SCOPE product is an AI analysis of a pathology slide: immune patterns from tumor infiltrating lymphocytes, scoring of immunohistochemistry stains, and gene mutation and protein expression predicted from H&E. The SCOPE line sells no scanner, stain or laboratory service, so without the models there is nothing to deliver.

BB on Autonomy and Oversight ModelThe oversight structure is described and one part is missing, commonly the threshold at which the system stops or what happens after it is wrong.
Vendor Published

Each output's role is described; what happens after an error is left open. SCOPE GP comes with models tuned either for high sensitivity or for high specificity, positioned for pre screening before molecular testing, so a confirming test follows the prediction. The IHC suite is described as helping pathologists, with custom scoring and cut offs for a therapy. All products are labeled research use only, not for diagnostic decisions.

The company does not say how a pathologist reviews SCOPE scores, or what Lunit does when a delivered score or prediction proves wrong.

BB on Model and Technology TransparencyThe approach or the suppliers are named without the version and update discipline behind them. Naming a supplier is the entry to this band both here and on Model Supply Chain Disclosure, which ask different questions of the same disclosure: who receives the data, and what produces the output.
Vendor Published

The approach behind the lead product is published, and the training data is sized, though versions and update practice are not. The 2022 Journal of Clinical Oncology study describes how SCOPE IO finds tumor infiltrating lymphocytes and outlines tumor and supporting tissue to classify the immune pattern. Lunit says SCOPE IO was trained on more than 50,000 whole slide images with over 10 million cell annotations, SCOPE PD-L1 on more than one million cell annotations by over 150 pathologists, and SCOPE uIHC on more than 18 IHC stains across over 20 primary tumor origins. SCOPE GP is built on Lunit's own foundation model and offers models tuned for sensitivity or specificity.

No model versions or release notes are published for the SCOPE line.

BB on Model Supply Chain DisclosureSubstantial partial disclosure, or a chain that is structurally short: an in house build, a cleared model that cannot be quietly swapped, or a deployment where the transfer does not occur at all. Naming the model provider exits the band below into this one; the axis rises from here on the completeness of the party list and on the terms that govern data once it arrives.
Vendor Published

Lunit builds the SCOPE models itself, on its own foundation model for SCOPE GP, so no outside model provider sits between a slide and a score. It names CellCarta and the Aperio AI Store, HALO AP and navify platforms as delivery routes.

The hosting provider, the parties that process slides on each delivery route, and any subprocessor list are not published.

BB on Clinical and Operational EvidenceNamed deployments with dated outcome figures and enough method to test them, or published research short of independent validation.
Peer Reviewed Publication

Peer reviewed research led by the company supports SCOPE IO and SCOPE PD-L1, short of independent validation of the products sold. A 2022 study in the Journal of Clinical Oncology (Park and colleagues, with Lunit authors, and Tony Mok of the Chinese University of Hong Kong as a joint senior author) classified 518 patients with non small cell lung cancer by the immune pattern SCOPE IO computes.

Patients with the inflamed pattern had higher response rates and longer progression free survival on checkpoint inhibitors, including those with PD-L1 tumor proportion scores of 1 to 49 percent. Progression free and overall survival differed across the three patterns. Lunit says SCOPE IO has been tested on more than 2,300 real world patient samples, with results in JCO and JITC. For SCOPE PD-L1 it cites a 2024 clinical validation in JCO Precision Oncology, a 2022 study of reader variation, and an observer study reporting a 32% cut in reading time. For SCOPE HER2 it cites a retrospective study in which the AI found 23.6% of slides scored HER2 negative to be HER2 ultralow.

The studies look back at past patients or measure reader agreement. None is an independent evaluation of a SCOPE product as sold.

CC on AI Safety and PHI StewardshipGeneral assurances of privacy and security that do not answer the questions artificial intelligence raises: what is retained, what reaches a model, and what happens to it there.
Vendor Published

Security assurances cover the company, with no answers for the pathology line. Lunit publishes an information security page with its ISO/IEC 27001 certification and a bug bounty program. Its radiology product pages say customer images are kept only for processing and not used for retraining.

Nothing published says whether the same holds for slides SCOPE analyzes, how long they are kept, or how training slides were sourced and deidentified. No safety engineering for the SCOPE models is published.

Regulatory and Compliance
DD on HIPAA and BAA PostureNo statement of status and no privacy document that reaches the product. Both halves are absent at this grade: a service level privacy document on its own lifts a vendor to the band above even where no status is stated. A website privacy notice, including one that says it does not cover customer data, does not reach the product, so a vendor whose only document is that notice and who states no status grades here.
Vendor Published

There is no stated business associate position, and Lunit's privacy notice covers its website rather than SCOPE. The notice, effective 6 November 2026, covers website visitors, marketing and events, job applicants and support contacts. It does not address slides or patient data analyzed by SCOPE products.

No business associate agreement or data processing agreement is published. The TrustShare trust center linked from Lunit's footer returns a page not found.

BB on Security Certifications and Trust CenterA recognized certification is named in the vendor own material without the artifact, or with a scope or renewal question the buyer has to raise. A recognized certification here is an audit of how the vendor operates security, such as SOC 2, HITRUST or ISO/IEC 27001. A certification has a scope and a clock, and both are part of this grade. A certification the vendor attaches to its data center, hosting facility or cloud provider, rather than presenting as its own, is the host's and does not count here.
Vendor Published

Lunit, the parent company, names a recognized certification without publishing it. Its information security page states that Lunit.io is ISO 27001 certified, and its manufacturing site holds ISO 13485:2016, MDSAP and Korea GMP certification. Lunit's footer links a bug bounty program and a TrustShare trust center, and the trust center link returns a page not found.

The certificate, its scope and its date are not shown, and nothing says whether the scope covers SCOPE analysis services.

BB on FDA and Regulatory StatusThe pathway is stated and in progress, or a clearance is named without the vintage and scope a buyer needs to match it to the product on offer. A product outside device regulation also sits here when the vendor publishes what it is for and in what setting and the product on offer matches, or names another regime it works under, such as a registration outside the United States. A general clause saying the service is not medical advice does not reach this band.
Vendor Published

Lunit names a CE mark for SCOPE PD-L1 without its date or scope, and labels every SCOPE product page for research use. Its information security and registration page says Lunit SCOPE PD-L1 has received CE mark, with no certificate, class, regulation or intended use shown.

The product pages, SCOPE PD-L1 included, say the products are for research use only and not for diagnostic procedures or decisions. Lunit's manufacturing site holds ISO 13485:2016, MDSAP and Korea GMP certification. The FDA's 510(k) database lists no SCOPE device, and Lunit's clearances there are for its radiology products.

CC on AI Governance and Bias DisclosureResponsible artificial intelligence is committed to in policy language with no evaluation behind it. Most of the index sits here.
Vendor Published

Validation across datasets is asserted, but the SCOPE pages compare no results across groups. Lunit says its SCOPE models are validated across diverse internal and external datasets, lists publications, and has a certified quality system.

The product pages set out no results by scanner, stain, laboratory, tumor type or patient group. No evaluation or monitoring process specific to the pathology models is described.

CC on AI Liability and RecourseMechanisms exist that let someone challenge an output, such as audit trails, source traceability or review before commit, with nothing standing behind the output and no route for the harmed party.
Vendor Published

A confirming test is built into the intended use, but nothing stands behind the output. SCOPE GP is positioned as pre screening before molecular testing, so a confirming test can catch a wrong prediction, and the products are labeled research use only.

No error rate at a stated operating point, warranty, remedy or correction route for a SCOPE analysis is published.

Integration and Deployment
BB on EHR and Interoperability DepthNamed systems with read access or one directional writing, or standards support with named deployments behind it.
Vendor Published

No clinical record integration exists, and research platform integration is named for one product. Lunit SCOPE PD-L1 lists integration into Leica Biosystems' Aperio AI Store, Indica Labs' HALO AP and Roche navify. Lunit says the SCOPE line works with a broad range of scanners and file types and deploys through digital pathology platforms and contract research partners, naming CellCarta.

No EHR, HL7 or FHIR interface or LIS connection is documented, and how deep the integration runs on each named platform is not described.

CC on Deployment Model and Data ResidencyA single hosted option with location implied rather than committed.
Vendor Published

SCOPE reaches buyers through partners, and where it runs is implied, not stated. Lunit describes SCOPE as deployable across digital pathology platforms and contract research partners, lists SCOPE PD-L1 on Aperio AI Store, HALO AP and Roche navify, and runs analyses for drug company programs.

Nothing published says where slides are processed or stored, which cloud is used, whether processing happens in Korea, or whether SCOPE can run on premises.

Commercial
CC on Commercial TransparencyNo price is published and the posture is discoverable: a buyer can establish how the product is sold and what drives the cost before contacting the vendor. Most of the index sits here.
Vendor Published

No price is published, but how the products are sold can be worked out. Lunit sells the SCOPE line to drug companies, contract research organizations and digital pathology platform providers. It sells through codevelopment, such as the AstraZeneca lung cancer pre screening tool, and through partners such as CellCarta, with custom scoring and cut offs built for specific therapies.

No license basis, per slide or per study price is published.

BB on Setting and Specialty CoverageCoverage is named with validation behind part of it.
Vendor Published

Each product's coverage is named, with peer reviewed validation behind part of it. SCOPE IO classifies the immune pattern on H&E slides and is validated in non small cell lung cancer, in the 2022 study of 518 patients. The IHC suite covers PD-L1, HER2 and a universal analyzer for other immunohistochemistry markers. SCOPE GP predicts lung cancer driver alterations from H&E. All are research use only.

Lunit sells the line to drug companies and contract research organizations for biomarker work tied to specific therapies and for trial pre screening.

Tracked Since Listing

Product updates

Material product, regulatory, evidence and commercial changes at Lunit SCOPE, each verified against a live source and tagged to the capability axis it bears on. Funding rounds and awards are not product changes and are not logged.

Nov 4, 2026Clinical evidencePartially verified

Lunit presented two SITC 2026 posters using Lunit SCOPE IO. With Labcorp, SCOPE IO quantified 22 immune and morphological features on 339 ovarian carcinoma slides and found about 1.5 fold higher mitotic cell density in HRD tumors (p=0.002), with inflamed phenotypes showing higher TMB (median 4.7 vs 3.9 mut/Mb, p=0.038). With CellCarta, SCOPE IO assessed tertiary lymphoid structures in ABACUS trial bladder cancer tissue alongside spatial analysis of response to neoadjuvant atezolizumab.

Bears on: Clinical and Operational EvidenceSource
Oct 20, 2026Clinical evidence

Lunit presented five ESMO 2026 posters using Lunit SCOPE. In 993 resected lung cancers, SCOPE PD-L1 agreed with the pathologist's tumor proportion score 73 percent of the time, and 204 of the 211 disagreements fell in the 1 to 10 percent band. In 181 pancreatic cancer patients, SCOPE IO linked a higher tumor to stroma ratio with worse progression free survival, except in KRAS G12D tumors, where the link reversed.

Bears on: Clinical and Operational EvidenceSource
Our read on these updates →Tracked since Oct 2026
Alternatives

Alternatives to Lunit SCOPE

Other products that do the same kind of work as Lunit SCOPE, and what each one does.

All Lunit SCOPE alternatives, with prices →

Biomy makes DeepPathFinder, cloud software that uses AI to map immune and cancer cells on tissue slides for drug research.

Imagene AI reads routine stained lung cancer slides and predicts which gene changes a tumor carries, without using more tissue for sequencing.

PictorLabs, which now styles its name Pictor Labs, makes software that stains tissue images with AI instead of chemicals, and a tissue intelligence platform built on it.

Aignostics builds foundation models for computational pathology and sells them, and applications built on them, to biopharmaceutical companies for target discovery, translational research, clinical trials and companion diagnostic development, and to pathology laboratories under research use terms.

Panakeia infers molecular status from routine H&E slide images, without sequencing, immunohistochemistry or additional tissue.

Mindpeak sells AI that scores cancer biomarkers on pathology slides, which laboratories use to quantify HER2, ER, PR, Ki-67 and PD-L1 on immunohistochemistry and H&E images.

Commercial

Pricing

Vendor-published figures are labeled as such. Figures labeled “Estimated” are derived from third-party sources and have not been confirmed by the vendor.

Entry Price Pricing Basis BAA Tier Implementation Source
Not published
Research use AI analysis for biopharma, contract research organizations and pathology platforms, through direct programs, codevelopment and partners; no unit of charge or price published. Not published. Lunit's privacy notice covers its websites and contacts; no business associate or data processing agreement is published. Not published. Vendor Published

Lunit sells the SCOPE line to biopharma and contract research organizations, including through codevelopment such as the AstraZeneca lung cancer pre screening tool and the CellCarta alliance, with custom scoring and cut offs per therapy, and publishes no price, unit of charge or license term.

Lunit SCOPE pricing in full →