MedulAI
MedulAI is an AI model from Spanish company SpotLab that supports cell counting and cytological analysis of bone marrow aspirates, classifying 23 cellular subtypes to help haematologists cut analysis time and reduce variability between observers.
What is MedulAI?
MedulAI is an artificial intelligence model built by SpotLab to support cell counting and analysis in bone marrow samples. It classifies cells across 23 cellular subtypes. The product page states a training dataset of more than 500,000 cells from over 1,000 patients, while the launch communications of June 2026 describe a corpus of more than six million cells and cite 18 scientific publications behind the system.
The clinical problem it addresses is well defined. A bone marrow aspirate is a routine diagnostic test for blood cancers, traditionally read under an optical microscope by a haematologist. That reading is slow and varies between observers. MedulAI automates the count and assists classification, which the vendor says standardises the process while reducing time and human variability. SpotLab also argues it gives a more comprehensive view of a patient's disease, which can feed predictive models and inform targeted therapy decisions.
MedulAI launched officially on 25 June 2026 and is presented as the first Spanish AI system for the digitisation and cytological analysis of bone marrow. It is described as in use across 14 hospitals in seven autonomous communities in Spain, among them Ramon y Cajal, La Paz, Fundacion Jimenez Diaz, Virgen del Rocio, Reina Sofia, Parc Tauli, Hospital del Mar and the Institut Catala d'Oncologia. The Spanish Society of Haematology and Haemotherapy and GSK, through the Gate2Health programme, are named as partners helping introduce it into the Spanish National Health System. SpotLab claims AI can cut haematology sample analysis time by up to 80 percent. In July 2026 MedulAI received an award for advances in haematological diagnosis with AI at La Razon's A Tu Salud awards.
The model runs on SpotLab's TeleSpot platform, a cloud architecture connected to web and smartphone applications, and can be deployed on the edge, in the cloud or through an API. SpotLab itself is a spin-off of the Universidad Politecnica de Madrid with a team of more than 20 people. The product is sold through direct engagement: there is no public price and no self-service sign-up.
What it does
- Digitise bone marrow aspirate slides
- Count cells automatically in bone marrow samples
- Classify cells across 23 cellular subtypes
- Reduce variability between observers on the same sample
- Standardise morphological analysis across hospital sites
- Cut the time haematologists spend on sample analysis
- Feed richer cell-level data into predictive models
When to use MedulAI / When not to
A quick filter to help you decide if MedulAI is the right fit.
When to use MedulAI
- Haematologists who read bone marrow aspirates and want automated differential cell counts
- Hospital haematology laboratories seeking to standardise morphological analysis across sites
- Smaller or remote centres without on-site cytology expertise, which the vendor explicitly targets
- Clinical research teams working on blood cancers, myelodysplastic syndromes or leukaemia
- Clinical informatics and digital pathology leads rolling out AI in a hospital network
When not to use MedulAI
- Patients or general consumers: this is a professional diagnostic aid, not a self-service health app
- Anyone expecting an autonomous diagnosis, since the vendor positions it as support for clinical judgement
- Teams looking for parasite or infectious disease screening, which is ParasitAI, a separate SpotLab product
- Buyers who need a published price, contract terms or a self-service sign-up before engaging
- Organisations that require a signed DPA, a subprocessor list or a stated hosting jurisdiction up front
How to use MedulAI
A typical end-to-end flow, from setup to results.
- Identify the clinical need, since MedulAI is scoped to bone marrow aspirate cytology
- Contact SpotLab to request a demonstration of the product, the route the privacy policy describes
- Work through SpotLab's assessment phase, in which it designs the technological and clinical roadmaps
- Agree the deployment mode: edge AI on mobile devices, cloud through the web platform, or API integration
- Connect the digitisation hardware, as the platform states it integrates with multiple hardware devices
- Have the platform configured to your technical requirements and organisation
- Digitise bone marrow aspirate slides through the platform
- Review the automated cell counts and the classification across the 23 subtypes
- Use built-in user, data and AI traceability to audit how results were produced
- Retain haematologist review, since the vendor positions the model as decision support
Pros & Cons
Pros
- An unusually strong evidence base: 18 claimed publications, with two peer-reviewed references cited on the product page itself
- Verifiable real-world deployment, with 14 named hospitals across seven Spanish autonomous communities
- Named institutional partners, the Spanish Society of Haematology and Haemotherapy and GSK through Gate2Health
- Independent recognition, with an award for haematological diagnosis with AI in July 2026
- Three deployment options, including an edge mode that runs without connectivity or data transfer
- An EU-based vendor with a Madrid registered office and a named Data Protection Officer
- A claimed reduction of up to 80 percent in haematology sample analysis time
Cons
- No public pricing of any kind: no pricing page exists and none of the 55 URLs in the sitemap is one
- No terms and conditions and no legal notice, so the contractual framework is entirely private
- No data processing agreement and no subprocessor list, despite the product handling health data
- The privacy policy covers website visitors only and says nothing about patient data
- No hosting country or region is stated for clinical data
- No documented way to exclude data from model training
- Regulatory status is unclear: ISO-13485 and IVDR appear as standards followed, not as certificates held
Pricing & Plans
No pricing information is published. SpotLab does not operate a pricing page, and none of the 55 URLs listed in the site's sitemap corresponds to one; direct probing of the usual paths returned the site's not-found page in every case. No free plan and no free trial are mentioned anywhere on the site. Access is arranged through direct contact with the vendor, with the privacy policy referring to a demo request route. Prospective customers should expect a quotation negotiated per institution rather than a published rate.
Data, GDPR & hosting
A consolidated view of how MedulAI handles your data.
GDPR overview
SpotLab claims GDPR compliance explicitly, with a GDPR compliant item on its Platform page. Concrete measures are published: a privacy policy and a cookie policy both dated 23 December 2025, a named Data Protection Officer reachable at dpo@spotlab.org, consent as the legal basis, and the full list of data subject rights (access, rectification, erasure, restriction, objection, portability and not being subject to automated individual decisions), exercisable by post or email with proof of identity. The Spanish supervisory authority, the Agencia Espanola de Proteccion de Datos, is named for complaints. However, these documents address website visitors only. There is no published record of processing for patient data, no legal basis for health data, no data protection impact assessment and no processor agreement.
Who owns the data?
SpotLab publishes no terms and conditions, so data ownership for MedulAI is not contractually documented anywhere on the site. The only published document is a privacy policy dated 23 December 2025, and it covers website visitors alone: name, surname, email address and cookie data, with SPOTLAB, S.L. of Paseo de Juan XXIII 36B, 28040 Madrid named as controller. Nothing states who owns the bone marrow images, the derived cell counts or the patient records processed by the product, nor whether the hospital or the vendor controls them. For a tool deployed in 14 hospitals, that silence is a material gap a buyer must close directly with SpotLab.
Reuse rights
There are no terms and conditions on the site, so no reuse rights are granted or withheld in writing. The privacy policy limits itself to website data and names three purposes: requesting a demo of the products and services, subscribing to the newsletter, and contacting SpotLab. Consent is the stated legal basis. Separately, the Platform page says SpotLab works with multimodal data that can be imported from clients databases and used to train AI models, without saying whether such data feeds only the customer's own model or SpotLab's wider models. Customers should therefore treat reuse rights over clinical data as undefined until contracted.
Data retention & training
Hosting summary
No hosting jurisdiction is published for MedulAI. SpotLab describes TeleSpot as a cloud architecture connected to web and smartphone applications, and offers an edge mode in which the AI runs in real time on mobile devices on site, without connectivity or data transfer; that mode is the only configuration for which a data location can be inferred, and only because data does not leave the device. Neither a country nor a region is named for the cloud deployment. The cookie policy does disclose one transfer: analytics cookie data goes to the United States, with an explicit warning that the level of protection there is not equivalent to the European Economic Area. That statement concerns website analytics only, not clinical data. The website itself is served from Automattic infrastructure, which says nothing about where the product processes patient information. Buyers should obtain hosting details contractually.
Things to keep in mind
Risks and trade-offs to weigh before adopting MedulAI.
- Regulatory status is not established on the site: ISO-13485 and IVDR are cited as standards followed, an in-vitro medical device manufacturing licence is claimed without number, body or validity date, and in November 2021 CE marking was described as an objective being pursued with Hardian Health
- The published privacy and cookie policies cover website visitors only; no document describes how patient data is processed, under which legal basis, or whether an impact assessment exists
- No subprocessor list is published for the product, and the only third party named anywhere on the site is Google Analytics, for website cookies
- No hosting country or region is stated for clinical data, so data residency cannot be assessed from public sources
- Automation bias is a real clinical risk: an automated differential count is persuasive, and teams may under-scrutinise borderline morphology that still requires expert review
- Training figures diverge between the vendor's own pages, with over 500,000 cells and 1,000 patients on the product page against more than six million cells in the 2026 announcements
- The absence of published terms means service levels, liability and data ownership are defined only in a private contract
Setup & Integrations
Technical difficulty
Deployment is a project, not a sign-up. There is no self-service registration; SpotLab begins with an assessment phase in which it designs the technological and clinical roadmaps, then adapts the platform to the customer's technical requirements and organisation. The institution must choose a deployment mode among edge, cloud and API, and connect slide digitisation hardware. Expect clinical validation, information security review and integration effort on the hospital side. No public technical documentation exists, so scoping depends entirely on the vendor.
Deployment
Behind MedulAI
Social
Resources
All the official URLs gathered for verification and reference.
Frequently asked questions
What exactly does MedulAI do?
Who develops MedulAI?
Is MedulAI actually deployed in hospitals?
Does MedulAI replace the haematologist?
How much does MedulAI cost?
Is there an API?
Is MedulAI a certified medical device?
Where is the data hosted?
Is customer data used to train the models?
How do I get in touch?
Should you pick MedulAI?
MedulAI is a credible, narrowly scoped clinical tool rather than a general-purpose AI product. The evidence behind it is unusually concrete for this market: peer-reviewed references cited on the product page itself, 18 claimed publications, a named list of 14 deploying hospitals across seven Spanish autonomous communities, institutional partners in the Spanish Society of Haematology and Haemotherapy and GSK, and a sector award in July 2026. For a haematology laboratory looking to standardise differential counts on bone marrow aspirates and reduce inter-observer variability, it is a serious candidate.
The reservations are documentary rather than clinical. SpotLab publishes no pricing, no terms and conditions and no legal notice, so the entire commercial and contractual relationship is private. More importantly for a product processing health data, the published privacy and cookie policies cover website visitors only: there is no processor agreement, no subprocessor list, no stated hosting jurisdiction, no retention rule for clinical data and no published impact assessment. Regulatory status also needs clarifying, since ISO-13485 and IVDR are presented as standards followed rather than certificates held, and CE marking was described in 2021 as an objective being pursued.
None of this makes the product doubtful, but it does shift the burden onto the buyer. Any institution evaluating MedulAI should ask SpotLab directly for its regulatory documentation, its data processing agreement, its hosting arrangements and its position on training with customer data, and should expect those answers in contract rather than on the website.
- Choosing a selection results in a full page refresh.
- Opens in a new window.