Symptoma · Project

AI Search Engine That Helps Doctors Diagnose Rare Diseases Faster

healthMarket-readyTRL 8

Imagine Googling your symptoms but instead of scary forum posts, you get a doctor-grade list of possible causes ranked by probability. That's what Symptoma built — a search engine covering over 44,000 diseases that outperforms every competitor in diagnostic accuracy. They added a smart chatbot that asks follow-up questions like a real doctor would, plus translated the whole system into Spanish, Japanese, and Arabic so it works worldwide.

By the numbers

€446 billion

Wasted costs from misdiagnoses in the EU

$750 billion

Wasted costs from misdiagnoses in the US

44,000

Disease conditions covered in Symptoma database

10-40%

Rate of misdiagnoses in healthcare

10%

Patient deaths caused by diagnostic errors

22 of 30 million

Rare disease patients in EU without a diagnosis

8 million

Patients who waited 10 years on average for rare disease diagnosis

30%

Healthcare budget that could be freed by solving misdiagnosis

Doctors satisfied with existing diagnostic research tools

EUR 1,597,400

EU contribution to this project

The business problem

What needed solving

Doctors face over 20,000 diseases but misdiagnose patients 10-40% of the time, causing 10% of patient deaths and €446 billion in wasted healthcare costs across the EU. Only 3% of doctors are satisfied with the diagnostic tools available to them. Rare disease patients suffer the most — 22 of 30 million in the EU still lack a diagnosis, with average wait times of 10 years.

The solution

What was built

Symptoma built an AI-powered diagnostic search engine covering 44,000 diseases, a deep-learning chatbot that guides users through intelligent question sequences to reach accurate diagnoses (D2.2), and multilingual versions in Spanish, Japanese, and Arabic with medically reviewed translations (D3.1).

Audience

Who needs this

Health insurance companies looking to reduce misdiagnosis-related claim costsPharmaceutical companies needing faster rare disease patient identification for clinical trialsHospital networks and telemedicine platforms seeking AI diagnostic decision supportNational health services aiming to reduce diagnostic wait times and wasted spendingMedical device companies wanting to integrate AI diagnostics into their products

Business applications

Who can put this to work

Health Insurance

enterprise

Target: Health insurance companies and managed care organizations

If you are a health insurer dealing with misdiagnosis costs — this project developed an AI diagnostic engine that reduces diagnostic errors, which currently account for €446 billion in wasted costs in the EU alone. Faster, more accurate initial diagnoses mean fewer unnecessary tests, fewer wrong treatments, and lower claim payouts. The system covers 44,000 conditions and has been validated to outperform all competitors in accuracy.

Pharmaceutical & Rare Diseases

enterprise

Target: Pharmaceutical companies focused on rare and orphan diseases

If you are a pharma company working on rare disease therapies struggling to find correctly diagnosed patient populations — this project built a tool that helps identify rare diseases faster. Currently 22 of 30 million rare disease patients in the EU remain undiagnosed, and 8 million waited 10 years on average for diagnosis. Symptoma's AI engine can accelerate patient identification, directly supporting clinical trial recruitment and therapy matching.

Digital Health & Telemedicine

any

Target: Telemedicine platforms and hospital IT departments

If you are a telemedicine provider looking to improve remote consultation quality — this project developed a deep-learning chatbot that guides patients through intelligent question sequences to reach accurate diagnoses. It works across multiple languages including Spanish, Japanese, and Arabic, making it suitable for international patient populations. With only 3% of doctors satisfied with existing research tools, integration could be a competitive differentiator.

Frequently asked

Quick answers

What does Symptoma cost and what is the pricing model?

Symptoma.com is free for basic use. The project states they are already operating at a profit with pilot customers, indicating a B2B licensing or subscription model for professional/enterprise users. Specific pricing tiers are not disclosed in the project data.

Can this scale to serve a national health system or large hospital network?

Yes. The platform already covers 44,000 conditions and has been expanded to multiple languages (Spanish, Japanese, Arabic) alongside its original offering. The deep-learning chatbot component was designed to work across all 44,000 conditions, demonstrating system-wide scalability. The SME-2 funding was specifically for scaling the business.

What is the IP situation — can we license or white-label this technology?

Symptoma GmbH is a private Austrian company and sole owner of the technology. As a single-partner SME instrument project, all IP stays with the company. Licensing or white-label arrangements would need to be negotiated directly with Symptoma GmbH.

How does diagnostic accuracy compare to existing tools?

The project states that Symptoma's diagnostic accuracy has outdone all competitors, making it the first and only viable solution in this field. They validated this through case report testing. However, specific accuracy percentages are not provided in the available project data.

Is this compliant with medical device regulations?

Based on available project data, specific regulatory certifications (CE marking, FDA clearance) are not mentioned. As an AI-based diagnostic support tool operating in the EU, it would fall under the Medical Device Regulation (MDR) and the AI Act. Prospective users should verify current regulatory status directly with Symptoma GmbH.

How long does integration take?

The project delivered a web-based search engine and chatbot, suggesting API-based integration is feasible. The multilingual expansion (D3.1) shows the platform architecture supports modular additions. Based on available project data, specific integration timelines are not disclosed.

What languages are supported?

The platform supports at minimum English, German, Spanish, Japanese, and Arabic. The translation pipeline combines machine translation from Google, Bing, Yandex, and DeepL with review by native-speaking medical professionals, ensuring clinical accuracy across languages.

Consortium

Who built it

This is a single-company project — Symptoma GmbH from Austria, a private SME that received the full EUR 1,597,400 through the competitive SME Instrument Phase 2 program. The absence of university or research partners is actually a strong commercial signal: this is not a research experiment but a business scaling operation. The company was already profitable with paying customers and used the funding to add AI capabilities and expand into new language markets. For potential business partners, this means you're dealing with one decision-maker, no consortium politics, and a company with proven commercial instincts.

SYMPTOMA GMBHCoordinator · AT

How to reach the team

Symptoma GmbH is based in Austria. Contact their business development team through symptoma.com.

Order a report on this consortium See SYMPTOMA GMBH profile →

Next steps

Talk to the team behind this work.

Want an introduction to the Symptoma team or a detailed briefing on how this technology fits your diagnostic workflow? Contact SciTransfer for a guided matchmaking session.

Order a report on this topic More in Health & Biomedical