Medical accuracy
96.7 %

*Evaluated based on 1,471 case studies.

Medical Quality

Leading-edge medical accuracy and reliability.

We verify the medical accuracy of our technology in medical stress tests. This happens continuously and on the basis of thousands of case studies from the literature. The goal is to identify the verified diagnosis from the case study as a possible outcome. This has proven to be the most reliable method of testing medical accuracy under objective conditions.

Benchmark with case studies from the standard literature.

In order to objectively assess the quality of XUND in comparison to other solutions, we assessed the 45 case studies from Semigran et al. 2015, one of the most frequently cited publications in the domain. These case studies are regularly used by researchers to evaluate medical accuracy, most recently by Ceney et al. 2021. The results showed that 93.3% of the cases were identified correctly.

ICD-10
Medical condition
Age
Result
ICD-10J00
Medical conditionCommon cold
Age30 Years
Result identified
30-year-old male with common cold
ICD-10: J00
Present symptoms:
  • Runny nose

  • Sore throat

  • Increased sweating

  • Headache

  • Cough

  • General muscle pain

  • Limb pain

  • Red throat

  • Swelling of the lymph nodes

Additional information:
Do you smoke or have you been a smoker in the past?
Yes
Source:
Intermittent low back pain referred from a uterine adenomyosis: a case report. J Chiropr Med. 2011;10(1):64-69. https://dx.doi.org/10.1016%2Fj.jcm.2010.08.004
Verified diagnosis identified as 3. result
Conditions identified by XUND:
  • Acute bronchitis
  • Pharynx inflammation
  • Common cold
ICD-10B27.0
Medical conditionPfeiffer's disease
Age16 Years
Result identified
16-year-old female with Pfeiffer's disease
ICD-10: B27.0
Present symptoms:
  • Fever
  • Sore throat
  • Physical exhaustion
  • Swallowing difficulties
  • General malaise
  • Swelling of the lymph nodes
  • Red throat
  • Rash
Additional information:
Do you feel like the rash is spreading to your whole body?
Yes
Source:
Evaluation of symptom checkers for self diagnosis and triage: audit study Hannah L Semigran, Jeffrey A Linder, Courtney Gidengil, Ateev Mehrotra 2015
Verified diagnosis identified as 2. result
Conditions identified by XUND:
  • Scarlet fever
  • Pfeiffer's disease
  • COVID-19 infection
ICD-10J02
Medical conditionPharynx inflammation
Age7 Years
Result identified
7-year-old female with pharynx inflammation
ICD-10: J02
Present symptoms:
  • Fever
  • Sore throat
  • Nausea
  • Vomiting
  • Swelling of the lymph nodes
  • Red throat
Source:
Evaluation of symptom checkers for self diagnosis and triage: audit study Hannah L Semigran, Jeffrey A Linder, Courtney Gidengil, Ateev Mehrotra 2015
Verified diagnosis identified as 2. result
Conditions identified by XUND:
  • Tonsil inflammation
  • Pharynx inflammation
  • COVID-19 infection
ICD-10H66
Medical conditionInflammation of the middle ear
Age2 Years
Result identified
2-year-old male with inflammation of the middle ear
ICD-10: H66
Present symptoms:
  • Sniffles
  • Cough
  • Sleeping difficulties
  • Loss of appetite
  • Fever
Source:
Evaluation of symptom checkers for self diagnosis and triage: audit study Hannah L Semigran, Jeffrey A Linder, Courtney Gidengil, Ateev Mehrotra 2015
Verified diagnosis identified as 2. result
Conditions identified by XUND:
  • Acute bronchitis
  • Otitis media of the infant or toddler
  • Common cold
  • COVID-19 infection
ICD-10D59.3
Medical conditionHemolytic uremic syndrome
Age4 Years
Result not identified
4-year-old male with hemolytic uremic syndrome
ICD-10: D59.3
Present symptoms:
  • Sniffles
  • Cough 
  • Sleeping difficulties
  • Loss of appetite
  • Fever
  • Irritability
  • Runny nose
Source:
Evaluation of symptom checkers for self diagnosis and triage: audit study Hannah L Semigran, Jeffrey A Linder, Courtney Gidengil, Ateev Mehrotra 2015
Verified diagnosis could not be identified
Conditions identified by XUND:
  • Rotavirus
  • Norovirus

The knowledge of millions built into one health solution.

Step 1
Data analysis with AI

Our artificial intelligence can analyze millions of medical publications from the literature and process data on over 4,000 condition concepts.

Step 2
Review by doctors

Our team of doctors then verifies this information qualitatively and enriches it with reference literature and practical experience.

Step 3
Medical stress test

We verify our technology with thousands of real life case studies. This is the most objective way to test accuracy of our system.

Step 4
Usability tests

Through ongoing testing with real users, we ensure that our medical algorithms work as intended and meet the defined requirements.

Step 5
Release new version

We are continuously updating XUND in line with medical device regulations and to improve accuracy and coverage.

Want to know more? Here are some of the most frequently asked questions for you.

Do you work together with medical doctors?

Yes, the medical team at XUND plays a pivotal role in managing the medical database, which forms the core of our AI solution. Our medical doctors verify the quality of every single piece of information in the medical database, ensuring its accuracy and reliability. This thorough verification process relies on reference literature and real-world medical practice experience. Learn more about the work of our medical team on our Medical Quality page, or read the interview with our Head of Medical.

How do you ensure the quality of the medical content?

Our in-house team of medical doctors and editors meticulously reviews and validates XUND's medical content before release, ensuring it meets ISO 13485 & MDR standards, and guaranteeing its highest quality. Additionally, the team collaborates with the Thieme Group to create fact sheets for frequently diagnosed conditions. On top of that, we continuously verify the medical accuracy of our technology through stress tests, drawing from thousands of literature case studies. For more details on our quality assurance process, please visit our Medical Quality page.

Is all content created or reviewed by medical doctors?

Yes, XUND's content is a blend of quantitative AI analysis and qualitative medical expertise. Using AI, we have analyzed millions of medical publications, which were then reviewed by our medical doctors with both literature and practical experience. This synergy, combined with our standardized quality assurance, ensures the highest level of accuracy and reliability in our content. To gain more insight into our medical quality processes, please see our Medical Quality page.

What diseases do you cover?

Our AI has the capability to automatically analyze millions of medical publications and process data on over 4,000 medical conditions. XUND prioritizes the most common medical conditions from this dataset, totaling over 500 illnesses, to provide statistically relevant and meaningful results.

Do you do clinical safety evaluations?

We continuously test our medical knowledge base and algorithms against cases reported in journals and training materials for doctors. We deem this the most reliable and scalable way to test the accuracy of the Medical API under verified, real-life circumstances. This is also the most commonly used approach to evaluate state-of-the-art devices by researchers and thus provides the best evidence base for a clinical evaluation of XUND with performance data.

Step into the future of healthcare.