Numbers in. Clarity out.
Upload your results and get a clear, plain-language explanation of every marker. Interpretation, not a diagnosis.
Five things you only get with Mediora.
If potassium, troponin, INR or 12 other danger thresholds are crossed, you get an email the same day plus an in-app banner.
How alerts work Your name, blacked outPatient names, IDs and dates of birth are visually blacked out on the PDF image before our AI ever sees it (AWS Textract + custom EN/RU/HE label regex). Privacy is enforced by construction, not policy.
Read our security stance Ask the report directlyEvery report has a first-party AI chat sidebar grounded in YOUR markers — not a generic chatbot. Compare to last year, draft questions for your doctor.
See the chat in action Compare every uploadOne reading is a number. Pivot every marker across every upload side-by-side — HbA1c trend, kidney panel drift, ferritin recovery.
Open the matrix Native EN · RU · HEHebrew flips the entire UI to RTL, not just the text. Interpretations are written natively in each language — never machine-translated.
See examplesGet your Mediora Health Score.
A composite 0–100 score across the markers you upload — shareable card, no account required.
How Mediora.AI works
Upload
Drop a PDF, drag in a scan, or just snap a phone photo of the paper printout — we read all three. If you upload the same file twice we'll notice (SHA-256 dedup) and open the existing report instead of re-billing the AI.
Analyze
We extract every marker and explain what each one means — in plain language.
Report
Get a clear summary, download PDF, and see how your markers change over time.
One test tells a snapshot. Six tell a story.
Most AI tools interpret a single lab report. Mediora.AI tracks your markers over months and reveals trends — so you can catch change, not just values.
Your lab report. Your language.
English, Russian, Hebrew — native interpretation, not translation.
“Your potassium is high enough to matter — please contact a doctor today.”
«Калий повышен настолько, что это важно — свяжитесь с врачом сегодня.»
”רמת האשלגן גבוהה משמעותית — מומלץ לפנות לרופא עוד היום.”
Your health data, your control.
Built with privacy as the foundation, not an afterthought.
Patient names, IDs and dates of birth are visually blacked out on the PDF image before it reaches our AI model (AWS Textract + EN/RU/HE label regex). Verified live in prod.
Vision reads the structure of your lab page; a separate text model interprets the numbers. The interpretation prompt never sees your name, ID or DOB — smaller surface area, fewer hallucinations.
If potassium, troponin, INR, glucose or 11 other danger thresholds are crossed, you get an email the same day, a clear in-app banner — and a reminder 24h later if you haven't opened the report yet.
Every upload is checked against PDF/JPG/PNG file signatures, not just the extension. A renamed evil.exe → results.pdf is rejected before it can reach our AI pipeline.
If the date of birth in the uploaded PDF doesn't match your profile, we flag it and ask you to confirm — so a misclick uploading someone else's report is caught, not silently processed.
A one-time code by email, no password to steal.
TLS for every connection; patient identifiers (name, DOB, test date) wrapped with .NET DataProtection on disk.
Export every byte of your data (Art. 15 / 20) and delete it permanently — two-step email-confirmed, no dark patterns.
Every marker, condition and symptom page passes a 2-of-3 physician review with SLA enforcement before going live.
Every login, data export and deletion request is recorded in a SHA-256 hash chain visible from your profile — tampering breaks the chain and you'll see it.
We don't sell, license, or transfer your test results to anyone.
No Google Analytics, no Meta Pixel, no third-party ad networks. Privacy-respecting first-party analytics only.
Free during open beta
You already googled. Now what?
- 10 conflicting sources, often alarming
- Generic — not your numbers
- No history, no trends
- Sounds confident, may be wrong
- No persistent record of your panel
- Doesn't cite any verified source
- Doctor-reviewed catalog (138 entries)
- Your numbers, tracked over time
- Native EN · RU · HE, never translated
Verified by practising physicians.
Every catalog page passes through a 2-of-3 doctor review pipeline. Reviewer specialties are stamped into the Schema.org metadata so search engines and AI agents can verify provenance. Preview some of the roster:
Watch your report write itself.
Click "Re-explain" on any report and Mediora streams the explanation token-by-token in your language — same UX as ChatGPT, bounded to your actual markers. Rate-limited and cached so you're never billed twice for the same explanation.
- Live token-by-token output
- Works natively in EN · RU · HE
- 60-min cache — instant replay
- Layered cost defence (auth · cooldown · cache · concurrency)
Not just an explanation. A conversation.
Every report carries a first-party AI chat sidebar — RAG-grounded against the markers in YOUR panel, not a generic chatbot. Ask follow-up questions, compare against last year, get a draft list of things to mention to your doctor.
- Grounded in your real markers (Qdrant vector search over your uploads)
- Safety filter blocks diagnostic claims and dosing instructions
- Works in EN, RU, HE — same chat, native answers
Doctor-reviewed catalog you can actually read.
Every marker, condition, symptom and sick-fund page on Mediora has been written and verified. Browse without uploading anything.
Blood-test markers
70 entries — ferritin, HbA1c, TSH, vitamin D, ANA, lactate, and 64 more. Each with reference ranges, what high/low means, and when to ask a doctor.
Browse all 70 🩺Clinical conditions
37 entries — type-2 diabetes, hypothyroidism, Addison's, Crohn's, PCOS, endometriosis, and more. ICD-10 coded with key markers.
Browse all 37 🌡️Patient symptoms
31 entries — fatigue, palpitations, tremor, tinnitus, chronic cough, blurred vision. Each maps to the most relevant lab panel.
Browse all 31 🏥Israeli sick funds
Clalit, Maccabi, Meuhedet, Leumit — how each works, who they're best for, how to access lab results, and how Mediora helps you interpret them.
Compare all 4 ⚡Get your Health Score
A composite 0–100 score across the markers Mediora has interpreted. Shareable card, free.
Calculate mine 📈Chronic-disease trajectory
HbA1c rising, eGFR falling, ferritin depleting — Mediora's trend analyser flags multi-year drift so you can catch it before symptoms.
Learn how it worksMediora speaks MCP. Cite our catalog from any AI agent.
Model Context Protocol is the open standard agentic AI tools use to fetch verified knowledge. Mediora ships an MCP server with 14 tools — 9 anonymous catalog read tools and 5 patient-JWT-gated tools (whoami · analyze_lab_pdf · get_patient_history · get_test_details · longitudinal_trend). Patient tools use SSRF-hardened fetch + multipart S3 upload, so Claude Desktop can analyse a lab PDF without ever visiting our website.
Prefer a regular HTTP API? Every catalog entry is also at api.mediora.ai/api/public/catalog/{markers, conditions, symptoms, sick-funds} — 1-hour CDN cache, anonymous, JSON.
curl https://api.mediora.ai/api/public/catalog/markers{
"mcpServers": {
"mediora": {
"type": "http",
"url": "https://mcp.mediora.ai/mcp"
}
}
}Then ask Claude to compare hypothyroidism markers, look up a sick-fund, or interpret a lab value — answers come straight from Mediora's catalog.
Got Claude Desktop? Analyse a lab PDF without opening our site.
Drop a PDF (or a public-URL link) into Claude with Mediora's MCP server connected. Claude calls our mediora__analyze_lab_pdf tool, forwards your file to our upload pipeline with your JWT, and the analysis lands back in the chat — markers, ranges, and explanations. Same SSRF-hardened path our website uses; never your file on a public mirror.
Analyse this please
✓ HbA1c 5.9 % · borderline
When Claude needs a verified lab fact, it asks us.
Live example of a Claude Desktop conversation grounded in Mediora's catalog via MCP. The agent calls our tool, cites the source, and stays inside what's medically verified.
My TSH came back at 4.5 mIU/L. What does that mean?
mediora__lookup_marker(slug: "tsh")
→ returned reference range, high/low meaning, related markersTSH of 4.5 mIU/L is at the upper end of the typical reference range (0.4–4.0). Per Mediora's catalog, this is borderline subclinical hypothyroidism — worth retesting in 6–8 weeks alongside free T4 before treating.
Source: mediora.ai/en/markers/tshBuilt mobile-first. Reads on the train, replies on the train.
Just below prediabetes — retest in 3 months.
HbA1c trending +0.4 since January.
Pilot reviews are coming. Want to be first?
Mediora opened beta this month. We're inviting early reviewers — clinicians, patient advocates, and curious patients — to write candidly about what we got right and where we have room to grow. Your name + a single sentence on the homepage in exchange for honest feedback.
Built like infrastructure.
Twelve operational + SEO + reliability details that aren't worth their own section but collectively explain why Mediora behaves like real software, not a demo.
- Hangfire job auto-retry (3x) with temp-file retention. If OpenAI/S3 hiccups, your upload retries — you never re-upload manually.
- Stuck-test auto-recovery within 1 hour. DetectStuckTestsJob hourly: any analysis stuck Parsing/Analyzing > 30 min becomes Failed with a clear message — and free CleanupOldDataJob drops it after 7 days.
- Visible monthly quota on /profile. Used / limit / reset date — no surprises, no opaque counter.
- Cookie banner shown once. After you decide, we never bother you again. Default behavior, not a setting.
- Schema.org Physician on every doctor profile. Google / Bing / Perplexity can verify each doctor's specialty, language and years of practice from structured data — not just our marketing copy.
- Schema.org reviewedBy on every catalog page. The two physicians who approved this marker / condition / symptom are credited as structured data with their specialty — AI agents can verify provenance.
- Dynamic /sitemap.xml — 500+ URLs. Every marker, condition, symptom and sick-fund is its own indexable page in three languages, not a single generic SEO root.
- Branded PDF reports. Downloads carry a Mediora.AI footer with the generation timestamp — when you bring it to a doctor, they know exactly what they're reading.
Mediora launched in beta in June 2026. As mentions land, they'll appear here. Press inquiries: press@mediora.ai
Stop guessing what your results mean.
Frequently asked questions
Is my medical data secure? ⌄
We sign you in with one-time codes instead of passwords, secure every connection with TLS, and wrap patient identifiers (name, date of birth, lab test date) with .NET DataProtection on disk. Your results stay between you and Mediora — never shared with third parties — and you can delete your account and all your data at any time.
Who reviews the catalog content? ⌄
Every page on our catalog (markers, conditions, symptoms, sick funds) goes through a 2-of-3 doctor review pipeline before it appears publicly. Each draft is routed to three licensed physicians whose specialty and language matches; the article ships when two have approved or escalates to a senior reviewer if they disagree. Catalog updates carry the reviewers' specialties as Schema.org metadata.
Can AI agents (Claude, Cursor) use Mediora's catalog? ⌄
Yes. We ship an MCP (Model Context Protocol) server at mcp.mediora.ai with 14 tools: 9 anonymous catalog read tools (no API key required) and 4 patient-JWT-gated tools for personal data. Add to Claude Desktop in 4 lines of config — see the section above for the snippet. We also publish /llms.txt and /llms-full.txt for general-purpose AI crawlers.
How accurate is the AI interpretation? ⌄
Mediora.AI uses large language models, and our interpretations reference established clinical ranges. However, we always recommend confirming results with your physician before making any health decisions.
Does this replace my doctor? ⌄
No. Mediora.AI is a preparation and understanding tool — it helps you arrive at your doctor's appointment informed, not to bypass medical consultation. Think of it as a friend who reads the report with you before the visit.
What types of lab tests do you support? ⌄
Blood panels, biochemistry, thyroid function, hormones, vitamins, lipid profiles, and more. Upload as PDF or photo — we read scanned documents too. If a file is unreadable, we don't charge the analysis.
Do you work in my country? ⌄
Mediora.AI works globally. Our interpretations are in English, Russian, and Hebrew. We use international clinical reference ranges, with regional context where relevant.
Can I use this for my family members? ⌄
Currently one account = one person. Family accounts with multiple profiles are coming in the Pro plan.
How is this different from just Googling my results? ⌄
Google gives you scattered information, often alarming or wrong. Mediora.AI gives you a single clear interpretation tailored to your values, tracked over time, grounded in established clinical ranges.
How many analyses can I do during open beta? ⌄
During open beta, you can analyze up to 10 lab reports per month. The counter resets on the first day of each calendar month. Paid plans may be introduced later; current users will be notified in advance.