History taking
The patient answers in character and tells you only what you ask. Find the right questions before the room does it for you.
MedExamHelperSign inMedExamHelper
Step into a timed station. Take the history. Examine.
Order the tests. Defend your thinking to the examiner.
Then get marked against the checklist. Every mark quotes your own words back at you.
An AI council of three judges does the grading. Clinicians calibrate the rubric and checklist. You practise anytime, anywhere, as many times as you need.
Free to start · No card · No sign-up to try
250+ stations · Timed study & exam modes · Voice and dictation · Every mark cites your transcript · Clinician-calibrated rubrics
This is a real station.
58M, central chest pain on exertion × 3 weeks. Relieved by rest. No radiation, no SOB. Pain character: dull, pressure-like. CV risk factors to ask: smoking, DM, family history. Differential to revisit: stable angina, MSK, GORD.
Your score: 39/50
Evidence: “Can you tell me more about the pain?” — you asked. “Do you smoke?” — you didn't.
Not a vibe. The checklist, the domains, and your own transcript — side by side.
Tells you
Read → highlight → forget
Tests recognition
Read → choose → explanation
Real talk, no marking
Their schedule, their patience, their guesswork
Watches you perform
Ask → examine → reason → decide → get marked
And the weekend course? One day, one room, one shot at feedback. This is every night until the exam.
Your examiner won't hand you four options. They'll watch you find the answer in a room, with a patient, on a clock.
The patient answers in character and tells you only what you ask. Find the right questions before the room does it for you.
Say what you'd examine and how. Get the findings the station was built to surface.
Build your differential as you go. The viva examiner will ask why.
Explain, counsel, reassure — out loud, to a patient, not in your head.
Study mode teaches. Exam mode grades. Same station, real pressure.
Graded by a council of three judges against the checklist. What you did, what you missed, and the exact moment it happened.
Every mark you miss becomes a card in your deck, scheduled by spaced repetition. You never lose the same mark twice.
A readiness score built from mastery, recency and streak, plus coverage per specialty — so the exam never finds a gap first.
| Marking domain | Result |
|---|---|
| Introduction & consent | 4/4 |
| History | 8/12 |
| Examination | 6/8 |
| Clinical reasoning | 5/8 |
| Management | 6/8 |
| Communication | 7/8 |
Open History → missed marks:
Cardiovascular risk factors not explored.
Evidence from your encounter: you asked about the pain, the onset, the radiation. You never asked about smoking, diabetes, or family history.
Feedback that quotes you can't be argued with. It can only be fixed.
choose your exam track and specialty. 250+ stations, from first-week basics to finals pressure.
type, dictate, or speak. The patient stays in character the whole way through.
your transcript, notes and differential become the record. Nothing you did is lost.
checklist, domains, and evidence from your own words. Ready in seconds.
Stations built for UKMLA-style clinical assessment.
Explore UKMLA stations →250+ stations, built from a clinician's syllabus map of what exams actually test. The essential set — nothing padded.
Pick a pack that runs to your exam — 3, 6, or 12 months. Every pack: the full station library, study and exam modes, flashcards, and uncapped council marking.
Every clinical case passes clinician review before it reaches you. The marking rubrics are calibrated with practising clinicians. And we're honest about what AI is: MedExamHelper is a practice tool, not a patient, and no simulation replaces clinical teaching. It exists so the room isn't the first place you perform.
Drop your email and we will let you know when there is something worth reading. No account, no marketing spam.
