CPETlab takes you from the physiology, on curves you can bend yourself, to real de-identified studies, to the thing the course is actually for: picking up an unfamiliar test and saying what it shows.
Every module is a hook question, a short idea, something to drive, and a handful of questions. Nothing is a video you sit through.
Four units, and the difference between them is what the data is — idealised, then real, then measured against a guideline, then handed to you cold.
Simulated. Idealised curves and a simulator you drive yourself — change a variable, watch the nine panels answer.
Real de-identified studies. Every number on the screen is fitted from the patient’s own breath file at build time, never typed.
Guideline-led. Where a measurement meets a threshold somebody drew — listing criteria, risk tiers, operability.
Interpretation. You measure the study yourself, then read it one finding at a time and commit to a report.
The simulator is the engine the physiology unit is built on. Change a variable — stroke volume, dead space, the ventilatory set-point — and watch all nine panels answer at once. It is illustrative rather than measured, which is exactly what makes it safe to break.
Breath-by-breath files from real tests, carried through the whole course. You meet the same patients again and again, and they get harder to dismiss.
The slopes, thresholds and percentages in the text are computed from the patient's own data when the module is built. The prose cannot drift away from the plot.
Place the peak, the threshold, the fitting windows. Every number in your report comes from an anchor you put there, against limits computed for that patient.
Several studies here have landmarks that cannot honestly be read. Saying so is graded as correct — because producing a number anyway is the error the course is about.