About and methods
How our EKG strips are made, and how we check them
ekgstrips.com is a practice tool. Its strips are drawn by software, not recorded from patients, so you are entitled to know exactly how they are made and how we know they are right.
Last updated November 12, 2026
Who we are
ekgstrips.com is built and maintained by the ekgstrips.com team, a small independent team of software engineers. We write the waveform engine, the question logic, the explanations and the tests ourselves. Everything on this site, including this page, is written by the team.
We are engineers, and the method below is built so that you do not have to take our word for anything: every rhythm is built from published criteria, and every strip is measured by an independent program before you see it. If you are a telemetry nurse, an educator or a cardiology professional and you see something that does not look right, we want to hear from you at [email protected]. We fix confirmed errors and credit reviewers who wish to be named.
How a strip is made
- A rhythm engine decides the timing. For each rhythm it sets when the atria and the ventricles fire: the rate, how regular it is, whether each P wave is followed by a QRS, and how the PR interval behaves from beat to beat. A Wenckebach strip, for example, is a sequence of PR intervals that lengthen by smaller and smaller steps until one QRS is dropped.
- Each beat is drawn as a sum of smooth waves. The P wave, the Q, R and S deflections and the T wave are each a bell-shaped curve placed at the right time with the right height and width. The idea of building a synthetic ECG from a small number of such curves comes from McSharry and colleagues (2003); the implementation here is our own.
- The QT interval follows the heart rate. It is set with Bazett's formula for a corrected QT of 0.40 s, so the T wave moves closer to the QRS as the rate rises.
- The trace is drawn to scale. Paper speed is 25 mm/s and gain is 10 mm/mV, the standard calibration. One small square is 0.04 s by 0.1 mV; one large square is 0.20 s by 0.5 mV. A little baseline wander and muscle artifact is added.
- Ambiguous strips are thrown away. A 6-second window can cut a rhythm so that it no longer shows what makes it that rhythm. Before a strip is used it must show enough to be read: a second-degree block must show a whole cycle including the dropped beat, a complete block must show at least three different PR intervals and more P waves than QRS complexes, and no QRS may be cut by the edge of the paper. We never generate 2:1 AV block as type I or type II, because the two cannot be told apart on such a strip. For the same reason atrial flutter is never drawn with fixed 2:1 conduction (2:1 cycles appear only next to longer ones that show the flutter waves), and premature beats are kept clear of the T wave in front of them so that they can be seen.
Every strip comes from a number called a seed. The same seed always draws the same strip, which is how the three daily strips are identical for every visitor and how any reported problem can be reproduced exactly.
The criteria each rhythm is built to
These are the textbook definitions the engine generates from and the tests check against. Rates are per minute.
| Rhythm | Rate | Rhythm | PR interval | QRS |
|---|---|---|---|---|
| Normal sinus rhythm | 60–100 | Regular | 0.12–0.20 s, constant | < 0.12 s |
| First-degree AV block | Underlying rhythm | Regular | > 0.20 s, constant | < 0.12 s |
| Second-degree AV block, type I | Atrial > ventricular | Grouped beats | Lengthens until a QRS is dropped | < 0.12 s |
| Second-degree AV block, type II | Atrial > ventricular | Irregular at the dropped beat | Constant before and after the dropped QRS | Narrow or wide |
| Third-degree AV block | Atrial > ventricular; ventricular 20–60 | P-P regular, R-R regular | Varies; no relationship | Narrow or wide |
| Atrial fibrillation | Variable | Irregularly irregular | None; no P waves | < 0.12 s |
| Supraventricular tachycardia | 150–250 | Regular | Not measurable | < 0.12 s |
| Idioventricular rhythm | 20–40 | Regular | None | ≥ 0.12 s |
| Ventricular tachycardia | 100–250 | Regular | None | ≥ 0.12 s |
| Ventricular fibrillation | Not countable | Chaotic | None | None |
| Asystole | 0 | None | None | None |
| Sinus bradycardia | < 60 | Regular | 0.12–0.20 s, constant | < 0.12 s |
| Sinus tachycardia | 101–150 | Regular | 0.12–0.20 s, constant | < 0.12 s |
| Sinus arrhythmia | 60–100 | R-R varies by more than 0.16 s, gradually | 0.12–0.20 s, constant | < 0.12 s |
| Sinus arrest | Underlying sinus rate | A pause over two cycles long, not a whole number of cycles, with no P wave in it | 0.12–0.20 s, constant | < 0.12 s |
| Premature atrial contractions | Underlying sinus rate | An early beat with its own P wave; pause less than compensatory | Normal | < 0.12 s |
| Atrial flutter | Atrial 250–350 | Regular (3:1, 4:1) or variable (2:1 to 4:1) | None; sawtooth flutter waves | < 0.12 s |
| Junctional rhythm | 40–60 | Regular | No upright P wave; an inverted one may sit just before the QRS (PR < 0.12 s) or just after it | < 0.12 s |
| Accelerated junctional rhythm | 60–100 | Regular | None; no upright P wave | < 0.12 s |
| Premature ventricular contractions | Underlying sinus rate | An early beat with no P wave; full compensatory pause | Normal on sinus beats | ≥ 0.12 s on the early beat |
| Ventricular bigeminy | Underlying sinus rate | Every other beat a PVC | Normal on sinus beats | Alternates |
| Accelerated idioventricular rhythm | 40–100 | Regular | None | ≥ 0.12 s |
| Torsades de pointes | 200–250 | Steady beat, amplitude in spindles | None | Wide, changing |
| Ventricular paced rhythm | 60–90, fixed | Regular | None; a spike starts each QRS | ≥ 0.12 s |
| Junctional tachycardia | 101–149 | Regular | No upright P wave; an inverted one may sit just before the QRS (PR < 0.12 s) | < 0.12 s |
| Ventricular trigeminy | Underlying sinus rate | Every third beat a PVC | Normal on sinus beats | Every third one wide |
| PVC couplet and triplet | Underlying sinus rate | Two or three PVCs in a row, over 100 a minute within the run | Normal on sinus beats | ≥ 0.12 s in the run |
| Multifocal PVCs | Underlying sinus rate | Two PVCs of opposite direction | Normal on sinus beats | ≥ 0.12 s on the PVCs |
| Atrial paced rhythm | 60–80, fixed | Regular | Normal; a spike starts each P wave | < 0.12 s |
| AV paced rhythm | 60–80, fixed | Regular | Two spikes a fixed delay apart | ≥ 0.12 s |
| Failure to capture | Below the set rate | One or two spikes with no QRS; the pause is two paced intervals | None | ≥ 0.12 s |
| Failure to pace | Below the set rate | A pause of two or three paced intervals with no spike in it | None | ≥ 0.12 s |
| Failure to sense | 60–100 (the patient’s own) | Regular sinus rhythm; spikes at a fixed interval fall 0.13–0.36 s after a QRS | 0.12–0.20 s | < 0.12 s |
| Sinus rhythm with bundle branch block | 60–100 | Regular | 0.12–0.20 s, constant | ≥ 0.12 s, notched |
| Wolff-Parkinson-White pattern | 60–100 | Regular | < 0.12 s | ≥ 0.12 s, starting with a delta wave |
| Wandering atrial pacemaker | Under 100 by count | Irregular | Changes with the P wave; at least three P wave shapes | < 0.12 s |
| Multifocal atrial tachycardia | Over 100 by count | Irregular | Changes with the P wave; at least three P wave shapes | < 0.12 s |
| Atrial fibrillation with PVCs | By count | Irregularly irregular; one or two wide beats, each followed by a pause | None | < 0.12 s; PVCs ≥ 0.12 s |
| First-degree AV block with PVCs | Sinus rate 60–100 | Regular with one or two early wide beats | > 0.20 s on sinus beats | < 0.12 s; PVCs ≥ 0.12 s |
| Sinus bradycardia with first-degree AV block | Under 60 | Regular | > 0.20 s, constant | < 0.12 s |
Pulseless electrical activity (PEA) is shown in the lethal rhythm section but is never asked as an identification question: on paper it is an organized rhythm, and it can only be recognized by knowing the patient has no pulse.
How every strip is checked
Knowing what we meant to draw is not the same as knowing what ended up on the paper. So the check is done by a second program that is deliberately kept ignorant.
A blind measurer
The measurer receives only the finished trace: a list of voltages. It is not told which rhythm it is looking at or where the beats are. It works the way a reader with calipers does. It finds the R waves, finds where each QRS leaves and returns to the baseline, finds the P waves by their shape, pairs each P wave with the QRS that follows it, and measures the intervals. It also looks for pacing spikes and flutter waves. It then applies the standard five-step read (rate, rhythm, P waves, PR interval, QRS) and names the rhythm.
What must agree
- The rhythm the measurer names must be the rhythm the engine says it drew. Naming a different rhythm, or being unable to decide, is a failure.
- The measured heart rate, PR interval, QRS width and atrial rate must match the reported values to within about half a small square (0.02 s).
- Every number quoted in an explanation, such as a run of lengthening PR intervals, must be measurable on that strip.
- The drawn line must follow the computed signal to within 0.045 mm.
- For each measurement question, the measured value must fit exactly one of the four choices.
How much is tested
- 69,000 strips are generated and measured on every test run: 1,500 of each of the 42 rhythms, plus 1,500 of each AV block at its most subtle setting.
- 12,000 strips of each rhythm were run as a stress test, which surfaced and led us to fix cases that occur about once in ten thousand strips.
- The next 1,100 days of daily strips are checked in advance, so each day's three questions have exactly one answer that fits the trace.
- The tests themselves are tested. We break the engine on purpose in more than forty different ways (a premature beat that is not early, flutter waves at the wrong rate, a pacing spike too small to see, a pause that is a whole number of cycles, and so on) and confirm that the tests catch every one.
- No giveaways. The height and width of the waves vary from strip to strip within the normal range, and a test checks that no rhythm can be told from its look-alike by anything other than the feature that defines it.
No change to the engine is released unless all of these pass.
How explanations are written
Explanations are produced by rules from the strip's own measurements. When an explanation says the PR interval runs 0.16, 0.21, 0.25 and then a beat is dropped, those are the values on the strip in front of you. We do not use an AI model to interpret strips or to write explanations.
How to use these strips
- Textbook strips, as on an exam. Each strip follows the textbook definition of its rhythm and is clean enough to measure, the way teaching and exam strips are.
- Your course comes first. Our tests prove that the strips match the criteria in the table above. Where your instructor, hospital or exam uses a different cut-off, follow your course.
- Lead II, six seconds. The rhythm strip format that dysrhythmia tests use. Findings that take a 12-lead, such as naming a bundle branch block right or left, are described in words on the rhythm pages.
- One right answer. A variant that cannot be settled from one strip, such as atrial flutter with fixed 2:1 conduction, is kept out of the questions and described on its rhythm page.
- Corrections are fixed and credited. When a nurse or an educator reports an error, we redraw the strip from its seed, check it against the criteria, fix what is wrong and name the reviewer if they wish.
- Treatment notes are study summaries. At work, follow current guidelines and your facility's policies.
ekgstrips.com is for education and practice only. See the medical disclaimer.
Sources
- Relias. Dysrhythmia (Basic) Exam Content Outline, 2021. (The share of each rhythm category in the “Relias Basic outline” practice test and in the printable paper.)
- McSharry PE, Clifford GD, Tarassenko L, Smith LA. A dynamical model for generating synthetic electrocardiogram signals. IEEE Transactions on Biomedical Engineering. 2003;50(3):289–294. (The idea of composing a beat from a few bell-shaped waves. We do not use the authors' software.)
- Kligfield P, Gettes LS, Bailey JJ, et al. Recommendations for the standardization and interpretation of the electrocardiogram, part I. Circulation. 2007;115(10):1306–1324. (Standard calibration: 25 mm/s and 10 mm/mV.)
- Bazett HC. An analysis of the time-relations of electrocardiograms. Heart. 1920;7:353–370. (Rate correction of the QT interval.)
- University of Arkansas. How to Interpret Rhythm Strips, in the open EKG manual. (The five-step method and the normal ranges.)
- American Heart Association. CPR and ECC algorithms. (Shockable and non-shockable arrest rhythms.)
Independence
ekgstrips.com is independent. It is not affiliated with or endorsed by Relias, NCSBN (NCLEX), NHA or the American Heart Association. Every strip and every question is our own. The site is paid for by its users, so there is no advertising.
Contact
Corrections, questions and review offers: [email protected]. See also the contact page.