This page describes how The ECG Quiz evaluates user interpretations using a hierarchical scoring system that incorporates Must and Should diagnoses, handles uncertainty, and measures diagnostic accuracy with precision, recall, and F1 metrics. Several case examples below illustrate how the scoring system works.

Interpreting ECGs requires a structured, repeatable approach. Experienced ECG readers use a systematic sequence to ensure that all essential features of the tracing are analyzed. This method helps distinguish objective findings (e.g., QRS presence) from more subjective assessments (e.g., nonspecific T-wave abnormalities). Because ECG interpretation combines both objective and subjective elements, The ECG Quiz incorporates a scoring algorithm that accounts for these differences while providing fair and consistent feedback across all skill levels.

The table below summarizes the standard review sequence used by expert interpreters.

Historically, ECG experts frequently disagreed when interpreting complex tracings—especially when findings involved borderline abnormalities or subtle repolarization patterns. This natural variability is one reason The ECG Quiz uses a hierarchical and skill-level–based scoring method, ensuring that users receive appropriate credit and feedback, whether the finding is objective or somewhat subjective.

To accommodate learner variability, The ECG Quiz classifies each required diagnosis for an ECG as either:

• Must — Required at the user’s selected skill level
• Should — Helpful to include, but optional

A Beginning Learner may only be expected to identify “Supraventricular Tachycardia,” while Intermediate and Advanced learners are expected to specify the mechanism, such as:

• Atrial Tachycardia
• Junctional Tachycardia
• Sinus Tachycardia

These Must/Should assignments vary by skill level and by each individual ECG.

The ECG Quiz also uses a hierarchical diagnosis system, in which:

• More general diagnoses sit higher in the hierarchy
• More specific diagnoses sit lower (e.g., “Atrial Flutter with Variable Block” is more specific than “Atrial Flutter”)

This hierarchical structure allows learners to receive appropriate credit when identifying more specific—or less specific—diagnoses depending on their skill level.

The ECG Quiz uses a standardized set of scoring terms to compare your interpretation with the correct interpretation. Each diagnosis you submit is categorized using the following definitions.

These terms form the basis for calculating:

• Precision = TP / (TP + FP)
• Recall = TP / (TP + FN)
• F1 Score = Harmonic mean of precision and recall

These metrics ensure fair, consistent feedback across all skill levels.

• If you include the diagnosis → True Positive (TP)
• If you omit it but include a more specific diagnosis within its hierarchy → Not Counted (NC)
• If you omit it and do not include any more specific alternative → False Negative (FN)

• If you include it → True Positive (TP)
• If you omit it → Not Counted (NC)

• → False Positive (FP)

Because ECG diagnoses form a hierarchy of general → specific patterns, the scoring system adjusts outcomes depending on which diagnoses you include.

These adjustments prevent users from receiving double credit or unfair penalties.

If you omit a Must diagnosis but include a more specific Should diagnosis on its hierarchy:

• Omitted Must → Not Counted (NC)
• Included Should → True Positive (TP)

Example: An ECG shows *Atrial Flutter with 2:1 Block*. At Beginner level:

• Must: Atrial Flutter
• Should: Atrial Flutter with 2:1 Ventricular Response

If you enter *Atrial Flutter with 2:1 Ventricular Response*, you receive a TP, and the Must (*Atrial Flutter*) becomes NC because you selected the more specific diagnosis.

If you omit the Must and include a specific diagnosis that is in the hierarchy but not in our Must/Should lists:

• Omitted Must → True Positive (TP)
• Your incorrect specific diagnosis → False Positive (FP)

Example: Entering *Atrial Flutter with Variable Response* instead of *Atrial Flutter with 2:1 Response*:

• Flutter (general) → TP
• Incorrect specific type → FP

You correctly identified *Flutter* but not the specific pattern.

If you include both a general Must diagnosis and a more specific Should diagnosis:

• Specific Should → True Positive (TP)
• General Must → Not Counted (NC)

This prevents double credit for overlapping diagnoses.

Example: Beginners analyzing a narrow-complex tachycardia:

• Must: Supraventricular Tachycardia
• Should: Junctional Tachycardia

Entering *both* yields:

• Junctional Tachycardia → TP
• Supraventricular Tachycardia → NC

You get credit for the more specific diagnosis.

You may reread ECGs as many times as you wish. Your statistical summary tracks performance trends for:

• Each ECG
• Each diagnosis
• Each interpretation attempt

You may archive your current results to reset your statistics at any time.

Good luck, and have fun interpreting ECGs!