Table 1.

Summary of the differences between clinical and modelling studies (examples of study design)

	Clinical study	Modelling study
Patient selection	Group of patients satisfying inclusion/exclusion criteria	Multiple variants of one or several models (modification of the substrate)
Indications for ablation	Drug‐resistant paroxysmal or persistent AF	AF episode of duration >10–30 s
Study design	Same therapy in many patients	Several therapies in the same model (or a few different models)
	Statistics over patients	Statistics over simulations in the same substrate
	The substrate is sometimes characterised	The substrate is specified
	AF mechanism is studied	AF mechanism is postulated and is an input to the model
Ablation procedure	Step‐wise ablation	Instantaneous application of all ablation lines
Procedure time	Up to a few hours	2–30 s of simulation
Side effects	Inflammation, risk of complications	None simulated
Criteria for AF termination	No AF after the procedure and until patient release	AF termination within 2–30 s
Criteria for AF prevention	No symptomatic AF episodes (48 h or several months of follow‐up)	All induction attempts failed (ectopic in the pulmonary vein)
Access to physiological data	ECG, electrical mapping	All variables everywhere all the time
Reproducibility	Limited by inter‐patient variability; AF episodes may vary in the same patient	Yes; variability can be controlled
Unique features	Real patient, real heart	Validation is critical
	Ablation is irreversible	Possibility to ‘undo’ an ablation line and try another one
	Risks for the patients, including mortality and morbidity	No human life at risk

AF, atrial fibrillation.