video1.0<iframe src="https://www.loom.com/embed/a55d0385302942a3ab46fb01087d79af" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/a55d0385302942a3ab46fb01087d79af-1ef33a40fa87199f.gif180.544This Loom explains Conduction Lens, a method to determine which ECG conduction parameters can be reliably inferred. The approach builds a heart model with a Purkinje network using anatomical parameters, generates a simulated ECG, and runs adversarial exchanges to test ECG calibration within that simulation. It uses an NPE estimator to rank, parameter by parameter, which ones provide the best information and where confidence is highest. The system trains a network to invert ECGs and output a parameter distribution, then uses simulations to grade calibration and report the anatomies and conductance parameters that produce observed ECGs.