Disclosed are various examples and embodiments of systems, devices, components and methods configured to detect a location of a source of at least one cardiac rhythm disorder in a patient's heart. In some embodiments, electrogram signals are acquired from a patient's body surface, and subsequently normalized, adjusted and/or filtered, followed by generating a two-dimensional spatial map, grid or representation of the electrode positions, processing the amplitude-adjusted and filtered electrogram signals to generate a plurality of three-dimensional electrogram surfaces corresponding at least partially to the 2D map, one surface being generated for each or selected discrete times, and processing the plurality of three-dimensional electrogram surfaces through time to generate a velocity vector map corresponding at least partially to the 2D map. The resulting velocity vector map maybe employed to classify a patient as one of an A-type patient, a B-type patient, and a C-type patient, and to guide therapy subsequently delivered to the patient. Trained atrial discriminative machine learning models that facilitate the foregoing systems and methods, and that provide predictions or results concerning a patient's condition, are also disclosed.
