Crucial Events Identify Emotion Granularity from Long-Term ECG Recordings

The increasing interest in improving the accessibility and implementation of psychiatric solutions in diagnosing and treating mental and neurological disorders is driven by the need for real-time patient monitoring. One promising approach is emotion recognition using physiological signal complexity detection. Complexity measures involving crucial events, which are brief intervals of intermittent turbulence that resemble fractal-like behaviour and a part of temporal biosignals have been used to analyze physiological signals, with the assumption that healthy and pathological signals differ in their levels of complexity. However, there is limited knowledge about the relationship between physiological signals, and psychopathology. Changes in emotion are reflected in heartbeat variations, and valence and arousal are psychological features of emotion. Crucial events, patterns in the heart rate that identify instances of change, can be detected using the novel multiscaled modified diffusion entropy analysis (MSMDEA), which has been shown to distinguish healthy from pathologic cardiac signals and different types of pathologic signals at high statistical significance (p < 0 0001) compared to using MDEA on its own.