Cameras offer new possibilities in assessing human states like stress. This work addresses the measurement of respiration rate with several approaches. Data from a mental stress study (> 40 h recording time, 54 participants) were evaluated and color channel combinations were examined using a hemispherical surface grid search with mean absolute error M AE to optimize the approaches. The grid search converged towards the green channel in the baseline modulation approach (M AE = 2.53 rpm). However, best results were achieved with the frequency modulation approach (M AE = 2.13 rpm) with the color channel combination optimal for heart rate measurement. Respiration rate increased highly significant during stress (p < 0.001, Mann-Whitney U-test). Deliberate selection of the color channel combination is crucial for respiration rate measurement with cameras in order to assess human states or pathophysiological processes.

Imaging photoplethysmography (iPPG) enables the extraction of physiological signals from standard RGB video recordings. For the assessment of the human health condition, pulse pressure is of utmost importance and is usually determined from conventional blood pressure signals.Within this work we present the fully automated estimation of pulse pressure using iPPG. We computed the pulse strength from the iPPG signals and performed a linear correlation analysis with the corresponding pulse pressure. We compared different algorithmic iPPG approaches amongst one is an artificial neural network. We measured a maximum pearson correlation of 0.65 for the artificial neural network and 0.63 for the best conventional approach. Our results show 0.1 increase in correlation coefficient compared to previous work based on manual processing, demonstrating the feasibility of automated contactless pulse pressure estimation from RGB videos.

Atrial fibrillation (AF) is our society's most common cardiac arrhythmic disease, leading to increased morbidity and mortality. Predicting AF episodes during sinus rhythm based on electrocardiograms (ECGs) allows timely interventions. It is known, that changes in selected ECG morphology features are a predictor for the onset of AF, but no systematic investigation of different ECG features' temporal changes has been performed so far. We split sinus rhythm episodes of 60 minutes preceding AF from the MIT-BIH AF database into segments of 5 minutes with 50% overlap (n = 644) and calculated 155 features of different domains per segment. Logistic regression analyses between the segments preceding AF and others revealed the most significant effects for segments ending 5 minutes before AF onset, with PQ interval slope (p < 0.01), PQ interval correlation (p < 0.05), and median RR time (p < 0.05) being the most relevant features. A decision tree ensemble, trained with all features, achieved an accuracy of 0.87 when distinguishing 8 segment clusters. Our results confirm expected changes in ECG features (e.g., PQ interval) before AF episodes, indicating impaired atrial excitation, and show that the combination of interpretable features is sufficient to discriminate at different points in time before AF onset. For advanced analyses, more extensive databases should be included.