Evaluation of the responsiveness pattern to caffeine through a smart data-driven ECG non-linear multi-band analysis

Area of Science:

• Cardiology
• Pharmacology
• Biomedical Engineering

Background:

• Caffeine administration methods impact nervous system activity.
• Electrocardiographic (ECG) signals can reflect physiological responses to caffeine.
• Novel delivery systems for caffeine require evaluation against traditional methods.

Purpose of the Study:

• To compare the nervous system modulation effects of caffeine administered via oral films (OF_caffeine) versus expresso coffee (EC).
• To assess the efficiency and controlled release characteristics of caffeine oral films.
• To utilize ECG analysis and machine learning for differentiating caffeine delivery methods.

Main Methods:

• Non-linear multi-band analysis of ECG signals using Discrete Wavelet Transform (DWT).
• Feature extraction from ECG data and Principal Component Analysis (PCA) for dimensionality reduction.
• Application of 23 machine learning models with leave-one-out cross-validation to discriminate between caffeine groups.

Main Results:

• Caffeine Oral Films (OF_caffeine) demonstrated higher accuracy (62-84%) in distinguishing from placebo compared to expresso coffee (EC) (30-70%) in ECG analysis.
• Significant differences in accuracy curves were observed between OF_placebo and OF_caffeine groups.
• Caffeine OFs showed rapid dissolution and controlled release, differentiating them from EC.

Conclusions:

• Caffeine Oral Films represent a superior technology for caffeine delivery compared to traditional methods, evidenced by ECG analysis.
• OFs offer efficient control over caffeine release kinetics.
• This study supports the potential of OFs for future caffeine delivery systems.