Artificial intelligence is increasingly being used in medicine, including clinical physiology. This is facilitated by the increase in computing processing power, the development of cloud services and datasets, and numerous scientific articles demonstrating the effectiveness and viability of such intelligent solutions. Although the approach to medical datasets development is generally similar, there are a number of key features and significant differences in clinical physiology ones. Artificial intelligence systems in clinical physiology may be effectively trained and applied in practice by following the recommendations proposed in this article.