Contributions to heart rate variability and respiratory rate analysis through video imaging techniques

The present doctoral thesis proposes a series of methods that have as a general objective the non-contact evaluation of heart rate variability (HRV) and respiratory rate (BR) through video imaging. The proposal of this technique has emerged in the last few years as an alternative to the traditional measurement systems, which assess these and other physiological parameters through direct contact with the person’s body. Due to this condition, the traditional systems may cause some complications in people who present a sensitive or fragile skin, such as neonates or patients with burn injuries. Furthermore, although the video imaging technique could be an excellent alternative, as in the cases mentioned above, it also aims at being a monitoring instrument for the general population in non-clinical environments. In fact, nowadays, the use of electronic devices such as the video cameras, smartphones, tablets and others, it starts being part of a healthy lifestyle. Thus, this promising technique, which could provide advantages such as the contactless measurement, portability, easy use and low-cost, it could be employed in the near future as part of the evaluation of the person’s health status in everyday life. On the other hand, despite the aforementioned advantages, the measurements obtained by this technique may be greatly affected by factors such as the movement of the person, lighting conditions, camera settings, among others. Hence, an analysis of the influence of various factors and conditions is carried out in this research study in order to obtain a better insight of the scope and limitations of the technique. For this purpose, a series of methods were developed and implemented in which face detection and tracking algorithms are employed, as well as image and signal processing techniques. The results obtained in each study are evaluated by using various statistical parameters and plots with the aim of measuring the agreement between the proposed methods and the reference systems. In general, the statistical analysis carried out shows a good level of agreement between the measurement systems. It is important to note the presence of an impact on the results, to a greater or lesser extent, by the factors and conditions analyzed in the different studies. In several cases, the results show a significant improvement in comparison with the data reported in related studies. By contrast, the results achieved in scenarios with a greater presence of artifacts show a