VIDA-TEC: Inferencia de signos vitales por video

Vital sign monitoring is a critical aspect of medical care and emergency situations. However, there are scenarios where traditional measurement methods are impractical or potentially harmful to the patient. This research project proposes the development of an innovative video-based system for non-invasive acquisition of vital signs, specifically heart rate and respiratory rate.
The proposed system utilizes advanced image processing and machine learning techniques to analyze subtle changes in skin coloration and chest movements, allowing the extraction of vital information without the need for physical contact. This approach is particularly valuable in situations where patients are inaccessible, such as in disaster areas, or when placing conventional sensors could be counterproductive, as in the case of patients with extensive burns or premature neonates.
The main objectives of this research include: Developing a robust algorithm for detecting and tracking regions of interest on the human body through video sequences. Implementing signal analysis techniques to extract heart and respiratory rates from temporal variations in the images. Evaluating the accuracy and reliability of the system compared to standard methods of vital sign measurement. Identifying and addressing technical challenges associated with different lighting conditions, patient movements, and interpersonal variability. Exploring potential use cases in clinical, emergency, and telemedicine environments.

Proponer algoritmos para la detección y seguimiento de regiones de interés en el cuerpo humano para seguimiento de signos vitales a través de secuencias de video.

Computing algorithms, AI and software design.