The Noninvasive Measurement of Hemoglobin by Light Sources in Various Wavelengths

Monitoring human body vital signs like heart rate, oxygen saturation and blood pressure, has a profound influence on recognition of cardiovascular diseases which are growing at unprecedented rate all over the world. In recent years, using imaging photoplethysmography (IPPG) signals is one of the most interesting issues among researchers to measure the vital signs of the human body. Decreasing the values of hemoglobin in blood, which is called Anemia, and it's more common among women, can be detected through the processing of the IPPG signals. In this article, the magnitude of hemogolobin level is measured by a suggested approach applied on the IPPG signals taken by means of a physical setup. To make the signals, after capturing video from the fingertip pulse of index right finger with various light sources in wavelengths consisting of white, 520nm and 980nm; the IPPG signals will be accessible as a result of applying the proposed algorithm on the videos. In the next step, providing appropriate signals to the implementation of the regarded method, the signals are preprocessed. Considering physics-based models, the time domain features are extracted. In the final step, utilizing the support vector regression, accuracy of the prediction is 82%, which is shown reliability, repeatability, and reproducibility of the designed configuration.