A personalized model for monitoring vital signs using camera of the smart phone

Abstract

Smart phones with optical sensors have created new opportunities for low cost and remote monitoring of vital signs. In this paper, we present a novel approach to find heart rate, perfusion index and oxygen saturation using the video images captured by the camera of the smart phones with mathematical models. We use a technique called principal component analysis (PCA) to find the band that contain most plethysmographic information. Also, we showed a personalized regression model works best for accurately detecting perfusion index and oxygen saturation. Our model has high accuracy of the physiological parameters compared to the traditional pulse oxymeter. Also, an important relationship between frame rate for image capture, minimum peak to peak distance in the pulse wave form and accuracy has been established. We showed that there is an optimal value for minimum peak to peak distance for detecting heart rate accurately. Moreover, we present the evaluation of our personalized models.