Divya Saxena

ORCID Identifier(s)

0000-0003-1420-6424

Graduation Semester and Year

2022

Language

English

Document Type

Thesis

Degree Name

Master of Science in Computer Science

Department

Computer Science and Engineering

First Advisor

Manfred Huber

Abstract

In recent times, people have increasingly self-assessed their health using different devices on their bodies that monitor physiological attributes such as their oxygen level and blood pressure (BP) to monitor their health. One of the most popular health concerns that became prominent during the COVID-19 pandemic was the blood oxygen saturation (SPO2) level. It became increasingly important to monitor SPO2 in patients, time and again to determine whether the right amount of oxygen is in the blood. Low oxygen levels usually indicate there may be an issue with oxygen circulation or supply and thus informs diagnostic and treatment decisions such as transfer to the hospital or ICU or the application of external oxygen. The use of existing devices requires active interaction of the user to obtain the information about blood flow needed for SPO2 or blood pressure measurements. However, a change in skin color could also provide information whether a body has a healthy blood flow or not. Skin color can easily be captured with the help of conventional cameras under different lighting conditions. However, under insufficient lighting, it is strenuous to capture accurate images. A digital camera with the ability to capture near Infrared (NIR) images can help resolve this issue by allowing for better illumination control without inconveniencing the user. If blood flow information can be effectively extracted, RGB and NIR image sequences can be used to estimate blood oxygen level (SPO2), systolic blood pressure (SBP), and diastolic blood pressure (DBP). In this study, methods are proposed that estimate various health parameters like blood oxygen saturation (SPO2), systolic blood pressure (SBP), and diastolic blood pressure (DBP) from multi-spectral video. An RGB-NIR camera was used to record the data. Colored and near-infrared images from the camera have been recorded to extract blood-related health details. Using the existing Photoplethysmography technology that is widely used in commercial devices for measuring oxygen saturation, and BP, we analyzed the photoplethysmogram (PPG) signals using an RGB-NIR camera. The camera was placed at an approximated distance of 3 meters from the subject. For our research, we recorded the subject’s facial area. Furthermore, we analyzed the traditional red, green, and blue channel combination, as well as the additional IR channel. The results were obtained using a multi-spectral camera with a frame rate of 30 Hz per second. Preliminary results showed statistically that an RGB-NIR camera could potentially be an efficient alternative to conventional medical devices to measure SPO2 and BP, achieving good prediction accuracy for SPO2 and diastolic blood pressure while lacking somewhat in the prediction of systolic pressure. The system proposed is convenient, safe, contact-less, and cost-effective. The pandemic is still rampant and with many companies resorting to contact-less services, it is only necessary and smart to have a contact-less method of arterial oxygen saturation and BP estimation. This technology has significant potential in advancing healthcare.

Keywords

SPO2-blood oxygen saturation, BP-blood pressure, HB-de-oxyhemoglobin, HBO2-oxyhemoglobin

Disciplines

Computer Sciences | Physical Sciences and Mathematics

Comments

Degree granted by The University of Texas at Arlington

Download

Share

COinS