|M.Sc Student||Busool Rami|
|Subject||Wearable Chemical Sensors for Monitoring of Infectious|
|Department||Department of Chemical Engineering||Supervisor||Professor Hossam Haick|
About 23% of the global population is estimated to be contaminated with Mycobacterium tuberculosis and among poor / low-income nations, tuberculosis (TB) accounts for 1.4 million deaths per year and one-fifth of adult deaths. When left untreated, each patient with active pulmonary TB is expected to infect 10-15 other people per year. Interrupting the transmission of disease is therefore of great importance and requires early detection in conjunction with appropriate therapy.
There are serious limitations on the markers / tests currently available for TB diagnosis, and none is a point-of-care (POC) diagnostic test. In this research, I studied, created and validated a new non-invasive wearable sensing patch from the skin for the detection of infectious disease at the point of care, specifically tuberculosis (TB), with the ability to serve as a tool for testing, tracking and infection control. The patch uses a novel, intelligent, nano-material-based sensor array to detect disease-specific volatile organic compounds (VOCs) from the skin's surface, allowing for rapid and highly accurate diagnosis using a small device. Through my work, I introduced and developed a wearable diagnostic device that could connect a variety of sensors, i.e. chips with 8 heterogeneous sensors, with different Monolayer-Capped Nanoparticles (MCNPs) and form an electronic acquisition network that acquires measured sensor responses and transmits them to computer labs or even cloud processing. The wearable device developed, i.e. the patch, was first validated in the laboratory before we used it in a prospective clinical study in Latvia. In total, we examined 51 subjects (21 TB patients, 30 healthy) in two different body areas; chest and arm. The data we received from the patch was analysed in different PCA methods and I obtained the best result from the implementation of Linear Discriminate Analysis (LDA) with 82.76% sensitivity, 100% specificity, 85.76% accuracy, respectively, and 0.93 receiver operating characteristic.
Furthermore, in the laboratory exposure experiment to TB and non-TB volatile organic compounds (VOCs), I experimented the same sensors used in the clinical study that imitate the released VOCs from the human body. The acquired data was evaluated and validated and provided the sensor performance optimisation in a temperature range of 32-38 ° C -human skin temperature range- and various concentrations ranging from part-per-billion (ppb) to part-per-million (ppm). As a result, the sensors show a good 6 ppb detection limit (LOD) and a strong capacity for differentiation between TB and non-TB VOCs. The developed wearable device (WED) is a promising non-invasive tool, compact, cheap (less than $5) and accurate proof of concept diagnostic system with a high potential to be used in third world countries and endemic regions, as a diagnostic method for TB detection. In addition, unlike current diagnostic approaches, the new patch can be applied either in the doctor's office or by the home patient allowing the patient to participate actively in TB prevention and detection.