AN INTELLIGENT WEARABLE REMOTE HEALTH MONITORING SYSTEM FOR PREVENTIVE MEDICINE
DATE: July 11, 2018
TIME: 11:00 am
PLACE: BRG 423
Associate Professor of Mechanical Engineering
Texas A&M University at Qatar
The pressure on available health care resources has increased in the recent years due to the rise in the number of patients suffering from physical disorders including lung diseases, diabetes, and cardiovascular disease. Remote health monitoring systems, which collect vital signs, can provide the opportunity to remotely access medical information quickly, interactively, and inexpensively, without the presence of a physician unless deemed necessary by the obtained information. However, technical challenges associated with physiological data acquisition, transmission, and accurate analysis has hindered this concept from being widely commercialized and clinically accepted. The aim is twofold. First, to design an advanced prototype of health monitoring system to remotely and securely track physiological data including continuous heart rate, respiratory rate, blood pressure, blood glucose, oxygen saturation, and electrocardiogram (ECG). Second, to develop an efficient software that can accurately determine the subject’s current health status and provide a medically meaningful data to a physician for a precise decision-making process. The objective is to prototype an advanced remote health monitoring system incorporating both innovative hardware and software designs and to develop intelligent algorithms, fuse and process the data acquired by different sensors and transmitted wirelessly to a central processing unit. This system provides healthcare providers valuable information about the health status of their patients. The analysis of the acquired sensory signals is performed through the application of advanced signal processing techniques including adaptive filtering and trend detection, removal of noise and artifacts, high quality signal transmission, secure wireless communication, extraction of relevant parameters and vital signs, pattern recognition, and summarizing the results in texts, graphs, and tables that are clinically interpretable. Our research team has already developed an efficient algorithm that can accurately detect the QRS features from the complex ECG data (99.6%) and identify the myocardial infarction (95.7% sensitivity and 94.6% specificity), even in the presence of waveform fluctuations in baseline and amplitudes. The developed algorithms are implemented in the hardware prototyping.
Light snacks will be served!