A Physiological Telemetry System to Detect the Onset of Autonomic Dysreflexia in Individuals with Spinal Cord Injuries
Every year persons with a spinal cord injury (SCI) are rehospitalized due to secondary health complications. One of the more common and serious post-SCI health conditions is Autonomic Dysreflexia (AD). Lack of awareness and poor management of AD can be life threatening. Key physiological indicators of AD include increased blood pressure, altered heart rate, cold and clammy skin, and pathological sweating above the level of injury. Utilizing commercially-available galvanic skin resistance (GSR), heart rate, and skin temperature sensors, we developed a mobile computer application to enable wearers to monitor many of these physiological indicators in real-time and throughout the day. A classifier algorithm was used to develop a machine learning model that could predict the occurrence of AD in real time with a high level of accuracy. The objective of this study is to use physiological signals to develop a universal recognition system to reliably detect the onset of AD in individuals with tetraplegia without prior training. The system uses a commercially available wearable device, the Microsoft Band to collect data with a high level of precision. Through this study we determined that skin temperature and GSR were reliable physiological parameters to detect the onset of AD and the system uses normalized GSR and skin temperature data collected from individuals with SCI to detect AD with almost 95% accuracy and predict the onset of AD at most 10 minutes prior to the recognition of symptoms by the individuals. The system was shown to be easy to use, comfortable to wear and most importantly, helpful to individuals who experienced AD. The goal of this physiological telemetry system (PTS) is to assist newly injured individuals to train themselves to identify their own AD triggers and symptoms and provide continuous medical oversight via telemetry.
Duerstock, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our