Date of Award

Spring 2019

Document Type



Santa Clara : Santa Clara University, 2019



First Advisor

Prashanth Asuri

Second Advisor

Emre Araci


Hand dexterity and motor control are critical in our everyday lives because a significant portion of the daily motions we perform are with our hands and require some degree of repetition and skill. Therefore, development of technologies for hand and extremity rehabilitation is a significant area of research that will directly help patients recovering from hand debilities sustained from causes ranging from stroke and Parkinson’s disease to trauma and common injuries. Cyclic activity recognition and assessment is appropriate for hand and extremity rehabilitation because a majority of our essential motions are cyclic in their nature. For a patient on the road to regaining functional independence with daily skills, the improvement in cyclic motions constitutes an important and quantifiable rehabilitation goal. However, challenges exist with hand rehabilitation sensor technologies preventing acquisition of long-term, continuous, accurate and actionable motion data. These challenges include complicated and uncomfortable system assemblies, and a lack of integration with consumer electronics for easy readout. In our research, we have developed a glove based system where the inertial measurement unit (IMU) sensors are used synergistically with the flexible sensors to minimize the number of IMU sensors. The classification capability of our system is improved by utilizing a fuzzy logic data analysis algorithm. We tested a total of 25 different subjects using a glove-based apparatus to gather data on two-dimensional motions with one accelerometer and three-dimensional motions with one accelerometer and two flexible sensors. Our research provides an approach that has the potential to utilize both activity recognition and activity assessment using simple sensor systems to help patients recover and improve their overall quality of life.