Movement sensor’s feedback acts as physical therapist

NSF REU study
Kindle, left, attaches sensor to study participant’s arm.

PULLMAN, Wash. – Brooks Kindle sits at his desk to demonstrate the simple stretch he has used all summer to collect data. He has divided the stretch into four parts: lowering his forearm to the desk, raising it back into the air, lowering it again and finally sitting back in his chair.

Kindle, a junior in computer science at Washington State University, does this stretch with a rectangular sensor about the size of a garage door opener attached to his arm. It collects movement data. As part of a WSU research experience for undergraduates (REU), Kindle is developing a sensor that can act as a physical therapist – reading a patient’s movements during exercise and suggesting improvements.
“With a sensor like this you could do stretches from the comfort of your home and get instant feedback on how well you are performing and what you can improve,” he said.
Supported by NSF
Nirmalya Roy, clinical assistant professor in the WSU School of Electrical Engineering and Computer Science (EECS), is Kindle’s advisor and the lead of the smart environments REU, which is funded by the National Science Foundation. NSF supports REUs all over the country to introduce undergraduates to research and provide a unique, out-of-classroom experience.
movement sensor on armKindle is one of more than 60 students from around the country and WSU participating in REUs with WSU faculty this summer. His part of the research focuses specifically on data mining, which involves collection, processing and interpretation of data.
“When I started a few months ago, I heard data mining and thought, ‘what’s that?’ It was a struggle to learn what it meant and how to do it, but I’ve definitely become more knowledgeable about it,” he said.
Gathering, interpreting data
The data comes from the sensor, which takes measurements of acceleration and orientation while the patient is exercising. Kindle transfers the sensor data to a computer, breaks the numbers into smaller groups based on the four parts of the stretch and then reduces the data further by finding averages and standard deviations.
Movement sensorOnce that is completed, he can begin interpreting data by giving verbal suggestions or a number score based on the performance of the exercise.
Gaining research experience
“The REU is a major step toward gaining experience that counts for future endeavors. Even if you don’t end up in graduate school, it is good experience in your field and helps you figure out what you are interested in,” Kindle said.
He will present his research at the Summer Undergraduate Research Poster Symposium, which will be 10 a.m.-1 p.m. Friday, Aug. 2, in the CUE atrium. As he looks to his next year of school, Kindle plans to explore some artificial intelligence classes and continue looking for research or internship experience.