Stroke costs the United States an estimated $33 billion each year, reducing mobility in more than half of stroke survivors age 65 and over, or approximately 225,000 people a year. Stroke motor rehabilitation is a challenging process, both for patients and for their care providers. On the patient-side, occupational and physical therapy often involves hours of training patients to perform functional movements. This process can be mundane and tedious for patients. On the physician’s side, medical care providers need quantitative data and metrics on joint motion while stroke patients perform rehabilitative tasks.

To learn more about the stroke motor skill rehabilitation process and pertinent needs in the area, our team interviewed Dr. Kara Flavin, a physician and clinical assistant professor of Orthopaedic Surgery and Neurology & Neurological Sciences at the Stanford University School of Medicine, who helps stroke patients with recovery. We were inspired by her thoughts on the role of technology in the stroke recovery process to learn more about this area, and ultimately design our own technology to meet this need.

What it does

Our product, Kinesis, consists of an interconnected suite of three core technologies: an Arduino-based wearable device that measures the range of motion exhibited by the joints in users’ hands and transmits the data via Bluetooth; a corresponding virtual reality experience in which patients engage with a virtual environment with their hands, during which the wearable transmits range of motion data; and a Bluetooth to Android application to MongoDB data collection and visualization mechanism to store and provide this information to health care professionals.

How we built it

Our glove uses variable-resistance flex sensors to measure joint flexion of the fingers. We built circuits powered by the Arduino microcontroller to generate range-of-motion data from the sensor output, and transmit the information via a Bluetooth module to an external data collector (our Android application.) We sought a simple yet elegant design when mounting our hardware onto our glove. Next, we built an Android application to collect the data transmitted over Bluetooth by our wearable. The data is collected and sent to a remote server for storage using MongoDB and Node.js. Finally, we the data is saved in .csv files, which are convenient for processing and would allow for accessible and descriptive visuals for medical professionals. Our virtual-reality experience was built in the Unity engine and was constructed for Google Cardboard, deployable to any smartphone device that supports Google Cardboard.


Our project proved to be an exciting yet difficult journey. We quickly found that the various aspects of our project - the Arduino-based hardware, Android application, and VR with Unity/Google Cardboard - were a challenging application of internet of things (IoT) and hard to integrate. Sending data via Bluetooth from the hardware (Arduino) to the Android app and parsing that data to useful graphs was an example of how we had to combine different technologies. Another major challenge was that none of our team members had prior Unity/VR-development/Google Cardboard experience, so we had to learn these frameworks from scratch at the hackathon.

Accomplishments that we’re proud of

We hope that our product will make motor rehabilitation a more engaging and immersive process for stroke patients while also providing insightful data and analytics for physicians. We’re proud of learning new technologies to put our hack together, building a cost-effective end-to-end suite of technologies, and blending together software as well as hardware to make a product with incredible social impact.

What we learned

We had a very well-balanced team and were able to effectively utilize our diverse skill sets to create Kinesis. Through helping each other with coding and debugging, we familiarized ourselves with new ways of thinking. We also learned new technologies (VR/ Unity/Google Cardboard, MongoDB, Node.js), and at the same time learnt something new about the ones that we are familiar with (Android, hardware/Arduino). We learnt that the design process is crucial in bringing the right tools together to address social causes in an innovative way.

What's next for Kinesis

We would like to have patients use Kinesis and productize our work. As more patients use Kinesis, we hope to add more interactive virtual reality games and to use machine learning to derive better analytics from the increasing amount of data on motor rehabilitation patterns. We would also like to extend the applications of this integrated model to other body parts and health tracking issues.

Share this project: