Segmentation and Classification of Object Usage in Smart Environments Using MetaWear Sensors - 2024
Research project in Gerontechnology funded by NIH, where I had the opportunity to work with Dr. Diane Cook, a Regents Professor and Huie-Rogers Chair Professor in the School of Electrical Engineering and Computer Science (EECS) at Washington State University. The long-term goal of this eight-week project was to monitor routine activities for older adults using item sensors placed on everyday objects such as toothbrushes, pill organizers, and coffee pots. We hypothesized that machine learning techniques could effectively monitor when these objects were being used for their intended functional purposes. Our models achieved an overall 95% accuracy in k-fold cross-validation, a success attributed to the quality of the feature extraction and class labeling processes we employed.