Course Projects

This is a list of some of the projects formulated, developed, and demonstrated by students in various courses taught by Dr. Hassan Ghasemzadeh.

Abnormal Gait and Fall Detection using Embedded Machine Learning

  • Author: Chia-Cheng Kuo

  • Course: Embedded Machine Learning

  • Semester: Spring 2022



It is crucial to provide emergency treatment for elderly or patients when they fall over. It is also important to provide warnings if the user has a high risk of falls due to abnormal gait. This project develops a real-time gait monitoring and fall detection system that integrates wearable inertial sensors and embedded machine learning while generating real-time feedback when falls are detected.

Real-Time Activity Recognition using Embedded Machine Learning

  • Authors: Emily Glagolev, Mihir Kotas, and Ahmed Musani

  • Course: Embedded Machine Learning

  • Semester: Spring 2022



Accurate estimation of activity types is central to human behavior modeling, activity-based interventions, and context-aware system design. We created an activity detection device using embedded machine learning using and Arduino Nano 33 BLE sense. This device can detect if a person is walking, jogging, using stairs or standing still, along with the duration of time they are doing the activity. Data was collected for each activity using the Arduino BLE connection and cleaned. A model was trained using Tensorflow. This trained model was used for real time testing with the Arduino BLE connected to a central. The central used bleak to detect the activity and record the duration of the activity.

Breathing Pattern Identification

  • Abdullah Mamun and Asiful Arefeen

  • Course: Embedded Machine Learning

  • Semester: Spring 2022



This project builds a system to detect abnormal breathing, coughs, and sneezes. The system differentiate abnormal breathing, coughs, and sneezes from normal breathing using Arduino and machine learning.

Low-Power Microgrid Disturbance Classification with Phasor Measurement Unit

  • Author: Zachary Lythgoe

  • Course: Embedded Machine Learning

  • Semester: Spring 2022



In comparison to the primary grid, microgrids are significantly more sensitive to local changes due to reduced local reserve generation and the switching of loads that make up a comparatively larger percentage of the total load. Given this sensitivity, it is critical for grid controllers to monitor local disruptions (i.e., sudden increase or decrease in load or generation) that the grid is experiencing. This work proposes a low-power phasor measurement unit (PMU) paired with an Arduino Nano BLE 33 Sense to classify a variety of disturbances on a simulated microgrid. The low-power low-cost solution makes such a system easier to implement in existing and future microgrids.

Embedded Machine Intelligence Lab
Embedded Machine Intelligence Lab
Research Lab

The current focus of our research in the Embedded Machine Intelligence Lab (EMIL) is on design, development, and validation of algorithms, tools, and technologies that enhance utilization and large-scale adoption of medical embedded systems. To validate and refine the new technology, we conduct clinical studies involving patients with heart failure, diabetes, cancer, visual impairment, and gait difficulties. Clinical studies are conducted in collaboration with partners from Elson S. Floyd College of Medicine, College of Nursing, College of Pharmacy, College of Education, and College of Agricultural, Human, and Natural Resource Science at WSU as well as our collaborators at Pullman Regional Hospital, UCLA School of Medicine, UCLA Stein Eye Institute, UC-Irvine Nursing Science, and Memorial Sloan Kettering Cancer Center (MSKCC). This end-to-end approach results in innovative, evidence-based and cost-conscious solutions for patients, doctors and medical centers.