Engineering Commons^™

Blood Glucose Predictor, Jessica Patterson

Electrical Engineering

For my senior project, I perform data analysis using statistical methods to determine body metrics that correlate with blood glucose levels. Working with Dr. Tina Smilkstein, I take repeat measurements from 6 different volunteers to establish trends in bodily metric data. The data taken includes weight, body fat, pulse rate, VO2, blood glucose, blood pressure, hours slept, and quality of sleep. Using these values, I use the program MiniTab to view results.

A few examples of correlations with blood glucose found in this project are:

• Systolic blood pressure for females had a regression line of 124.0 -0.3366*Blood Pressure. This indicates …

Eeg Classification Using Residual Neural Networks, Eren Esener

Electrical Engineering

There is a growing desire to understand the EEG (electroencephalogram) signals related with brain activities. In order to analyze EEG signals, they first must be measured by sensors, which induces a lot of noise; then these signals are classified to understand the intended actions. In many cases a neural network is used as the algorithms for classification. A residual neural network (ResNet) is an artificial neural network (ANN) of a kind that are related with the pyramidal cells in the cerebral cortex. The goal of this work is to investigate the viability of ResNets for classifying EEG signals for hand …

Portable Electrocardiogram Device And Signal Processing Design, Ryan F. Blaalid

Electrical Engineering

Full 12-lead electrocardiogram (ECG) measurements require inconvenient and time consuming adhesive electrode placement. This project proposes a design for a Bluetooth based ECG for remote patient measurement. The device is designed to measure up to 6-leads and utilizes 5 dry (non-adhesive) electrodes to accomplish this. The device delivers the ECG data to the user’s mobile smart phone and can then be sent to the patient’s doctor for analysis. Since the contact of the dry electrodes to the skin is not perfect, low-frequency noise called baseline wandering is introduced. A signal processing technique borrowed from image processing called morphological filtering is …