Presenter: Maryam Shuaib – Human Physiology
Faculty Mentor(s): Mike McGeehan, Keat Ghee Ong
Session: (In-Person) Oral Panel—Stimuli and Response, Poster Presentation
There is an increasing need to measure shear force in biomedical applications. Many shear force sensors exist, but are often impractical as they can be bulky, require large amounts of power, and are sensitive to electromagnetic interference. The goal of this project is to apply new optoelectronic sensing principles to measure shear strain. Optoelectronic sensors have various advantages including a smaller design that is able to measure multi-axial shear strain. This particular sensor functions through optical coupling of an LED that emits red, green, and blue (RGB) light, which is then reflected off of an adjacent surface displaying a color pattern consisting of randomized color pixels (Figure 1A). Shearing between these surfaces is measured using a photodiode, which senses changes in the RGB light intensities due to the shifts in the color pattern’s position. The purpose of this study was to compare the efficacy of various color patterns and classification algorithms to determine multi-axial shear strain. The optimal sensor configuration was found to be Pattern 3 (Figure 1B) with a Weighted K-Nearest-Neighbor algorithm with an accuracy of 98%, and a misclassification cost of 0.07 millimeters. The accuracy and robustness of the sensor-derived measurements, along with the practical and scalable design, support the use of this sensor in a multitude of biomedical applications.