Researchers are using sonomicrometry to study the biomechanics of the human brain. While at times the signals collected do not require any preprocessing, more frequently they do require some denoising or are too noisy to analyze. Currently, researchers are manually categorizing the quality of thousands of these sonomicrometry signals and preprocessing them individually. RC is helping researchers develop a machine learning model to classify the signals and to determine the necessary preprocessing steps.
Preliminary sonomicrometry data have been collected, and RC is working to classify, prepare, and normalize the data for use in a machine learning model. RC is currently developing preliminary models to classify the data by signal quality and preprocess automation techniques that will later be applied to noisy signals.
PI: Matthew Panzer (Center for Applied Biomechanics)