Download Project (504 KB)
I.ObjectivesFor the study of brain function and behavior, the mouse brain serves as a valuable model system. Methods to accurately analyze the complex signals produced by the mouse brain are becoming increasingly important due to developments in neuroimaging and optogenetics. Techniques that can take the information contained in mouse brain signals and turn it into useful biological insights are especially needed. In brain analysis, instance segmentation is commonly used for measuring and visualizing the brain's anatomical structures, for analyzing brain changes, for delineating pathological regions, and for surgical planning and image-guided interventions. In the last few decades, various segmentation techniques of different accuracy and degree of complexity have been developed and reported in the literature.II.MethodsThis interdisciplinary project explores the image segmentation with fluorescence microscopy images of mouse brain tissue. In particular, we will develop a model to segment dendrites, cell body, and axon from images of mouse brain tissue. The segmented results will be extremely helpful to detect synaptic proteins that are important for neuronal communication. We will focus on the excitatory synaptic proteins VGLUT1 and VGLUT2 which are specifically expressed circuits in the brain. To this end, we will use an image dataset of mouse brain tissue provided by Dr. Aaron Sathyanesan. Then, we will annotate the dendrites, cell body, and axonal regions by using our in-house annotation tool provided by Dr. Tam Nguyen. Next, we will train an image segmentation model on the newly annotated dataset. For the evaluation, we will use the performance metrics such as accuracy and IoU.
Course Project 202310 CPS 595 P2
Van Nguyen, Aaron Sathyanesan
Primary Advisor's Department
Stander Symposium, College of Arts and Sciences
Institutional Learning Goals
Community; Diversity; Scholarship
"Instance Segmentation to Identify Mouse Brain Cell Types" (2023). Stander Symposium Projects. 3049.