Browsing by Author "Efird, Cory"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
Item Creative, interdisciplinary undergraduate research: an educational cell biology video game designed by students for students(2020) Sperano, Isabelle; Shaw, Ross; Andruchow, Robert; Cobzas, Dana; Efird, Cory; Brookwell, Brian; Deng, WilliamIn a three-year, practice-based, creative research project, the team designed a video game for undergraduate biology students that aimed to find the right balance between educational content and entertainment. The project involved 7 faculty members and 14 undergraduate students from biological science, design, computer science, and music. This nontraditional approach to research was attractive to students. Working on an interdisciplinary practice-based research project required strategies related to timeline, recruitment, funding, team management, and mentoring. Although this project was time-consuming and full of challenges, it created meaningful learning experiences not only for students but also for faculty members.Item Hippocampus segmentation on high resolution dffusion MRI(2021) Efird, Cory; Neumann, Samuel; Solar, Kevin G.; Beaulieu, Christian; Cobzas, DanaWe introduce the first hippocampus segmentation method for a novel high resolution (1×1×1mm3) diffusion tensor imaging (DTI) protocol acquired in 5.5 minutes at 3T. A new augmentation technique uses subsets of the DTI dataset to create mean diffusion weighted images (DWI) with plausible noise and contrast variations. The augmented DWI along with fractional anisotropy (FA) and mean diffusivity (MD) maps are used as inputs to a powerful convolutional neural network architecture. The method is evaluated for robustness using a second diffusion protocol.Item A UNet pipeline for segmentation of new MS lesions(2021) Efird, Cory; Miller, Dylan; Cobzas, DanaA pipeline for the second multiple sclerosis segmentation challenge (MSSEG-2) hosted by MICCAI is proposed. Two FLAIR images taken at different time-points are used as a multi-channel input to a 3D CNN to detect new lesions. Patch sampling strategies are adopted to keep the input volume shape manageable in terms of memory requirements. To further improve results, multiple models and patch orientations are ensembled. Performance is evaluated against nn-UNet.