Abstract:
Understanding brain functions requires digital reconstruction or tracing of 3D neurons. Existing automatic tracing algorithms work well for a clean neuronal image with a single neuron but not for a neuron surrounded by nerve fibers.
We propose 3D U-Net Plus to segment the neuron from surrounding fibers before applying tracing algorithms. BigNeuron, the largest neuronal image dataset, contains only clean neurons without nerve fibers, making it impractical to train the segmentation network.
By fusing BigNeuron images with extracted nerve fibers, we synthesize a SYNethic TAngled NEuronal Image dataset (SYNTANEI) to train the proposed network.
The proposed 3D U-Net Plus network segmented synthetic and real tangled neuronal images using dropout, trous convolution, and trous Spatial Pyramid Pooling (ASPP). The tracing algorithm’s segmentation result neurons match the ground truth better than the original images.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
