Visual Pathway
The visual pathway, responsible for processing visual information from the retina to the brain, is a complex system currently under intense investigation. Research focuses on understanding its two main branches (ventral and dorsal streams) using deep learning models, including convolutional neural networks (CNNs), vision transformers (ViTs), and spiking neural networks (SNNs), to improve image and video representation learning and object-centric representation. These computational models are crucial for developing advanced neuroprostheses and improving the accuracy of medical image analysis techniques like retinogeniculate pathway segmentation from MRI data, ultimately aiding in the diagnosis and treatment of visual disorders. Furthermore, research is exploring how these models can reveal the functional organization and information processing mechanisms within the visual pathway across different species.
Papers
Modality Exchange Network for Retinogeniculate Visual Pathway Segmentation
Hua Han, Cheng Li, Lei Xie, Yuanjing Feng, Alou Diakite, Shanshan Wang
LESEN: Label-Efficient deep learning for Multi-parametric MRI-based Visual Pathway Segmentation
Alou Diakite, Cheng Li, Lei Xie, Yuanjing Feng, Hua Han, Shanshan Wang