Tracking the visual system—from the optic chiasm to primary visual cortex

Abstract

Epilepsy surgery is a well-established method of treatment for pharmacoresistant focal epilepsies, but it carries an inherent risk of damaging eloquent brain structures. This holds true in particular for visual system pathways, where the damage to, for example, the optic radiation may result in postoperative visual field defects. Such risk can be minimized by the identification and localization of visual pathways using diffusion magnetic resonance imaging (dMRI). The aim of this article is to provide an overview of the step-by-step process of reconstructing the visual pathways applying dMRI analysis. This includes data acquisition, preprocessing, identification of key structures of the visual system necessary for reconstruction, as well as diffusion modeling and the ultimate reconstruction of neural pathways. As a result, the reader will become familiar both with the ideas and challenges of imaging the visual system using dMRI and their relevance for planning the intervention.

Deep Learning-Based Detection of Malformed Optic Chiasms From MRI Images

“…Effectively, it is unknown which anatomical features of the chiasm may be employed in an individualized detection of malformations or whether such a detection is possible in the first place. Recently, the application of anatomy-sensitive diffusion MRI (dMRI), capable of estimating the proportion of crossing and non-crossing nerves via tractography ( Puzniak et al, 2021 ), has demonstrated chiasmal malformations in albinism at the group level ( Ather et al, 2018 ) with potential for an individualized diagnostic utility ( Puzniak et al, 2019 ). It must be noted, however, that dMRI as compared to aMRI is time consuming at the level of both data acquisition and data analysis.…”

Correlation between retinal structure and brain multimodal magnetic resonance imaging in patients with Alzheimer’s disease

“…As suggested by Puzniak et al (2021), to perform tractography of the visual system, the individual structures of optic chiasm, LGN, and V1 need to be provided at least. In our case, the locations of these structures were estimated automatedly as outlined above.…”