The accurate perception of visual information once it is received in the visual cortex is crucial for successfully interacting with the world. It helps us navigate, identify objects and recognise people.
The Brain Imaging and Perception lab at the University of Wollongong is headed by Dr Mark Schira and spans the Schools of Psychology, Computer Science and Engineering.
We have a broad range of research projects under way in the areas of retinotopic mapping, natural image processing, psychophysics, susceptibility artefact correction, high resolution anatomical brain atlasing and automated segmentation of functional MR images.
This project aims to create a high resolution atlas of the human brain. We are using 7T MRI and multiple sequences to gather hundreds of images of the living human brain.
Given the vast diversity of natural scenes, which can range from a sparse desert to dense foliage, it is not immediately obvious what property the visual system may have become tuned to. However, in the last few decades, many statistical properties have been discovered across natural scenes which may have been exploited by the visual system throughout evolution.
One such property is the 1/f a amplitude spectrum which describes a distribution of luminance intensities across spatial and temporal scales (i.e. pixels close to each other in space and time are more likely to have similar intensity values compared to pixels further away from each other). This property is also considered to underlie the scale-invariant, fractal properties of natural scenes (i.e. the branching pattern of a tree is self-similar from the largest branch to the smallest twig).
Research has been conducted in an attempt to determine the extent in which the visual system is tuned to such properties. However, at present, most of this work has been conducted at the behavioural level using psychophysics, while little has been done at the cortical level.
For this project we use both psychophysics and fMRI to measure responses to highly controlled synthetic noise stimuli, which only differ in their 1/f a amplitude spectrum (manipulation of a) or their fractal dimension (fractal D). By measuring discrimination sensitivity and BOLD activity, we aim to better understand the degree in which behaviour matches activity in early visual cortex toward the 1/f a amplitude spectrum and its underlying fractal properties.
Venous contributions to the BOLD signal in fMRI are currently not completely understood.
This project uses fMRI to elucidate the retinotopic organisation of human V4 (hV4) after decades of experiment and debate. The prevalence of apparently 'incomplete' maps of hV4 has yet to be thoroughly explained however, it is likely that a variety of factors contribute to difficulty in accurately mapping this region, in particular artefact from nearby veins.
We use a combination of novel techniques to examine and correct retinotopic maps of hV4, including depth-dependent fMRI, susceptibility artefact correction and inverted voxel correction.
In this project we try to correct the susceptibility artefact in fMRI images, especially in high-resolution images in which the distortions are severe. We developed a susceptibility artefact method by using a T1w anatomy image and two fMRI images acquired using the identical sequences but with opposite polarities of the phase-encoding gradient.
The proposed method is called anatomy-guided inverse-gradient susceptibility artefact (AISAC). The proposed AISAC method was demonstrated that it is more robust, less blurred and run faster than the existing susceptibility artefact.
How is the visual system tuned to the movement of natural scenes? It has been robustly observed that visual sensitivity is optimal when still images share similar statistical properties to those that exist in natural scenes. What underlying features drive this sensitivity?
This project aims to demonstrate the sensitivity of the visual system to stimuli that approximate natural movement and understand which structural properties inherent in moving, natural images play a key role in visual perception.
Magnetic field inhomogeneities are common in the pre-frontal cortex due to differences in magnetic susceptibility between air-filled cavities and tissue/bone in the front of the human head.
Due to individual differences in sinus anatomy and relative position to the pre-frontal cortex, artefact correction must be performed on a subject-to-subject basis. This project aims to test the efficacy of the Anatomy-guided Inverse Gradient Susceptibility Artefact Correction method in pre-frontal cortex.
On November 7th and 8th, 2019, the Social Sciences Faculty Research Group MIBF presented the 3rd UOW MRI workshop. This workshop focussed on analysis methods for the hippocampus.
The practical components of the workshop were realised using the [email protected] workbench.
We were honoured to have presentations given by Tom Shaw from the University of Queensland, and Yann Chye from Monash University in Melbourne.
Thank you to our excellent presenters, organisers and attendees for making this workshop a success! We also thank the Faculty of Social Sciences Research Group Program at UOW for their support.
Links to the video recordings of these presentations as well as power point slides and other resources are included below.
Credit to Dr Steffen Bollmann and Thomas Shaw for these slides.
Magnetic Resonance Imaging. 10.1016/j.mri.2020.04.004., , Bouzerdoum, A., & (2020). An unsupervised deep learning technique for susceptibility artifact correction in reversed phase-encoding EPI images. Magnetic Resonance Imaging.
Journal of Neuroscience Methods. 336. 10.1016/j.jneumeth.2020.108625., , Bouzerdoum, A. , , Puckett, A., & (2020). Susceptibility Artifact Correction for Sub-millimeter fMRI using Inverse Phase Encoding Registration and T1 Weighted Regularization.
Oliveira, ML., Pang, JC., Robinson, PA., Liu, X., & PLoS Comp. Biol. 15. e1007418. 10.1371/journal.pcbi.1007418. (2019). Feasibility of Funtional Magnetic Resonance Imaging of Ocular Dominance and Orientation Preference in Primary Visual Cortex.
Mancini, F., Wang, AP., J. of Neurosci. 10.1523/JNEUROSCI.2005-18.2019, , McAuley, JH., Iannetti, GD., Sereno, MI., Mosley, GL., & Rae, CD. (2019). Fine-grained mapping of cortical somatotopies in chronic Complex Regional Pain Syndrome.
Viengkham, C., Axiomathes. 10.1007/s10516-019-09444-z, & Spehar, B. (2019). Fractal-Scaling Properties as Aesthetic Primitives in Vision and Touch.
PLoS ONE 14(6): e0204388. 10.1371/journal.pone.0204388, , , & (2019). Vascular effects on the BOLD response and the retinotopic mapping of hV4.
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP): Proceedings., , Phung, L., Bouzerdoum, A., & (2018). Anatomy-guided inverse-gradient susceptibility artifact correction method for high-resolution fMRI.
Pang, J., Journal of Neuoscience Methods, 308. 10.1016/j.jneumeth.2018.07.009, Robinson, P., Lacy, T., & (2018). Biophysically based method to deconvolve spatiotemporal neurovascular signals from fMRI data.
NeuroImage, 146. 10.1016/j.neuroimage.2016.10.013, , & Spehar, B. (2017). The tuning of human visual cortex to variations in the 1/f α amplitude spectra and fractal properties of synthetic noise images.
NeuroImage, 139. 10.1016/j.neuroimage.2016.06.019, , Robinson, P., Breakspear, M., & (2016). The spatiotemporal hemodynamic response function for depth-dependent functional imaging of human cortex.
Current Biology, 22 (24). 10.1016/j.cub.2012.11.003, Tyler, C., & Rosa, M. (2012). The (Un)folding of striate cortex.
Journal of Neuroscience, 29 (28). 10.1523/JNEUROSCI.1760-09.2009, Tyler, C., Breakspear, M., & Spehar, B. (2009). The foveal confluence in human visual cortex.
Journal of Neurophysiology, 97. 10.1152/jn.00972.2006, Wade, A., Tyler, C. (2007). Two dimensional mapping of the central and parafoveal visual field to human visual cortex.