Our projects
Our research initiatives span neuroimaging and brain analysis, including topology-preserving neural networks, unsupervised segmentation techniques, multimodal MRI connectivity mapping, and deep learning approaches for detecting cerebral microbleeds and brain lesions across various imaging modalities.
Topology Preserving Reservoirs (TPR)
Improving zero-shot image recognition for accurate classification of both seen and unseen objects
Unsupervised dense prediction via Normalized Cuts
Enhancing self-supervised computer vision with Differentiable Normalized Cuts
Multimodal brain MRI analysis
Analysing multimodal MRI to reveal how subcortical stroke changes brain connectivity tied to executive function
Deep learning for brain lesion detection and segmentation
Using 3D deep learning to detect and segment cerebral microbleeds in MRI scans
KōreroNET: listening to nature
Developing autonomous acoustic sensor networks to monitor New Zealand’s native wildlife and pests
Extreme weather modelling with satellite imagery
Applying deep learning and remote sensing to detect, forecast and analyse extreme weather events such as cyclones and floods
Plant decision support via spectral imaging
Developing an end-to-end workflow for early virus detection in plants using hyperspectral imaging and advanced AI