Nicola K Dinsdale
I am currently working as a post-doctoral research associate in the Oxford
Machine Learning in NeuroImaging Lab (OMNI), working with Dr. Ana
Namburete, in the Department of Computer Science.
I studied for my DPhil (PhD) in the Analysis Group at the Wellcome Centre for Integrative
Neuroimaging at the University of Oxford, where I researched deep
learning based approaches for neuroimaging analysis, supervised by Prof. Mark
Jenkinson and Dr. Ana
Namburete, funded by the UKRI EPRSC/MRC as part of the ONBI DTC.
My research uses computer vision and deep learning to solve medical imaging
problems. I am especially interested in exploring methods to overcome the
barriers to clinical translatability of deep learning methods and robust deep
learning, and I am open to collabortion opportunities.
Email  / 
Google
Scholar  / 
Github
|
|
|
UniFed: A unified deep learning framework for
segmentation of partially labelled, distributed
neuroimaging data
Nicola K Dinsdale, Mark Jenkinson, Ana IL Namburete
bioRxiv, 2024
Project Page / Paper / Code
We, therefore, propose UniFed, a unified federated harmoni-
sation framework, which enables three key processes to be completed: 1) the training of a federated partially labelled
harmonisation network, 2) the selection of the most appropriate pretrained model for a new unseen
site, and 3) the incorporation of a new site into the harmonised federation.
|
|
SFHarmony: Source Free Domain Adaptation for Distributed
Neuroimaging Analysis
Nicola K Dinsdale, Mark Jenkinson, Ana IL Namburete
ICCV, 2023
Project Page / Paper / Code
We propose an Unsupervised Source-Free Domain Adaptation (SFDA) method,
SFHarmony, and demonstrate the approach for classification, regression and
segmentation tasks.
|
|
Challenges for machine learning in clinical translation of big data imaging studies
Nicola K Dinsdale, Emma Bluemke, Vaanathi Sundaresan, Mark Jenkinson, Steve Smith, Ana IL Namburete
Neuron , 2022
Paper / Code
Review article, explores: data availability, interpretability, model bias and data privacy.
|
|
Prototype Learning for Explainable Regression
Linde S Hesse, Nicola K Dinsdale, Ana IL Namburete
IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2023
Paper / Code
In this work we present ExPeRT: an explainable prototype-based model specifically designed for regression tasks.
|
|