Automated Detection of Candidate Subjects with Cerebral Microbleeds using Machine Learning

V Sundaresan, C Arthofer, G Zamboni, R A Dineen, P M Rothwell, S N Sotiropoulos, D P Auer, et al.

Frontiers in Neuroinformatics, vol.15, 2022 

Project Page / Paper / Code

Detected microbleed candidate subjects from a larger dataset, UK Biobank, using a machine learning-based, computationally light pipeline.

Triplanar ensemble U-Net model for white matter hyperintensities segmentation on MR images

V Sundaresan, G Zamboni, P M Rothwell, M Jenkinson, and L Griffanti

Medical Image Analysis, vol. 73, 102184, 2021

Project Page / Paper / Code

Proposed an ensemble triplanar network that uses anatomical information regarding WMH spatial distribution in loss functions to provide an accurate WMH segmentation.

White matter hyperintensities classified according to intensity and spatial location reveal specific associations with cognitive performance

L Melazzini, C E Mackay, V Bordin, S Suri, E Zsoldos, N Filippini, A Mahmood, V Sundaresan, M Codari, E Duff and A Singh-Manoux et al.

NeuroImage: Clinical, vol.30, 102616. 2021

Project Page / Paper 

Developed an automatic method that sub-classifies WMHs using MRI data from the Whitehall II study.

Automated lesion segmentation with BIANCA: impact of population-level features, classification algorithm and locally adaptive thresholding

V Sundaresan, G Zamboni, C Le Heron, P M Rothwell, M Husain, M Battaglini, N De Stefano, M Jenkinson, and L Griffanti

NeuroImage, vol. 202, 116056. 2019 

Project Page / Paper / Code

Improve our recently developed FSL tool for white matter hyperintensity segmentation, BIANCA, in order to better deal with the sources of lesion variability.

BIANCA (Brain Intensity AbNormality Classification Algorithm): a new tool for automated segmentation of white matter hyperintensities

L Griffanti, G Zamboni, A Khan, L Li, G Bonifacio, V Sundaresan, U G Schulz et al.

Neuroimage, vol. 141, p. 191-205. 2016

Project Page / Paper / Code

Proposed a fully automated, supervised method for white matter hyperintensity (WMH) detection, was integrated with FSL as the first WMH  segmentation tool.

Brain Tumour Segmentation Using a Triplanar Ensemble of U-Nets on MR Images

V Sundaresan, L Griffanti, M Jenkinson

In: Crimi A., Bakas S. (eds) Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. BrainLes 2020, MICCAI 2020. Lecture Notes in Computer Science, vol 12658. Springer Cham 2021

Project Page / Paper / Code