Salil Patel

I am a physician-scientist, currently a DPhil candidate at the University of Oxford and NIHR academic clinical fellow (ACF) in radiology at Guy's and St Thomas' hospitals, London.  

My research focuses on understanding and diagnosing neurological conditions using computational methods, with a specific interest in geometric deep learning and time series analysis. I investigate neural mechanisms underlying both healthy and diseased states, aiming to uncover new insights into these complex disorders.

I have a particular interest in neurodegenerative conditions - especially Parkinson's disease and other Parkinsonian disorders. These diseases are heterogenous, with a multitude of clinical phenotypes.

By using novel computational methods, we probe such phenotypes. This has the potential to aid diagnosis, monitoring and the discovery of novel treatments – both symptomatic and disease-modifying in nature.

Recent Preprints

2D and 3D Deep Learning Models for MRI-based Parkinson's Disease Classification: A Comparative Analysis of Convolutional Kolmogorov-Arnold Networks, Convolutional Neural Networks, and Graph Convolutional Networks (July 2024)
SB Patel, V Goh, JJ FitzGerald, CA. Antoniades

Hierarchical Machine Learning Classification of Parkinsonian Disorders using Saccadic Eye Movements: A Development and Validation Study (July 2024)
SB Patel, OB Bredemeyer, JJ FitzGerald, CA Antoniades

Recent Presentations

Computational Phenotyping of Parkinson’s Disease (March 2023), AD/PD, Sweden

1st Prize, Natural Language Processing for Neuroimaging Report Classification (May 2023), Royal Society of Medicine, London

Classification of Alzheimer's Disease Severity from MRI Scans: Vision Transformer versus Convolutional Neural Network Approach (November 2023), RSNA, Chicago

Multimodal Medical Image Classification: A Comparative Study of Vision Transformers and Convolutional Neural Networks using MRI, CT and Ultrasound 2D slices (November 2023), RSNA, Chicago

Hierarchical Machine Learning Classification of Parkinsonian Disorders using Saccades (September 2024) MDS, Philadelphia

Linear Mixed-Effect Modelling of Antisaccades in Parkinson's Disease and Progressive Supranuclear Palsy, (September 2024) MDS, Philadelphia

Upcoming (Late 2024):

Graph Neural Network-based Classification of Parkinson's Disease using MRI Brain Scans, RSNA, Chicago

Geometric Deep Learning for Brain Tumor Classification in MRI: A Graph Convolutional Network Approach, RSNA, Chicago

Selected Papers

Patel, S.B, Bredemeyer, O., Williamson, D.J., Struyven, R.R., Zhou, Y., Denniston, A.K., Petzold, A., Antoniades, C.A., Keane, P.A., and Wagner, S.K. (2024). Retinal morphological differences in atypical Parkinsonism: A cross-sectional analysis of the AlzEye cohort. Biomarkers in Neuropsychiatry

Su ZH, Patel S, Gavine B, Buchanan T, Bogdanovic M, Sarangmat N, et al (2023). Deep Brain Stimulation and Levodopa Affect Gait Variability in Parkinson Disease Differently. Neuromodulation. 

Bredemeyer, O, Patel, S.B, Fitzgerald, J and Antoniades, C (2022) Oculomotor deficits in Parkinson's disease: Increasing sensitivity using multivariate approaches. Front. Digit. Health.

Su ZH, Patel S, Bredemeyer O, FitzGerald JJ, Antoniades CA (2022). Parkinson's disease deficits in time perception to auditory as well as visual stimuli – A large online study. Front Neurosci . 

Patel, S.B, Fitzgerald, J and Antoniades, C (2019) Oculomotor effects of medical and surgical treatments of Parkinson's disease. A Progress in Brain Research. 

Patel, S.B, Brzezicki, M.A, Fitzgerald, J and Antoniades, C (2019) The influence of deep brain stimulation on eye movements. Advances in Translational Neuroscience of Eye Movement Disorders. 

Patel, S.B (2019) Should the National Health Service be independent like the judiciary?  Journal of the Royal Society of Medicine.

Patel, S.B. (2017) Can the National Health Service learn from the rise and fall of ancient empires? Journal of the Royal Society of Medicine.

Patel, S.B and Gutowski, N. (2015) The difficulty in diagnosing X linked adrenoleucodystrophy and the importance of identifying cerebral involvement. BMJ Case Reports