Hello ! My name is Simon and I am a PhD student at King’s College London and member of the MeTriCs Lab. My PhD project is supervised by Dr.Emma C. Robinson and Pr. Daniel Rueckert.

My research work lies in developing new deep learning methods for studying neuroimaging data, especially for MRI and fMRI, with a particular interest in studying neonatal cortical development and neurological disorders.

Currently, I am developing deep learning methodologies for studying brain surface features using vision transformers.

Updates:

  • Our latest work on Surface Vision Transformers is currently under review for the MIDL 2022 conference (link)

Publications

You can also visit my Google Scholar

2023

  • Surface Masked AutoEncoder: Self-Supervision for Cortical Imaging Data, Simon Dahan, Daniel Rueckert, Emma Claire Robinson.

    Code

  • The Multiscale Surface Vision Transformer, Simon Dahan, Abdulah Fawaz, Mohamed A Suliman, Mariana Da Silva, Logan Zane John Williams, Daniel Rueckert, Emma Claire Robinson.

    Arxiv Code

  • Geneneralising the HCP multimodal cortical parcellation to UK Biobank, Logan Zane John Williams, Matthew F Glasser, Fidel Alfaro-Almagro, Simon Dahan , Abdulah Fawaz, Timothy S Coalson, Sean Patrick Fitzgibbon, Mohamed Suliman, David C Van Essen, Stephen M Smith, A David Edwards, Emma Claire Robinson.

2022

  • Surface Vision Transformers: Attention-Based Modelling applied to Cortical Analysis, Simon Dahan, Abdulah Fawaz, Logan Zane John Williams, Chunhui Yang, Timothy S. Coalson, Matthew Glasser, A David Edwards, Daniel Rueckert, Emma Claire Robinson.

    Accepted for Oral Presentation at MIDL 2022

    Publication Arxiv Code

  • Surface Analysis with Vision Transformers, Simon Dahan, Logan Zane John Williams, Abdulah Fawaz, Daniel Rueckert, Emma Claire Robinson.

    Arxiv Code

2021

  • Benchmarking Geometric Deep Learning for Cortical Segmentation and Neurodevelopmental Phenotype Prediction, Abdulah Fawaz, Logan ZJ Williams, Amir Alansary, Cher Bass, Karthik Gopinath, Mariana da Silva, Simon Dahan, Chris Adamson, Bonnie Alexander, Deanne Thompson, Gareth Ball, Christian Desrosiers, Hervé Lombaert, Daniel Rueckert, A David Edwards, Emma C Robinson.
    Under review.
    Biorxiv Code

  • Improving Phenotype Prediction using Long-Range Spatio-Temporal Dynamics of Functional Connectivity, Simon Dahan, Logan ZJ Williams, Daniel Rueckert, Emma C Robinson.
    Published at the International Workshop on Machine Learning in Clinical Neuroimaging (MLCN 2021)
    Publication Arxiv Code

2019

  • Unravelling Machine Learning–Insights in Respiratory Medicine, Elsa Angelini, Simon Dahan, Anand Shah.
    Published in the European Respiratory Journal
    Publication

Background

A little bit more details about my educational background:

Graduate Logo I graduated from Telecom Paris in December 2019 and obtained an Engineer’s degree (MEng) specialising in Data Science and Image Processing. At Telecom Paris, I was particularly interested in applying machine learning and deep learning models, especially convolution neural networks to imaging datasets, comparing performances with image processing techniques. I especially enjoyed image processing and machine learning modules lead by Isabelle Bloch (medical imaging), Florence Tupin (satellite imaging) and Pietro Gori (medical imaging).

Graduate Logo In 2018-2019, I did an exchange in London to follow a MSc in Computing (specialism Machine Learning) at Imperial College London. This was an extraordinary programme with world-leading AI researcher such as Dr. Michael Bronstein or Dr. Stefanos Zafeiriou (Deep Learning module), Dr. Ben Glocker and Pr. Daniel Rueckert (Machine Learning for medical imaging module) or Pr. Marc Deisenroth (Probabilistic Inference module).

Graduate Logo In 2020, I was selected to join the Smart Medical Imaging CDT for a 4 years PhD programme. Note:

Contact

Please, feel free to contact me at simon.dahan@kcl.ac.uk