Development of Data-driven Prediction Model using 3D Multimodal Deep Neural Networks for Estimating the Evolution of White Matter Hyperintensities Associated with Small Vessel Disease in Brain MRI

Collaborating institutions: RIKEN Center for Brain Science (Japan), School of Informatics at The University of Edinburgh

About the project

Project title: "Development of Data-driven Prediction Model using 3D Multimodal Deep Neural Networks for Estimating the Evolution of White Matter Hyperintensities Associated with Small Vessel Disease in Brain MRI".

Prognosis of vascular disease is a need, and this collaboration is primarily focused on developing data-driven solutions to build predictive models for the progression of small vessel disease (SVD), a vascular disease that underpins ageing and dementia progression, using MRI scans as the main diagnostic tool. Specifically, the aim of this project is to develop algorithms for predicting the evolution of white matter hyperintensities, which are the key feature of this progressive disease.

Collaborating scientists

RIKEN Center for Brain Science (Japan)

The University of Edinburgh (UK) 

Dr Muhammad Febrian Rachmadi

Special Postdoctoral Researcher

Email Address: febrian.rachmadi@riken.jp

 Dr Maria Valdes-Hernandez

Row Fogo Lecturer in Medical Image Analysis
Affiliated departments:
  • Row Fogo Centre for Research into Ageing and the Brain
  • Edinburgh Imaging
  • Centre for Clinical Brain Sciences

Email address: M.Valdes-Hernan@ed.ac.uk

 

Professor Taku Komura

Professor of Computer Graphics
Affiliated departments:
  • Institute of Perception, Action and Behaviour
  • School of Informatics

Email address: TKomura@inf.ed.ac.uk

 

Awarded grants

This project is funded by the Grants-in-Aid for Scientific Research  (KAKENHI) Program (project no 20K23356) - awarded 2,861,000 ¥ Jan 2021- Dec 2022.

 

Publications

Publication title: "Probabilistic Deep Learning with Adversarial Training and Volume Interval Estimation - Better Ways to Perform and Evaluate Predictive Models for White Matter Hyperintensities"

Authors: Muhammad Febrian Rachmadi, Maria del C. Valdes-Hernandez, Rizal Maulana, Joanna Wardlaw, Stephen Makin, and Henrik Skibbe

Publication date: September 2021

Publication link on Springer (external website)

 

Recognitions and awards

Above publication received the Best Publication Award at 24th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2021).

Read more about the event

MICCAI website (external link)

 

Developed algorithms

IMPORTANT NOTE: Developed by us algorithms are publically available, free of charge. You are free to use developed by us tools for research purposes. Hovewer, please include adequate citations and acknowledgments whenever you present or publish results that are based on it.

Algorithm title: "Probabilistic Deep Learning with Adversarial Training and Volume Interval Estimation - Better Ways to Perform and Evaluate Predictive Models for White Matter Hyperintensities Evolution"

Authors: Muhammad Febrian Rachmadi

Topics: deep neural networks, deep learning, brain MRI, WMH segmentation, WMHs

Algorithm link

GitHub link (external website)

 

 

Dr Maria Valdes Hernandez

Key contact

Please, get in touch with Dr Maria Valdes-Hernandez for more information about this project and further collaboration.

 M.Valdes-Hernan@ed.ac.uk

Dr Maria Valdes-Hernandez research profile

This article was published on