- 07.10.2020: The workshop proceedings are now available for download " here! "
- 15.08.2020: Best paper award will be a Titan RTX GPU, sponsored by NVIDIA!
- 27.06.2020: The submission deadline has been extended by a week to July 7th! " Submission link! "
- 16.06.2020: An interesting talk by " Jonny Hancox " on NVIDIA's cuGraph library titled ‘Accelerating Graph Analytics with GPUs’ will be part of GRAIL2020! Stay tuned for more interesting updates!
- 09.05.2020: GRAIL2020 will be held ONLINE along with MICCAI2020! Workshop registration fees are reduced by about half!
- 06.05.2020: Program Committee members are now online!
- 05.05.2020: CMT System is now open for submissions of both abstract and complete papers. "Click here to submit your manuscript!".
- 13.05.2020: Our second Keynote Speaker will be " Prof. Hervé Lombaert ", with his keynote titled "Geometric Data Analysis in Medical Imaging".
- 27.04.2020: Our first Keynote Speaker is confirmed to be " Dr. Ahmad Ahmadi ", who will be giving a talk on "Graph Convolutional Networks in Disease Prediction".
- 30.03.2020: This year, we will accept papers in two modalities: Complete papers (8-12 pages, including references) and Abstract papers (2-3 pages). Read more about it in the section "Submit a paper".
- 30.03.2020: The GRAIL Website is up and running!
GRAIL 2020 is the third international workshop on GRaphs in biomedicAl Image anaLysis, organised as a satellite event of MICCAI 2020 in Lima, Peru.
Graph-based models have been developed for a wide variety of problems in computer vision and biomedical image analysis. Applications ranging from segmentation, registration, classification, and shape modelling, to population analysis have been successfully encoded through graph structures, demonstrating the versatile and principled nature of the graph based approaches.
Graphs are powerful mathematical structures which provide a flexible and scalable framework to model objects and their interactions in a readily interpretable fashion. As a consequence, an important body of work has been developed around different methodological aspects of graph including, but not limited to, graphical models, graph-theoretical algorithms, spectral graph analysis, graph dimensionality reduction, and graph-based network analysis. However, new topics are also emerging as the outcome of interdisciplinary studies, shedding light on areas like deep structured models and signal processing on graphs.
With this workshop we aim to highlight the potential of using graph-based models for biomedical image analysis. Our goal is to bring together scientists that use and develop graph-based models for the analysis of biomedical images and encourage the exploration of graph-based models for difficult clinical problems within a variety of biomedical imaging contexts.
The covered topics include but are not limited to:
- Deep/machine learning on graphs with regular and irregular structures
- Probabilistic graphical models for biomedical image analysis
- Discrete and continuous optimization for graphical models
- Signal processing on graphs for biomedical image analysis
- Deep/machine learning on structured and unstructured graphs
- Convolutional neural networks on graphs
- Graphs for large scale population analysis
- Graph-based shape modeling and dimensionality reduction
- Combining imaging and non-imaging data through graph structures
- Graph-based generative models for biomedical image analysis
- Graph spectral methods
- Algorithms on graphs
- Graphs in neuroimaging
- Applications of graph-based models and algorithms to biomedical image analysis tasks (segmentation, registration, classification, etc.)
|09:00 – 09:05||Welcome and opening remarks|
|09:05 – 09:40||Keynote Speaker 1: Xavier Bresson "Data Science, Graph Deep Learning, Spectral Graph Theory"|
|09:40 – 10:50||Session 1:
|10:50 – 11:25||Keynote 2: Seyed-Ahmad Ahmadi "Graph Convolutional Networks in Disease Prediction"|
|11:25 – 12:40||Session 2:
|12:40 – 13:20||Keynote 3: Jonny Hancox "Tutorial on NVIDIA's cuGraph"|
|13:20 – 13:30||Best paper award and closing remarks|
Dr. Ahmad Ahmadi
LMU TUM, Munich, Germany
Machine learning, deep learning, clinical translation: FCNNs, RNNs, auto-encoders, GANs, transfer learning
Seyed-Ahmad Ahmadi is currently the group leader for medical AI at the German Center for Vertigo and Balance Disorders (DSGZ, Ludwig-Maximilians-University (LMU), Munich, Germany), and a senior post-doc affiliate at the chair for Computer Aided Medical Procedures (Prof. Dr. Nassir Navab, Technical University of Munich TUM, Germany). Prior to that, he pursued a three-year post-doc fellowship at the department of neurology (LMU, Munich, Germany). In 2013, he finished his PhD degree with distinction under the supervision of Prof. Dr. Nassir Navab at TUM. His main research interest lies in the translation of methods from data science and machine learning (DS/ML) to clinical data in the wild, in close collaboration with medical doctors. During his 15-year track-record on applied DS/ML in the clinic, his projects involved surgical workflow modeling, medical image analysis, spatio-temporal sensor data processing and finally, a multi-modal fusion of complementary sensor data towards computer-aided decision support systems (CDSS) and computer-aided diagnosis and prediction (CADx). Since 2017, a particular focus lies on the application of geometric deep learning and graph neural networks towards CDSS/CADx. He has led and/or contributed to several successful national (DFG, BMBF) and international (EU FP7, HORIZON 2020) grant proposals. He has co-authored over 50 conference and journal papers, including the widely known paper on the V-Net architecture and Dice loss for fully convolutional (3D) image segmentation (>1800 citations). Beyond academia, he is a freelancer consultant to local med-tech startups and companies with a focus on applied DS/ML.
Prof. Xavier Bresson
Data Science, Graph Deep Learning, Sparse Convex Optimization, Spectral Graph Theory
Xavier Bresson is associate professor in Computer Science at NTU, Singapore. His research focuses on graph deep learning, a new framework that combines graph theory and deep learning techniques to tackle complex data domains. In 2016, he received the Singaporean NRF Fellowship of USD 2.5M to develop these new techniques. He was also awarded several research grants in the U.S. and Hong Kong. He co-authored one of the most cited works in this field (shorturl.at/ijS39) and he has recently introduced with Yoshua Bengio a benchmark (shorturl.at/dLO46) to evaluate the graph neural network architectures. He has organized several workshops and tutorials on graph deep learning such as the IPAM 2021 workshop on "Deep Learning and Combinatorial Optimization" (shorturl.at/AEL04), the IPAM 2019 workshop on "Deep Geometric Learning of Big Data" (shorturl.at/deKVW), the IPAM 2018 workshop on "New Deep Learning Techniques" (shorturl.at/jFUZ3), and the NeurIPS 2017 tutorial on "Geometric deep learning on graphs and manifolds" (shorturl.at/dmLPX). He is regular invited speaker from universities and companies to present this research topic. He was speaker at ICML 2020 workshop on "Graph Representation Learning and Beyond" (shorturl.at/dfBU4) and ICLR 2020 workshop on "Deep Neural Models and Differential Equations" (shorturl.at/esOVZ). He has been teaching graduate courses on graph deep learning at NTU and as guest lecturer at NYU for Yann LeCun's course (shorturl.at/FTZ57).
Mr. Jonny Hancox
Accelerating Graph Analytics with GPUs
Abstract: Nvidia started off in gaming but branched out into High Performance Computing and Deep Learning. Recently it has taken a similar change of tack to enter address some of short comings the Machine Learning community. Jonny will talk about some of the critical tools in the data scientist’s kit bag that RAPIDS provides. Jonny will show many of the components of RAPIDS can be leveraged to create graph-based medical imaging workflows. Bio Jonny is a Senior Data Scientist in the Healthcare team at Nvidia. He joined Nvidia from Intel in 2018 where he held a similar position. Prior to that Jonny was CTO for a software development company. Most of Jonny’s work is related to AI and medical imaging and, in particular, computational pathology. Jonny originally trained as a Product Designer before a foray into music and eventually finding out that computers were actually quite cool.
Submissions should be done through the GRAIL CMT portal
This year we will accept papers in 2 formats:
- Complete papers: Papers describing original research with length between 10 and 12 pages, including references. These papers are eligible for the "Best Paper Award" and will be published in the GRAIL Open Proceedings and Lecture Notes in Computer Sciences (LNCS) series. Submissions should be anonymous, and formatted following the LNCS Style. All complete papers will be peer-reviewed by 3 members of the program committee, with double-blinded review process. The selection of the papers will be based on significance of results, technical merit, relevance and clarity of presentation.
- Abstracts: In order to encourage interesting discussions within the community, this year we are also inviting papers describing on-going, recently published or original research in a short paper of 2-3 pages. Depending on the number of submissions, we will organize a poster or oral session during GRAIL, where accepted short papers will be invited to present. The selection of the short abstracts will be made by members of the program committee and/or organizers. Submissions should be anonymous, and formatted following the LNCS Style.
|27 April 2020||Submission system opens.|
|7 July 2020||Full paper submission deadline.|
|29 July 2020||Notification of full paper decisions.|
|4 August 2020||Camera-ready full papers due.|
|14 September 2020||Abstract paper submission deadline.|
|21 September 2020||Notification of Abstract decisions.|