NEWS!

  • 12.09.2024: NVIDIA sponsoring DLI course vouchers for Best Paper Awards. Check details in the Sponsor section.
  • 12.09.2024: Agenda update: check detailed talk titles and order of talks in the Program section.
  • 24.07.2024: Join our Workshop Journal Club! Check the GRAIL Journal Club.
  • 24.06.2024: The submission deadline has been extended until 29 June 2024, the maximum allowed date by MICCAI workshop chairs. Check the Important Dates.
  • 11.06.2024: The submission portal on CMT has been opened. Please submit here.
  • 05.04.2024: Excellent News - our proposal for GRAIL 2024 was accepted as an in-person workshop at MICCAI 2024. Stay tuned for further announcements!
  • 05.04.2024: The GRAIL 2024 Website is up and running! .
  • 18.04.2024: The deadlines have been released, check the Important Dates.

SCOPE

GRAIL 2024 is the sixth international Workshop on GRaphs in biomedicAl Image anaLysis, organised as an in-person satellite event of MICCAI 2024 in Marrakesh, Marocco.

Graphs are powerful mathematical structures that provide a flexible and scalable framework to model unstructured information, objects and their interactions in a readily interpretable fashion. Graph theory provides a solid mathematical foundation for models and algorithms, such as spectral analysis, dimensionality reduction, and network analysis. Since 2017, geometric deep learning has married the field of graph signal processing with the flexibility and rapid advancements of deep neural architectures. Since then, applications of graph neural networks (GNNs) in medicine have been steadily increasing, ranging from medical imaging and shape understanding, brain connectomics, population models and patient multi-omics to discovery and design of novel drugs and therapeutics. In the leading computer vision and ML conferences (CVPR, ICLR, NeurIPS), GNNs have been among the hottest topics in 2022 and 2023 and will likely increase in interest in 2024. With this workshop, we aim to provide a platform for understanding and application of graph-based models as versatile and powerful tools in biomedical image analysis and beyond. Our goal is to bring together scientists that develop graph-based models, and encourage their application to difficult clinical problems within a variety of biomedical data contexts. GRAIL 2023 featured keynote presentations from leading researchers in the graph community, and paper submissions on a wide range of topics, from brain connectomics and whole-slide image analytics, over semantic priors from biomedical knowledge graphs, explainable AI approaches for GNNs, to GNN-based genome alignment and their incorporation into multi-omics patient representations. For 2024, we expect a similarly exciting lineup of keynotes and paper submission.

In more detail, the scope of methodology topics includes but is not limited to:

  • Deep/machine learning on graphs with regular and irregular structures
  • Probabilistic graphical models for biomedical data analysis
  • Signal processing on graphs for biomedical image analysis, including non-learning based approaches
  • Explainable AI (XAI) methods in geometric deep learning
  • Big data analysis with graphs
  • Graphs for small data sets
  • Semantic graph research in medicine: Scene graphs and knowledge graphs
  • Modeling and applications of graph symmetry and equivariance
  • Graph generative models
  • Combination of graphs with other SOTA domains (e.g. self-supervised learning, federated learning)

Applications covered include but are not limited to:

  • Image segmentation, registration, classification
  • Graph representations in pathology imaging and whole-slide image analysis
  • Graph-based approaches in intra-operative surgical support
  • Graph-based shape modeling and dimensionality reduction
  • Graphs for large scale patient population analyses
  • Combining multimodal/multi-omics data through graph structures
  • Graph analysis of brain networks and connectomics
  • REGISTRATION

    As GRAIL 2024 will be an in-person workshop, registration will happen via the main conference website. Please visit the MICCAI 2024 Registration page for details.

    Workshop Agenda
    (Timezone: Marrakesh local, CET/GMT+1)

    Start Time End Time Description
    08:00 08:20 Welcome notes
    08:20 08:50 Session 1: Graph Learning in Medical Imaging
    • Supervised contrastive learning for image-to-graph transformers
      Anna Banaszak, Alexander H Berger, Laurin Lux, Suprosanna Shit, Daniel Rueckert, Johannes C. Paetzold

    • Graph Neural Networks: A suitable Alternative to MLPs in Latent 3D Medical Image Classification?
      Johannes Kiechle, Daniel M Lang, Stefan M Fischer, Lina Felsner, Jan C Peeken, Julia A Schnabel

    • Graph Residual Noise Learner Network for Brain Connectivity Graph Prediction
      Oytun Demirbilek, Alaa Bessadok, Tingying Peng
    08:50 09:20 Session 2: Disease Classification and Diagnosis with Graph Neural Networks
    • Prediction of radiological diagnostic errors from eye tracking data using graph neural networks and gaze-guided transformers
      Anna Anikina, Reza Karimzadeh, Diliara Ibragimova, Tamerlan Mustafaev, Claudia Mello-Thoms, Bulat Ibragimov

    • Exploring Graphs as Data Representation for Disease Classification in Ophthalmology
      Laurin Lux, Alexander H Berger, Maria Romeo-Tricas, Martin J Menten, Daniel Rueckert, Johannes C Paetzold

    • GAMMA-PD: Graph-based Analysis of Multi-Modal Motor Impairment Assessments in Parkinson's Disease
      Favour Nerrise, Alice Heiman, Ehsan Adeli
    09:20 10:00 Keynote 1
    Title: Modeling multimodal fusion through graph neural networks.
    Tanveer Sayeda Mehmood
    10:00 10:30 Coffee Break
    10:30 11:15 Keynote 2
    Title: Topological Deep Learning: Frontiers and Opportunities in Drug Discovery.
    Dr. Mustafa Hajij
    11:15 11:45 Session 3: Graph-Based Histopathology and Tissue Analysis
    • Multi-Resolution Histopathology Patch Graphs for Ovarian Cancer Subtyping
      Jack Breen, Katie Allen, Kieran Zucker, Nicolas M Orsi, Nishant Ravikumar

    • HistoGraphCoarse: Strategizing Graph Coarsening Techniques for Efficient Analysis of Gigapixel Histopathological Images
      Ekta Srivastava, Syed Mohammed Danish, Kumar Arjun, Manoj Kumar, Mohit Kataria, Syed Farhan Abbas, Ishaan Gupta, Sandeep Kumar

    • Mesh registration via geometric feature homogenization and offset cross-attention: application to 3D photogrammetry
      Ines Alejandro Cruz Guerrero, Connor Elkhill, Jiawei Liu, Phuong Nguyen, Brooke French, Antonio R Porras
    11:45 12:15 Session 4: Graph-Based Models for Surgical and Vascular Applications
    • SANGRIA: Surgical Video Scene Graph Optimization for Surgical Workflow Prediction
      Caghan Koksal, Ghazal Ghazaei, Felix Holm, Azade Farshad, Nassir Navab

    • DVasMesh: Deep Structured Mesh Reconstruction from Vascular Images for Dynamics Modeling of Vessels
      Dengqiang Jia, Xinnian Yang, Xiong Xiaosong, Shijie Huang, Feiyu Hou, Li Qin, Kaicong Sun, Kannie Wai Yan Chan, Dinggang Shen

    • DuoGNN: Topology-aware Graph Neural Network with Homophily and Heterophily Interaction-Decoupling
      Kevin Mancini, Islem Rekik
    12:15 12:30 Closing and Award Ceremony

    Keynote Speakers

    Dr. Tanveer Syeda-Mahmood
    IBM Research

    Talk title: Modeling multimodal fusion through graph neural networks.

    Dr. Tanveer Syeda-Mahmood is an IBM Fellow and Global Imaging AI Leader at IBM Research, advancing multimodal and bioinspired AI. Her group's work draws inspiration from computational models of brain functions, aiming to develop AI technologies for healthcare decision support, scientific discovery, and hybrid cloud data management. She previously led the Medical Sieve Radiology Grand Challenge project, which contributed to AI advancements in radiology and the creation of IBM Watson Health Imaging products. Her team has received over 10 best paper awards at major healthcare forums, including AMIA's Homer Warner Award. Dr. Syeda-Mahmood graduated from MIT with a Ph.D. in Computer Science and has over 250 refereed publications and 120 patents. She has held significant roles in conferences such as MICCAI 2023 and IEEE ISBI 2022 and is a Fellow of IEEE, AIMBE, and MICCAI. Dr. Syeda-Mahmood has received several key awards, including the IBM Corporate Award and Best of IBM Award.

    Dr. Mustafa Hajij
    University of San Francisco

    Talk title: Topological Deep Learning: Frontiers and Opportunities in Drug Discovery

    Dr. Hajij is an Assistant Professor specializing in AI and Data Science at the University of San Francisco’s Master of Science in Data Science Program. With over 9 years of research and industrial experience, he has delved into graph neural networks, topological data analysis, intelligent transportation, and topological deep learning. His expertise extends to industrial AI applications, with a focus on topological deep learning, geometric data processing, time-varying data, and predictive modeling. He co-founded AltumX, a startup utilizing deep learning for intelligent road network systems, and actively participates in AI-related workshops and conferences. He published more than 70 publications in journal and conference papers, as well as patents. Dr. Hajij served as the main organizer for MICCAI TDA workshops in 2021 and 2022. He made contributions to the tech industry, spearheading the development of innovative software solutions for both KLA Corporation and AltumX Inc.

    Submit a paper

    Submissions should be done through the GRAIL 2024 CMT portal.

    This year we will accept papers in the following format:

    • Complete papers: Papers describing original research with MICCAI length (max 8 pages text + max 2 pages references, supplemental material allowed). 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. Style. All complete papers will be peer-reviewed by 3 members of the program committee, with a double-blinded review process. The selection of the papers will be based on the significance of results, technical merit, relevance, and clarity of presentation.

    GRAIL Best Paper Award

    It will be announced during the award ceremony.

    Important dates

    29 June 2024 Abstract submission deadline (Intention to submit), 23:59 pm, PST
    29 June 2024 Submission deadline, 23:59 pm, PST
    8 July 2024 Reviews due, 23:59 pm, PST
    15 July 2024 Notification of acceptance, 23:59 pm, PST
    22 July 2024 Camera ready paper versions due, 23:59 pm, PST

    GRAIL Journal Club

    Welcome to the GRAIL Workshop Journal Club! Our mission is to create a vibrant and engaging community where we read and discuss the latest cutting-edge papers, bringing the Graph Deep Learning (GDL) community closer together. Join us monthly to foster collaborations, brainstorm innovative ideas, and connect with fellow GDL researchers. Whether you're looking to stay updated on the newest advancements or seeking a platform to share your insights or discuss your research and gain feedback, the GRAIL Workshop Journal Club is the perfect place for you. Let's advance the frontiers of GDL together! If you cannot access the form and want to be part of the journal club, please send an email to akazi1@mgh.harvard.edu and niharika.dsouza@ibm.com

    Registration: Register here

    Next Meeting: Kindly sign up to receive the most up-to-date information.

    Meeting platform: Via Zoom

    Organising Committee

    General Chairs: Co-Chairs:

    Past Events


    Sponsor

    We are very grateful to our sponsor NVIDIA for their valuable support, who are awarding six tuition wavers to self-paced courses of the Deep Learning Institute (DLI). These vouchers will be gifted to the winners of the GRAIL 2024 Best Paper Award.