Instructors: Prof. Dr. Nassir Navab, Anees Kazi,  Mahsa Ghorbani, Matthias Keicher, Tamara Mueller, Kamilia Mullakaeva, Tobias Czempiel, Ege Ozsoy, Felix Holm

 

Announcements

  • 26.04.2022 - Seminar starting date
  • 06.04.2022 - List of papers will be announced
  • 23.03.2022 - Deadline for deregistration 
  • 08.03.2022 - Notification for acceptance to the seminar
  • 03.02.2022 - Premilinary meeting
  • 03.03.2022 - Contact information. If you have any questions about the seminar, feel free to contact  Anees Kazi (anees.kazi@tum.de)
  • 10.02.2022 - Registrations are open through the TUM Matching Platform. Additionally, let us know your interest through our application form
  • The maximum number of participants: 20.
  • Join the lecture virtually at: 

    Zoom Meeting
    https://tum-conf.zoom.us/j/69782153188

    Meeting ID: 697 8215 3188
    Passcode: 095144

Introduction 

Graph Deep Learning is a new exotic branch in many fields like computer Vision and Medical Imaging. Many real-world medical and non-medical datasets can be represented in the form of graphs, providing a powerful source of information for machine learning models. This graph-based data, combined with the success of the convolutional neural networks, has motivated to translate the key ingredients of deep learning models into the graph domain. Many communities such as healthcare, social media, and computer vision are moving towards analyzing the data using Graph Convolutions. This seminar provides a space for discussion of the recent scientific publications on GCN with a focus on their medical applications and others.

Registration 

  • Interested students should attend the introductory meeting on 03.02.2022 to enlist in the course. Details are available at TUM online. 
  • Students can only register from 10.02. bis 15.02.2022 through TUM Matching Platform themselves if the maximum number of participants hasn't been reached (please pay attention to the Deadlines).
  • A maximum number of participants: 20.

Requirements and Course Structure

For this course, previous background in Machine/Deep Learning is required. 

During the course, students will select and present a scientific publication from the given list. Additionally, students will also contribute to the course blog post with their own summary of the selected article. Requirements are described in the image below: 

List of Topics

TopicNumberTitleJournal/ConferenceTutorStudentLink
Architecture Design1Improving Graph Neural Networks with Simple Architecture DesignPreprint

Franz Rieger 

franz.rieger@bi.mpg.de

Berfin Elif Erdoğanhttps://arxiv.org/pdf/2105.07634v1.pdf
Road Graphs and Graph-Tensor Encodings2Sat2Graph: Road Graph Extraction through Graph-Tensor EncodingECCV 2020

Johannes Paetzold

johannes.paetzold@tum.de

Marcos Balle Sanchezhttps://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123690052.pdf
Brain Mapping32D histology meets 3D topology: Cytoarchitectonic brain
mapping with Graph Neural Networks
MICCAI 2021

Alina Dima

alina.dima@tum.de

Merve Elif Demirtashttps://arxiv.org/pdf/2103.05259.pdf
Registration4GraphRegNet: Deep Graph Regularisation Networks on Sparse Keypoints for Dense Registration of 3D Lung CTsIEEE Transactions on Medical Imaging 2021

Mohammad Farid Azampour

mf.azampour@tum.de 

Ha Young Kimhttps://ieeexplore.ieee.org/abstract/document/9406964
Registration5IGCN: Image-to-graph Convolutional Network for 2D/3D Deformable RegistrationPreprint

Mohammad Farid Azampour

mf.azampour@tum.de 

Ata Jadid Aharihttps://arxiv.org/pdf/2111.00484.pdf
Scene Graphs6Spatial-Temporal Transformer for Dynamic Scene Graph GenerationICCV 2021

Ege Özsoy

ege.oezsoy@tum.de

Caghan Koksalhttps://arxiv.org/pdf/2107.12309.pdf
Scene Graphs7Target Adaptive Context Aggregation for Video Scene Graph GenerationICCV 2021

Ege Özsoy

ege.oezsoy@tum.de

Haichuan Lihttps://arxiv.org/pdf/2108.08121.pdf
Surgical Scene Graphs 8Global-Reasoned Multi-Task Learning Model for Surgical Scene UnderstandingROL 2022

Felix Holm

felix.holm@tum.de

Yuhan Lihttps://arxiv.org/pdf/2201.11957.pdf
Surgical Scene Graphs 9Learning and Reasoning with the Graph Structure Representation in Robotic SurgeryMICCAI 2020

Felix Holm

felix.holm@tum.de

Chengzhi Shenhttps://arxiv.org/pdf/2007.03357.pdf
Scene Graph Generation10RelTR: Relation Transformer for Scene Graph GenerationPreprint

Matthias Keicher

matthias.keicher@tum.de 

Priyank Upadhyahttps://arxiv.org/abs/2201.11460
Medical Report Generation, Few-shot Learning11MetaDT: Meta Decision Tree for Interpretable Few-Shot LearningPreprint

Matthias Keicher

matthias.keicher@tum.de 

Chaehyeon Simhttps://arxiv.org/abs/2203.01482
Anomaly detection,  chest x-ray application 12

Chest Radiograph Disentanglement for COVID-19 Outcome Prediction  

MICCAI 2021

Kamilia Mullakaeva 

kamilia.mullakaeva@tum.de

Diyorbek Rustamovhttps://arxiv.org/pdf/2105.09937.pdf
GNN Architecture13Neural Trees for Learning on GraphsNeurIPS 2021

Evin Pınar Örnek

evin.oernek@tum.de 

Niklas Vesthttps://openreview.net/pdf?id=UwSwML5iJkp
Self-supervision14Context Matters: Graph-based Self-supervised Representation Learning for Medical ImagesAAAI 2021

Mahsa Ghorbani

mahsa.ghorbani@tum.de

Xinyue Zhanghttps://arxiv.org/pdf/2012.06457.pdf
Anomaly detection using GCN15Airway Anomaly Detection by Graph Neural Network
MICCAI 2021

Kamilia Mullakaeva 

kamilia.mullakaeva@tum.de


https://miccai2021.org/openaccess/paperlinks/2021/09/01/046-Paper0604.html
Explainability of GNNs16On Explainability of Graph Neural Networks via Subgraph ExplorationsICML 2021

Tamara Mueller

tamara.mueller@tum.de

Malika    Sanhinovahttp://proceedings.mlr.press/v139/yuan21c/yuan21c.pdf

Scene Graphs

17Decoupling Object Detection from Human-Object Interaction RecognitionPreprint

Tobias Czempiel

tobias.czempiel@tum.de

Xinyu    Chenhttps://arxiv.org/pdf/2112.09828.pdf

Dynamic Scene Graphs

18Exploiting Long-Term Dependencies for Generating Dynamic Scene GraphsPreprint

Tobias Czempiel

tobias.czempiel@tum.de


https://arxiv.org/pdf/2112.09828.pdf
Pooling19Hierarchical Graph Representation Learning with Differentiable PoolingAdvances in neural information processing systems 2018

Anees Kazi

anees.kazi@tum.de

Daniel    Stollhttps://arxiv.org/abs/1806.08804
Uncertainty20Edge-variational Graph Convolutional Networks for Uncertainty-aware Disease PredictionMICCAI 2020

Anees Kazi

anees.kazi@tum.de

Rafael    Cabral Muchachohttps://arxiv.org/abs/2009.02759


Schedule


Sessions will take place Tuesdays from 12:00 to 14:00 hrs. A group of Two to Three students will present their papers during each session. 

Sessions will start on 26.04 with introductory lectures by the organizers. Room is 03.09.012.

DateTimeTopicAdditional Info/ slides
26.04.22 12:00 - 14:00Lecture: Intro to GDLMA

03.05.22





12:00 - 14:00Improving Graph Neural Networks with Simple Architecture Design
Sat2Graph: Road Graph Extraction through Graph-Tensor Encoding
Hierarchical Graph Representation Learning with Differentiable Pooling
10.05.2212:00 - 14:00On Explainability of Graph Neural Networks via Subgraph Explorations
MetaDT: Meta Decision Tree for Interpretable Few-Shot Learning
Edge-variational Graph Convolutional Networks for Uncertainty-aware Disease Prediction
17.05.2212:00 - 14:00Talk 1: Dr. Seyed  Ahmad Ahmadi (NVIDIA)




24.05.2212:00 - 14:00Chest Radiograph Disentanglement for COVID-19 Outcome Prediction
Context Matters: Graph-based Self-supervised Representation Learning for Medical Images
Airway Anomaly Detection by Graph Neural Network
31.05.2212:00 - 14:00

2D histology meets 3D topology: Cytoarchitectonic brain

mapping with Graph Neural Networks


GraphRegNet: Deep Graph Regularisation Networks on Sparse Keypoints for Dense Registration of 3D Lung CTs
IGCN: Image-to-graph Convolutional Network for 2D/3D Deformable Registration
14.06.2212:00 - 14:00Talk 2 : Alaa Bessadok (Helmholtz)




21.06.2212:00 - 14:00Lecture: Intro to Scene Graphs




28.06.22


12:00 - 14:00Neural Trees for Learning on Graphs
RelTR: Relation Transformer for Scene Graph Generation
Decoupling Object Detection from Human-Object Interaction Recognition
5.07.2212:00 - 14:00Spatial-Temporal Transformer for Dynamic Scene Graph Generation
Target Adaptive Context Aggregation for Video Scene Graph Generation
Sat2Graph: Road Graph Extraction through Graph-Tensor Encoding
12.07.2212:00 - 14:00Global-Reasoned Multi-Task Learning Model for Surgical Scene Understanding
Learning and Reasoning with the Graph Structure Representation in Robotic Surgery


19.07.2212:00 - 14:00Talk 3 : Azade Farshad (CAMP)




26.07.2212:00 - 14:00

Paper session - Backup






Resources and Material

An introductory lecture to GCNs by Prof. Xavier Bresson on Alfredo Canziani Youtube channel:


If you find more interesting material you would like to share in this section, feel free to contact us. 

  • No labels