ISBI 2020 Workshop

Interaction of Geometry and Topology in Biomedical Imaging


Iowa City, Iowa April 4, 2020

Overview

Geometric approaches have been very effective in quantifying and characterizing complex anatomical shape differences and changes in biomedical images. In image segmentation, various topological approaches such as level sets, graph cuts and fuzzy connectedness have been effective. However, it's very difficult to separate topology from geometry in images. Often the combinations of geometric and topological approaches are more effective in quantifying complex images. For instance, topological constraints are enforced to have consistent shape preserving image deformation. Theoretically, the Gauss-Bonnet theorem connects geometry and topology through a single mathematical equation. Recently, topological data analysis (TDA) has been popular in revealing topological features that are persistent over multiple scales. TDA often employs geometric methods in quantifying topological changes. The main aim of this workshop (within ISBI 2020 conference) is to increase the awareness of the interaction between geometrical and topological approaches to the ISBI community. The program will include invited talks, as well as regular oral and poster sessions with contributed research papers. Best paper and poster awards will be given.


Invited Oral Session

Anuj Srivastava, Professor of Statistics and Distinguished Research Professor, Florida State University

Recent Advances in Geometric analysis of topologically-varying shapes


Jong Chul Ye KAIST Endowed Chair Professor of Bio and Brain Engineering, Korea Advanced Institute of Science & Technology (KAIST)

Geometric understanding of convolutional neural networks


Anqi Qiu, Dean’s Chair and the Deputy Head for Research & Enterprises, Associate Professor of Biomedical Engineering, National University of Singapore

Polynomial-based spectral graph convolutional neural network for diagnosis of Alzheimer’s disease


Punam Saha, Professor of Electrical & Computer Engineering and Radiology, University of Iowa.

Topological and geometric methods in osteoporotic imaging


Yuan Wang, Assistant Professor of Biostatistics, University of South Carolina.

Topological signal processing in neuroimaging studies

Contributed Oral Session

Yue Pan, University of Iowa

Pulmonary blood vessel and lobe surface varifold (PvSV) registration

Chao Chen, Stony Brook University

End-to-end training of neural networks with topological loss.

Poster Session


Rudrasis Chakraborty, University of California-Berkeley

A GMM based point-cloud generation algorithm and its application to neuroimaging

Moo Chung, University of Wisconsin-Madison

Parametric representation of sulcal and gyral trees

Xiaoyang Guo, Florida State University

Representations, metrics and statistics for shape analysis of elastic graphs


Won Hwa Kim, University of Texas-Arlington

Multi-resolution Graph Neural Network for Detecting Variations in Brain Connectivity

Arman Kulkami, University of Wisconsin-Madison

Investigating heritability across resting sate brain networks via heat kernel smoothing on persistence diagrams

Hangfan Liu, University of Pennsylvania

Cerebral Microbleed detection via Fourier descriptor with dual domain distribution modeling

Romuere Rodrigues Veloso Silva, Universidade Federal do Piauí

Fusion of color bands using genetic algorithm to segment melanoma

Tananun Songdechakraiwut, University of Wisconsin-Madison

Stationarity of Barcodes in Time Series of Brain Images

Kehong Yuan, Tsinghua University

Deformable registration using average geometric transformations for brain MR images

Kehong Yuan, Tsinghua University

GSRGAN: Medical image super-resolution using a generative adversarial network

Qing Zou, University of Iowa

Generative union of surfaces model: deep architectures re-explained


Topics

Any topic related to geometry or/and topology

  • Segmentation (level sets, graph cuts, fuzzy connectedness, deep learning)
  • Shape models and analysis
  • Image registration and deformable models
  • Geometric and spectral methods
  • Manifold-valued data, statistics on manifolds
  • Heterogeneous data including tubular structures (lung airways, skeletonization, neuronal trees, vessel trees)
  • Surface mesh processing and analysis
  • Topology correction
  • Topological data analysis
  • Geometry and topological structure of deep learning and CNN
  • Brain and biological network analysis


Paper Submission

Paper submission Deadline [1 page abstract]: January 29, 2020 Informally extended to sometime in February

  • Author Notification Date: TBD
  • Camera-ready Deadline: TBD
  • Workshop Date: April 4, 2020 (Saturday)
  • Final paper submission Deadline [4 page IEEE format]: April 15, 2020

The ISBI workshop proceedings will be archived in the IEEE Xplore Digital Library. The workshop paper format (initially 1 page abstract, optional 4 page final version) follows that of the ISBI 2020 main conference [link]. Paper should be submitted [here] with submission CODE: 67cg7


Reviewers

Moo K. Chung, Associate Professor, Department of Biostatistics and Medical Informatics, University of Wisconsin-Madison

Won Hwa Kim, Assistant Professor, Department of Computer Science and Engieering, University of Texas, Arlington

Joseph Reinhardt, Professor, Biomedical Engineering, University of Iowa


Organizers

  • Joseph Reinhardt, Professor and Chair of Biomedical Engineering, University of Iowa (joe-reinhardt@uiowa.edu)
  • Moo K. Chung, Associate Professor of Biostatistics and Medical Informatics, University of Wisconsin-Madison (mkchung@wisc.edu )