PDF
Beyond Diagnostic Classification of Autism

Award Number and Duration

R01 1R01EB022876-01

Oct 1st, 2016 and June 30, 2019 (Estimated)

Points of Contact

Bei Wang
Assistant Professor
School of Computing, Scientific Computing and Imaging Institute
University of Utah
beiwang AT sci.utah.edu
http://www.sci.utah.edu/~beiwang

Tom Fletcher
Associate Professor
Electrical and Computer Engineering, Computer Science
University of Virginia
ptf8v AT virginia.edu
https://engineering.virginia.edu/faculty/tom-fletcher

Investigators

Bei Wang
Assistant Professor
School of Computing, Scientific Computing and Imaging Institute
University of Utah
beiwang AT sci.utah.edu
http://www.sci.utah.edu/~beiwang

Tom Fletcher
Associate Professor
Electrical and Computer Engineering, Computer Science
University of Virginia
ptf8v AT virginia.edu
https://engineering.virginia.edu/faculty/tom-fletcher

Brandon A. Zielinski
Assistant Professor of Pediatrics and Neurology
School of Medicine
University of Utah
brandon.zielinski AT hsc.utah.edu

Jeffrey S. Anderson
Associate Professor of Radiology
School of Medicine
University of Utah
J.Anderson AT hsc.utah.edu

Abstract

Autism spectrum disorder (ASD) is a heterogeneous disorder characterized by repetitive and stereotyped behavior and difficulties in communication and social interaction. It is now one of the most prevalent psychiatric disorders in childhood, but it is also a lifelong condition, adversely affecting an individual's social relationships, independence and employment well into adult. A major barrier to creating effective treatments for autism is the lack of understanding of the specific brain mechanisms involved and how these are related to specific behavioral symptoms. We propose to develop novel statistical methods for combining heterogeneous imaging and behavioral data to understand how properties of complex brain networks give rise to behavioral phenotypes in autism and other neuropsychiatric disorders. The first contribution of this project is to develop novel image analysis methods to extract individualized features of complex brain networks from imaging data. This includes powerful method for describing the shape of gray matter in brain networks based on diffeomorphic image registration and a rigorous method for inferring an individual's functional connectivity based on a hierarchical Bayesian model. The next contribution is a novel method to capture the topology of brain networks simultaneously across all scale levels of connection strength. Finally, we will develop Bayesian statistical methods for finding correlations in high-dimensional and heterogeneous data, and we will use this to analyze the relationship between brain networks and behavior. This project includes a strong collaborative and multi-disciplinary team with expertise in computer science, statistical data analysis, neuroimaging, and clinical autism care. A primary goal of this project is to create open-source software that is used by the neuroscience community to advance research in understanding the brain basis of behavior in neuropsychiatric disorders.

Public Health Relevance: This grant is focused on developing novel analysis of complex brain networks from structural and functional imaging and using this to link brain features to behavioral phenotypes. These algorithms will be used to discover brain biomarkers of behavioral traits in Autism and other neuropsychiatric disorders.

Publications

PDF Revisiting Abnormalities in Brain Network Architecture Underlying Autism Using Topology-Inspired Statistical Inference.
Sourabh Palande, Vipin Jose, Brandon Zielinski, Jeffrey Anderson, P. Thomas Fletcher and Bei Wang.
International Workshop on Connectomics in NeuroImaging (CNI) at MICCAI, 2017.
Brain Connectivity, 9(1):13-21, 2019
PDF Topological Data Analysis of Functional MRI Connectivity in Time and Space Domains.
Keri Anderson, Jeffrey Anderson, Sourabh Palande, and Bei Wang.
International Workshop on Connectomics in NeuroImaging (CNI) at MICCAI, 2018. Best Paper Award!
In: Wu G., Rekik I., Schirmer M., Chung A., Munsell B. (eds) Connectomics in NeuroImaging.
CNI 2018. Lecture Notes in Computer Science, vol 11083. Springer, Cham, 2018
Additional Material: Supplement. Publisher Version.
PDF Nonparametric Aggregation of Geodesic Trends for Longitudinal Data Analysis.
Kristen M. Campbell and P. Thomas Fletcher,
International Workshop on Shape in Medical Imaging (ShapeMI) at MICCAI, 2018. Best Paper Award!
In: Reuter M., Wachinger C., Lombaert H., Paniagua B., Luthi M., Egger B. (eds) Shape in Medical Imaging.
ShapeMI 2018. Lecture Notes in Computer Science, vol 11167. Springer, Cham, 2018.
Publisher Version.
PDF Riemannian Regression and Classification Models of Brain Networks Applied to Autism.
Eleanor Wong, Jeffrey S. Anderson, Brandon A. Zielinski and P. Thomas Fletcher.
International Workshop on Connectomics in Neuroimaging (CNI) at MICCAI, 2018.
In: Wu G., Rekik I., Schirmer M., Chung A., Munsell B. (eds) Connectomics in NeuroImaging.
CNI 2018. Lecture Notes in Computer Science, vol 11083. Springer, Cham, 2018.
Publisher Version.
PDF Efficient Parallel Transport in the Group of Diffeomorphisms via Reduction to the Lie Algebra.
Kristen M. Campbell and P. Thomas Fletcher.
In: Cardoso M. et al. (eds) Graphs in Biomedical Image Analysis, Computational Anatomy and Imaging Genetics.
GRAIL 2017, MFCA 2017, MICGen 2017. Lecture Notes in Computer Science, vol 10551. Springer, Cham, 2017.
Publisher Version.

Software Tools

GitHub repositories deployed and under active development:

  • FlashC: A free C++ library of a fast diffeomorphic image registration algorithm, including the latest (2018) code on parallel translation.

GitHub repository in documentation phase:

  • Autism Classification Using Topological Features and Deep Learning (Upcoming)
  • Topological Inference for scMRI (Upcoming)
  • Structure Averages of Merge Trees (Upcoming, demo video available)
  • Visualizing Topological Profiles of Brain Networks (Upcoming, demo video available): Our tool aims to incorporate the usage of topological data analysis (TDA) with brain network analysis to complement traditional visualization and analysis techniques.

Manuscripts

PDF A Kernel for Multi-Parameter Persistent Homology
René Corbet, Ulderico Fugacci, Michael Kerber, Claudia Landi, Bei Wang.
Manuscript, 2018.
arXiv Version: arXiv:1809.10231.
Computational Geometry Young Researchers Forum at International Symposium on Computational Geometry, abstract presented, 2018.
Autism Classification Using Topological Features and Deep Learning: A Cautionary Tale.
Archit Rathore, Sourabh Palande, Jeffrey Anderson, Brandon Zielinski, P. Thomas Fletcher, Bei Wang.
Manuscript, 2019.

Presentations, Educational Development and Broader Impacts

Bei Wang Invited Talk: Relating Functional Brain Network Topology to Clinical Measures of Behavior in Autism, at BIRS Workshop Topological Methods in Brain Network Analysis, May 7-12, 2017.

Bei Wang Invited Talks: Topological Thinking in Visualization and Structural Inference of Point Clouds, at Topological Data Analysis and Related Topics (TDART), Advanced Institute for Material Science (AIMR), Tohoku University, Japan, Feburary 8-10, 2017.

Bei Wang Workshop Organizer and Speaker, at International Workshop on Topological Data Analysis in Biomedicine (TDA-Bio), part of the 7th ACM Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB), Oct 2, 2016.

Students

Kris Campbell
Eleanor Wong
Atefeh Ghanaatikashani
Sourabh Palande
Lin Yan
Archit Rathore
Yaodong Zhao
Yiliang Shi (former)
Kyle Benson (former)
Sravan Neerati (former)
Chetal Patil (former)
Vipin Jose (former)

Acknowledgement

This material is based upon work supported or partially supported by the National Institutes of Health under Grant No. R01 1R01EB022876-01, project titled "Beyond Diagnostic Classification of Autism".

Any opinions, findings, and conclusions or recommendations expressed in this project are those of author(s) and do not necessarily reflect the views of the National Institutes of Health.

Web page last update: April 10, 2019.