Explore topological data analysis and its application to biological image classification in this 30-minute conference talk. Learn how persistent homology, local homology, and average persistence landscapes can be combined with machine learning techniques to analyze high-resolution images of a cell's actin cytoskeleton. Discover a general approach that can be applied to various image classes, covering topics such as data science, persistence landscapes, stability, and support vector regression. Gain insights into the challenges and potential of this innovative method for analyzing biological structures through visualization techniques and real-world examples.
Peter Bubenik - Topological Data Analysis for Biological Images
Applied Algebraic Topology Network via YouTube
Overview
Syllabus
Introduction
Persistent homology
Structure theorem
Real data
Data science
Persistence landscape
Stability
Joint project
Actin cytoskeleton
Profilin
Subsample patches
Persistence landscapes
Support vector regression
Results
Visualization
Thank you
Wrap up
Parallel lines
Challenges
Questions
Taught by
Applied Algebraic Topology Network
Related Courses
-
Persistence Images
-
Limit Theorems in Topological Data Analysis
-
A Canonical Framework for Summarizing Persistence Diagrams
-
Multiparameter Persistence for Machine Learning in Topological Data Analysis
-
Identifying and Assessing Damage in Infrastructure Using Topological Data Analysis and Machine Learning
-
Machine Learning
4.0
