Print  Close

Spatially Dependent Elastic Shape Analysis of Curves

Abstract Number:

2865 

Submission Type:

Contributed Abstract 

Contributed Abstract Type:

Speed 

Participants:

Ye Jin Choi (1), Karthik Bharath (2), Sebastian Kurtek (1)

Institutions:

(1) The Ohio State University, Columbus, OH, (2) University of Nottingham, Nottingham, England

Co-Author(s):

Karthik Bharath  
University of Nottingham
Sebastian Kurtek  
The Ohio State University

First Author:

Ye Jin Choi  
The Ohio State University

Presenting Author:

Ye Jin Choi  
N/A

Abstract Text:

In this work, we model shapes of boundaries of two-dimensional objects that are dependent in space, i.e., spatially dependent shapes of parameterized curves. The proposed framework leverages an appropriately modified definition of the trace-variogram, which is commonly used to capture spatial dependence in functional data. This modified trace-variogram accounts for all shape-preserving transformations, e.g., scale, orientation and reparameterization, via distance-based registration, and can be used in various statistical tasks including (i) spatially informed shape clustering, and (ii) local spatially dependent shape summarization through kriging. The efficacy of our framework is demonstrated through extensive simulation studies. We highlight the method's practical utility and potential via an application to shapes of cell nuclei extracted from histopathology images.

Keywords:

shape|clustering|trace-variogram|kriging| |

Sponsors:

Section on Statistics in Imaging

Tracks:

Spatial Maps

Can this be considered for alternate subtype?

Yes

Are you interested in volunteering to serve as a session chair?

No

I have read and understand that JSM participants must abide by the Participant Guidelines.

Yes

I understand that JSM participants must register and pay the appropriate registration fee by June 1, 2024. The registration fee is non-refundable.

I understand