Facilitating Valid Statistical Inference in Biomedical Image Synthesis

Abstract Number:

2605 

Submission Type:

Contributed Abstract 

Contributed Abstract Type:

Poster 

Participants:

Jiangmei Xiong (1), Shunxing Bao (1), Harsimran Kaur (1), Cody Heiser (2), Eliot McKinley (2), Joseph Roland (2), Robert Coffey (2), Martha Shrubsole (2), Ken Lau (1), Simon Vandekar (1), Siyuan Ma (1)

Institutions:

(1) Vanderbilt University, Nashville, TN, USA, (2) Vanderbilt University Medical Center, Nashville, TN, USA

Co-Author(s):

Shunxing Bao  
Vanderbilt University
Harsimran Kaur  
Vanderbilt University
Cody Heiser  
Vanderbilt University Medical Center
Eliot McKinley  
Vanderbilt University Medical Center
Joseph Roland  
Vanderbilt University Medical Center
Robert Coffey  
Vanderbilt University Medical Center
Martha Shrubsole  
Vanderbilt University Medical Center
Ken Lau  
Vanderbilt University
Simon Vandekar  
Vanderbilt University
Siyuan Ma  
Vanderbilt University

First Author:

Jiangmei Xiong  
Vanderbilt University

Presenting Author:

Jiangmei Xiong  
N/A

Abstract Text:

Image synthesis is an important and growing field of research fueled by rapid progress in image-based artificial intelligence and has been employed in neuroimaging, multiplexed immunofluorescence (MxIF), and imaging spatial transcriptomics. The number of publications on image synthesis has nearly tripled in the last decade, but there is no evaluation of the consequence of using it in medical research. Currently, biomedical image synthesis 1) does not include relevant clinical information, and 2) fails to provide statistical uncertainty. As postulated by multiple imputation theory, neglecting either issue can lead to invalid downstream statistical analysis. In this paper, we systematically examine these issues in state-of-the-art image synthesis algorithms with real-world imaging data. We demonstrate that 1) current imputation tools often lead to biased point estimates and anti-conservative standard errors, and 2) such issues can be alleviated by simple, post-hoc augmentation steps derived from multiple imputation literature. This work is pioneer in highlighting invalid findings on synthesized biomedical imaging data, and providing expeditious solutions.

Keywords:

Statistical imaging|Machine learning|Spatial genomics|Multiple imputation| |

Sponsors:

Section on Statistics in Genomics and Genetics

Tracks:

Miscellaneous

Can this be considered for alternate subtype?

Yes

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

Yes

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