Abstract #4731

Beyond Dice Coefficient: Evaluating Shape Biomarker Preservation in Neural Network Segmentations

Claudia Iriondo^1,2, Valentina Pedoia¹, Michael Girard³, and Sharmila Majumdar¹

¹Radiology and Biomedical Imaging, University of California, San Francsico, San Francisco, CA, United States, ²Bioengineering, University of California, Berkeley, Berkeley, CA, United States, ³Center for Digital Health Innovation, University of California, San Francsico, San Francisco, CA, United States

High accuracy scores in volumetric overlap metrics, such as Dice Similarity Coefficient, have not been proven to be reliable indicators of shape biomarker preservation. This study proposes a novel approach towards quantitative evaluation of segmentations from neural networks using PCA and contrastive PCA.

This abstract and the presentation materials are available to members only; a login is required.

Join Here