Optimization with soft Dice can lead to a volumetric bias

6 November 2019

Papers citing "Optimization with soft Dice can lead to a volumetric bias"

7 / 7 papers shown

Title
Theoretical analysis and experimental validation of volume bias of soft Dice optimized segmentation maps in the context of inherent uncertainty J. Bertels D. Robben Dirk Vandermeulen P. Suetens 30 19 0 08 Nov 2022
Weakly Supervised Medical Image Segmentation With Soft Labels and Noise Robust Loss B. Felfeliyan A. Hareendranathan G. Kuntze S. Wichuk Nils D. Forkert Jacob L. Jaremko J. Ronsky NoLa 41 2 0 16 Sep 2022
Hypernet-Ensemble Learning of Segmentation Probability for Medical Image Segmentation with Ambiguous Labels Sun-Beom Hong A. Bonkhoff Andrew Hoopes Martin Bretzner M. Schirmer A. Giese Adrian Dalca Polina Golland N. Rost UQCV 30 7 0 13 Dec 2021
Calibrating the Dice loss to handle neural network overconfidence for biomedical image segmentation Michael Yeung L. Rundo Yang Nan Evis Sala Carola-Bibiane Schönlieb Guang Yang UQCV 25 30 0 31 Oct 2021
Unified Focal loss: Generalising Dice and cross entropy-based losses to handle class imbalanced medical image segmentation Michael Yeung Evis Sala Carola-Bibiane Schönlieb L. Rundo 32 394 0 08 Feb 2021
SoftSeg: Advantages of soft versus binary training for image segmentation C. Gros A. Lemay Julien Cohen-Adad 33 70 0 18 Nov 2020
Confidence Calibration and Predictive Uncertainty Estimation for Deep Medical Image Segmentation Alireza Mehrtash W. Wells C. Tempany Purang Abolmaesumi Tina Kapur OOD FedML UQCV 24 264 0 29 Nov 2019