TY - GEN
T1 - Generalized Probabilistic U-Net for Medical Image Segementation
AU - Bhat, Ishaan
AU - Pluim, Josien P.W.
AU - Kuijf, Hugo J.
N1 - Funding Information:
Acknowledgements. This work was financially supported by the project IMPACT (Intelligence based iMprovement of Personalized treatment And Clinical workflow supporT) in the framework of the EU research programme ITEA3 (Information Technology for European Advancement).
Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2022
Y1 - 2022
N2 - We propose the Generalized Probabilistic U-Net, which extends the Probabilistic U-Net [14] by allowing more general forms of the Gaussian distribution as the latent space distribution that can better approximate the uncertainty in the reference segmentations. We study the effect the choice of latent space distribution has on capturing the uncertainty in the reference segmentations using the LIDC-IDRI dataset. We show that the choice of distribution affects the sample diversity of the predictions and their overlap with respect to the reference segmentations. For the LIDC-IDRI dataset, we show that using a mixture of Gaussians results in a statistically significant improvement in the generalized energy distance (GED) metric with respect to the standard Probabilistic U-Net. We have made our implementation available at https://github.com/ishaanb92/GeneralizedProbabilisticUNet.
AB - We propose the Generalized Probabilistic U-Net, which extends the Probabilistic U-Net [14] by allowing more general forms of the Gaussian distribution as the latent space distribution that can better approximate the uncertainty in the reference segmentations. We study the effect the choice of latent space distribution has on capturing the uncertainty in the reference segmentations using the LIDC-IDRI dataset. We show that the choice of distribution affects the sample diversity of the predictions and their overlap with respect to the reference segmentations. For the LIDC-IDRI dataset, we show that using a mixture of Gaussians results in a statistically significant improvement in the generalized energy distance (GED) metric with respect to the standard Probabilistic U-Net. We have made our implementation available at https://github.com/ishaanb92/GeneralizedProbabilisticUNet.
KW - Image segmentation
KW - Uncertainty estimation
KW - Variational inference
UR - http://www.scopus.com/inward/record.url?scp=85138830438&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-16749-2_11
DO - 10.1007/978-3-031-16749-2_11
M3 - Conference contribution
AN - SCOPUS:85138830438
SN - 9783031167485
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 113
EP - 124
BT - Uncertainty for Safe Utilization of Machine Learning in Medical Imaging - 4th International Workshop, UNSURE 2022, Held in Conjunction with MICCAI 2022, Proceedings
A2 - Sudre, Carole H.
A2 - Sudre, Carole H.
A2 - Baumgartner, Christian F.
A2 - Dalca, Adrian
A2 - Dalca, Adrian
A2 - Wells III, William M.
A2 - Qin, Chen
A2 - Tanno, Ryutaro
A2 - Van Leemput, Koen
A2 - Van Leemput, Koen
A2 - Wells III, William M.
PB - Springer Science and Business Media Deutschland GmbH
T2 - 4th Workshop on Uncertainty for Safe Utilization of Machine Learning in Medical Imaging, UNSURE 2022, held in conjunction with 25th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2022
Y2 - 18 September 2022 through 18 September 2022
ER -