TY - GEN
T1 - Modeling Barrett’s Esophagus Progression Using Geometric Variational Autoencoders
AU - van Veldhuizen, Vivien
AU - Vadgama, Sharvaree
AU - de Boer, Onno
AU - Meijer, Sybren
AU - Bekkers, Erik J.
N1 - Publisher Copyright: © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2023
Y1 - 2023
N2 - Early detection of Barrett’s Esophagus (BE), the only known precursor to Esophageal adenocarcinoma (EAC), is crucial for effectively preventing and treating esophageal cancer. In this work, we investigate the potential of geometric Variational Autoencoders (VAEs) to learn a meaningful latent representation that captures the progression of BE. We show that hyperspherical VAE (S -VAE ) and Kendall Shape VAE show improved classification accuracy, reconstruction loss, and generative capacity. Additionally, we present a novel autoencoder architecture that can generate qualitative images without the need for a variational framework while retaining the benefits of an autoencoder, such as improved stability and reconstruction quality.
AB - Early detection of Barrett’s Esophagus (BE), the only known precursor to Esophageal adenocarcinoma (EAC), is crucial for effectively preventing and treating esophageal cancer. In this work, we investigate the potential of geometric Variational Autoencoders (VAEs) to learn a meaningful latent representation that captures the progression of BE. We show that hyperspherical VAE (S -VAE ) and Kendall Shape VAE show improved classification accuracy, reconstruction loss, and generative capacity. Additionally, we present a novel autoencoder architecture that can generate qualitative images without the need for a variational framework while retaining the benefits of an autoencoder, such as improved stability and reconstruction quality.
KW - Equivariance
KW - Geometric Deep Learning
KW - Oncology
KW - Pathology
KW - Representation Learning
KW - Variational Autoencoders
UR - http://www.scopus.com/inward/record.url?scp=85176003265&partnerID=8YFLogxK
U2 - https://doi.org/10.1007/978-3-031-45350-2_11
DO - https://doi.org/10.1007/978-3-031-45350-2_11
M3 - Conference contribution
SN - 9783031453496
VL - 14295 LNCS
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 132
EP - 142
BT - Cancer Prevention Through Early Detection - 2nd International Workshop, CaPTion 2023, Held in Conjunction with MICCAI 2023, Proceedings
A2 - Ali, Sharib
A2 - van der Sommen, Fons
A2 - van Eijnatten, Maureen
A2 - Kolenbrander, Iris
A2 - Papież, Bartłomiej W.
A2 - Jin, Yueming
PB - Springer Science and Business Media Deutschland GmbH
T2 - 2nd International Workshop on Cancer Prevention through early detecTion, CaPTion 2023
Y2 - 12 October 2023 through 12 October 2023
ER -