Modeling Barrett’s Esophagus Progression Using Geometric Variational Autoencoders

Vivien van Veldhuizen, Sharvaree Vadgama, Onno de Boer, Sybren Meijer, Erik J. Bekkers

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review


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.
Original languageEnglish
Title of host publicationCancer Prevention Through Early Detection - 2nd International Workshop, CaPTion 2023, Held in Conjunction with MICCAI 2023, Proceedings
EditorsSharib Ali, Fons van der Sommen, Maureen van Eijnatten, Iris Kolenbrander, Bartłomiej W. Papież, Yueming Jin
PublisherSpringer Science and Business Media Deutschland GmbH
Number of pages11
Volume14295 LNCS
ISBN (Print)9783031453496
Publication statusPublished - 2023
Event2nd International Workshop on Cancer Prevention through early detecTion, CaPTion 2023 - Vancover, Canada
Duration: 12 Oct 202312 Oct 2023

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume14295 LNCS


Conference2nd International Workshop on Cancer Prevention through early detecTion, CaPTion 2023


  • Equivariance
  • Geometric Deep Learning
  • Oncology
  • Pathology
  • Representation Learning
  • Variational Autoencoders

Cite this