Temporal Consistency Objectives Regularize the Learning of Disentangled Representations

Gabriele Valvano, Agisilaos Chartsias, Andrea Leo, Sotirios A. Tsaftaris

Research output: Chapter in Book/Report/Conference proceedingConference contribution


There has been an increasing focus in learning interpretable feature representations, particularly in applications such as medical image analysis that require explainability, whilst relying less on annotated data (since annotations can be tedious and costly). Here we build on recent innovations in style-content representations to learn anatomy, imaging characteristics (appearance) and temporal correlations. By introducing a self-supervised objective of predicting future cardiac phases we improve disentanglement. We propose a temporal transformer architecture that given an image conditioned on phase difference, it predicts a future frame. This forces the anatomical decomposition to be consistent with the temporal cardiac contraction in cine MRI and to have semantic meaning with less need for annotations. We demonstrate that using this regularization, we achieve competitive results and improve semi-supervised segmentation, especially when very few labelled data are available. Specifically, we show Dice increase of up to 19\% and 7\% compared to supervised and semi-supervised approaches respectively on the ACDC dataset. Code is available at: https://github.com/gvalvano/sdtnet .
Original languageEnglish
Title of host publicationDomain Adaptation and Representation Transfer and Medical Image Learning with Less Labels and Imperfect Data
Subtitle of host publicationFirst MICCAI Workshop, DART 2019, and First International Workshop, MIL3ID 2019, Shenzhen, Held in Conjunction with MICCAI 2019, Shenzhen, China, October 13 and 17, 2019, Proceedings
ISBN (Electronic)978-3-030-33391-1
ISBN (Print)978-3-030-33390-4
Publication statusE-pub ahead of print - 13 Oct 2019
EventInternational Conference on Medical Image Computing and Computer Assisted Intervention - InterContinental Shenzhen, Shenzhen, China
Duration: 13 Oct 201917 Oct 2019
Conference number: 22


ConferenceInternational Conference on Medical Image Computing and Computer Assisted Intervention
Abbreviated titleMICCAI 2019
Internet address


  • Disentangled Representations
  • Semi-supervised Learning
  • Cardiac Segmentation


Dive into the research topics of 'Temporal Consistency Objectives Regularize the Learning of Disentangled Representations'. Together they form a unique fingerprint.

Cite this