AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Capturing implicit hierarchical structure in 3D biomedical images with self-supervised hyperbolic representations

Joy Hsu, Jeffrey Gu, Gong Her Wu, Wah Chiu, Serena Yeung

NeurIPS 2021 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Self-supervised neuroimaging representations · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use hyperbolic instead of Euclidean latent space (base) (ablation)3D conv VAE · synthetic Level 1Dice: 0.784 -> 0.832Improved
Use hyperbolic instead of Euclidean latent space (base) (ablation)3D conv VAE · synthetic Level 3Dice: 0.109 -> 0.135Improved
Add hierarchical triplet loss in hyperbolic space (ablation)hyperbolic VAE · synthetic Level 2Dice: 0.352 -> 0.534Improved
Add hierarchical triplet loss in hyperbolic space (ablation)hyperbolic VAE · synthetic Level 3Dice: 0.135 -> 0.222Improved
Add hierarchical triplet loss in Euclidean space (ablation)Euclidean VAE · synthetic Level 1Dice: 0.784 -> 0.761Hurt
Add hierarchical triplet loss in Euclidean space (ablation)Euclidean VAE · synthetic Level 3Dice: 0.109 -> 0.153Improved
Add gyroplane convolutional layer (ablation)hyperbolic VAE · synthetic Level 1Dice: 0.832 -> 0.905Improved
Add gyroplane convolutional layer (ablation)hyperbolic VAE · synthetic Level 2Dice: 0.352 -> 0.473Improved
Add gyroplane conv to hyperbolic + triplet model (ablation)hyperbolic VAE + triplet · synthetic Level 3Dice: 0.222 -> 0.216No clear effect
Add gyroplane conv to hyperbolic + triplet model (ablation)hyperbolic VAE + triplet · synthetic Level 1Dice: 0.945 -> 0.952Improved
Use hyperbolic unsupervised method instead of best prior unsupervised baselineOurs · BraTS 2019 WTDice: 0.495 -> 0.684Improved
Use hyperbolic unsupervised method instead of best prior unsupervised baselineOurs · BraTS 2019average Hausdorff distance: 87.704 -> 77.940Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy