# SepVAE: a contrastive VAE to separate pathological patterns from healthy ones Source: [SepVAE: a contrastive VAE to separate pathological patterns from healthy ones]() Collection index: [llms.txt]() Data updated: 2025-10-26T02:35:49.064Z Authors: Robin Louiset, Edouard Duchesnay, Antoine Grigis, Benoit Dufumier, Pietro Gori Published: 2023-07-12T14:52:21.000Z ## Abstract Contrastive Analysis VAE (CA-VAEs) is a family of Variational auto-encoders (VAEs) that aims at separating the common factors of variation between a background dataset (BG) (i.e., healthy subjects) and a target dataset (TG) (i.e., patients) from the ones that only exist in the target dataset. To do so, these methods separate the latent space into a set of salient features (i.e., proper to the target dataset) and a set of common features (i.e., exist in both datasets). Currently, all models fail to prevent the sharing of information between latent spaces effectively and to capture all salient factors of variation. To this end, we introduce two crucial regularization losses: a disentangling term between common and salient representations and a classification term between background and target samples in the salient space. We show a better performance than previous CA-VAEs methods on three medical applications and a natural images dataset (CelebA). Code and datasets are available on GitHub https://github.com/neurospin-projects/2023\_rlouiset\_sepvae. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Latent Factor Separation on FFHQ No Glasses vs. Glasses](): Latent Factor Separation; FFHQ No Glasses vs. Glasses; C Score: 0.59. - [Latent Factor Separation on FFHQ (Male vs Female)](): Latent Factor Separation; FFHQ Male vs. Female; C Score: 0.65. - [Image Reconstruction on FFHQ (SSIM, LPIPS, ID-Sim)](): Image Reconstruction; FFHQ; SSIM: 35.8. - [Latent Factor Separation on FFHQ Smile vs. Non-Smiling](): Latent Factor Separation; FFHQ Smile vs. Non-Smiling; C Score: 0.65. - [Salient Attribute Swapping on FFHQ Glasses Attribute](): Salient Attribute Swapping; FFHQ Glasses Attribute; Accuracy: 19.6. - [Latent Factor Separation on FFHQ Head Pose: Frontal vs. Right/Left](): Latent Factor Separation; FFHQ Head Pose: Frontal vs. Right/Left; C Metric: 0.69. - [Attribute Swapping on FFHQ Swap X to Y high-quality (test)](): Attribute Swapping; FFHQ Swap X to Y high-quality (test); ID Similarity: 0.142. - [Attribute Swapping on FFHQ Swap Y to X high-quality (test)](): Attribute Swapping; FFHQ Swap Y to X high-quality (test); ID-Sim: 0.14. - [Image Reconstruction on FFHQ X dataset high-quality (test)](): Image Reconstruction; FFHQ X dataset high-quality (test); LPIPS: 0.556. - [Image Reconstruction on FFHQ Y dataset high-quality (test)](): Image Reconstruction; FFHQ Y dataset high-quality (test); LPIPS: 0.573. ## Original sources and implementation [Original paper]() [PDF]()