Interpreting Latent Spaces of Generative Models for Medical Images using Unsupervised Methods

TL;DR

This paper demonstrates that unsupervised methods for discovering interpretable directions in the latent spaces of generative models can be effectively applied to medical images, revealing meaningful transformations and 3D structures.

Contribution

It extends unsupervised latent space interpretation techniques from natural images to medical images, specifically thoracic CT scans, showing their effectiveness and generalization.

Findings

Identified interpretable directions related to rotation and breast size.

Revealed that models capture 3D structure from 2D data.

Demonstrated generalization of methods from GANs to VAEs.

Abstract

Generative models such as Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs) play an increasingly important role in medical image analysis. The latent spaces of these models often show semantically meaningful directions corresponding to human-interpretable image transformations. However, until now, their exploration for medical images has been limited due to the requirement of supervised data. Several methods for unsupervised discovery of interpretable directions in GAN latent spaces have shown interesting results on natural images. This work explores the potential of applying these techniques on medical images by training a GAN and a VAE on thoracic CT scans and using an unsupervised method to discover interpretable directions in the resulting latent space. We find several directions corresponding to non-trivial image transformations, such as rotation or breast size. Furthermore, the directions show that the generative models capture 3D structure despite being presented only with 2D data. The results show that unsupervised methods to discover interpretable directions in GANs generalize to VAEs and can be applied to medical images. This opens a wide array of future work using these methods in medical image analysis.