Deep Variational Clustering Framework for Self-labeling of Large-scale Medical Images

TL;DR

This paper introduces a Deep Variational Clustering framework that effectively learns representations and clusters large-scale medical images in an unsupervised manner, improving generalization across different datasets and modalities.

Contribution

The novel DVC framework combines probabilistic convolutional encoding and decoding with clustering loss for unsupervised medical image analysis.

Findings

Achieves superior clustering performance on multiple medical datasets.

Generalizes well across different imaging modalities.

Offers a self-training approach that refines latent space iteratively.

Abstract

We propose a Deep Variational Clustering (DVC) framework for unsupervised representation learning and clustering of large-scale medical images. DVC simultaneously learns the multivariate Gaussian posterior through the probabilistic convolutional encoder and the likelihood distribution with the probabilistic convolutional decoder; and optimizes cluster labels assignment. Here, the learned multivariate Gaussian posterior captures the latent distribution of a large set of unlabeled images. Then, we perform unsupervised clustering on top of the variational latent space using a clustering loss. In this approach, the probabilistic decoder helps to prevent the distortion of data points in the latent space and to preserve the local structure of data generating distribution. The training process can be considered as a self-training process to refine the latent space and simultaneously optimizing cluster assignments iteratively. We evaluated our proposed framework on three public datasets that represented different medical imaging modalities. Our experimental results show that our proposed framework generalizes better across different datasets. It achieves compelling results on several medical imaging benchmarks. Thus, our approach offers potential advantages over conventional deep unsupervised learning in real-world applications. The source code of the method and all the experiments are available publicly at: https://github.com/csfarzin/DVC