Unsupervised Segmentation of 3D Medical Images Based on Clustering and Deep Representation Learning

TL;DR

This paper introduces an unsupervised deep learning approach for segmenting 3D medical images, combining clustering and CNN-based feature learning to automatically identify pathological regions without manual annotations.

Contribution

It extends joint unsupervised learning (JULE) to 3D medical images using 3D convolutions, enabling effective unsupervised segmentation of complex medical scans.

Findings

Successfully segmented lung cancer micro-CT images into pathological and normal regions.

Demonstrated potential of unsupervised deep features for medical image segmentation.

Achieved promising results without requiring manual annotations.

Abstract

This paper presents a novel unsupervised segmentation method for 3D medical images. Convolutional neural networks (CNNs) have brought significant advances in image segmentation. However, most of the recent methods rely on supervised learning, which requires large amounts of manually annotated data. Thus, it is challenging for these methods to cope with the growing amount of medical images. This paper proposes a unified approach to unsupervised deep representation learning and clustering for segmentation. Our proposed method consists of two phases. In the first phase, we learn deep feature representations of training patches from a target image using joint unsupervised learning (JULE) that alternately clusters representations generated by a CNN and updates the CNN parameters using cluster labels as supervisory signals. We extend JULE to 3D medical images by utilizing 3D convolutions throughout the CNN architecture. In the second phase, we apply k-means to the deep representations from the trained CNN and then project cluster labels to the target image in order to obtain the fully segmented image. We evaluated our methods on three images of lung cancer specimens scanned with micro-computed tomography (micro-CT). The automatic segmentation of pathological regions in micro-CT could further contribute to the pathological examination process. Hence, we aim to automatically divide each image into the regions of invasive carcinoma, noninvasive carcinoma, and normal tissue. Our experiments show the potential abilities of unsupervised deep representation learning for medical image segmentation.