Joint Embedding of 2D and 3D Networks for Medical Image Anomaly Detection

TL;DR

This paper introduces a joint embedding approach combining 2D and 3D networks with self-supervised learning to improve anomaly detection in medical images, addressing limitations of existing methods.

Contribution

It proposes a novel method that integrates 2D and 3D networks via joint embedding and a self-supervised pretask, enhancing anomaly detection performance.

Findings

Outperforms state-of-the-art methods in classification tasks.

Achieves superior segmentation accuracy.

Effectively learns from normal images only.

Abstract

Obtaining ground truth data in medical imaging has difficulties due to the fact that it requires a lot of annotating time from the experts in the field. Also, when trained with supervised learning, it detects only the cases included in the labels. In real practice, we want to also open to other possibilities than the named cases while examining the medical images. As a solution, the need for anomaly detection that can detect and localize abnormalities by learning the normal characteristics using only normal images is emerging. With medical image data, we can design either 2D or 3D networks of self-supervised learning for anomaly detection task. Although 3D networks, which learns 3D structures of the human body, show good performance in 3D medical image anomaly detection, they cannot be stacked in deeper layers due to memory problems. While 2D networks have advantage in feature detection, they lack 3D context information. In this paper, we develop a method for combining the strength of the 3D network and the strength of the 2D network through joint embedding. We also propose the pretask of self-supervised learning to make it possible for the networks to learn efficiently. Through the experiments, we show that the proposed method achieves better performance in both classification and segmentation tasks compared to the SoTA method.