Latent Anomaly Detection: Masked VQ-GAN for Unsupervised Segmentation in Medical CBCT

TL;DR

This paper introduces an unsupervised segmentation method using a masked VQ-GAN trained on normal medical images to detect anomalies in CBCT scans, aiding diagnosis and treatment planning for jaw osteoradionecrosis.

Contribution

The study presents a novel two-stage training pipeline with masked VQ-GAN for unsupervised anomaly detection in 3D medical imaging, addressing data scarcity issues.

Findings

Successful segmentation on simulated data

Effective anomaly detection on real patient scans

Potential for integration with 3D printing workflows

Abstract

Advances in treatment technology now allow for the use of customizable 3D-printed hydrogel wound dressings for patients with osteoradionecrosis (ORN) of the jaw (ONJ). Meanwhile, deep learning has enabled precise segmentation of 3D medical images using tools like nnUNet. However, the scarcity of labeled data in ONJ imaging makes supervised training impractical. This study aims to develop an unsupervised training approach for automatically identifying anomalies in imaging scans. We propose a novel two-stage training pipeline. In the first stage, a VQ-GAN is trained to accurately reconstruct normal subjects. In the second stage, random cube masking and ONJ-specific masking are applied to train a new encoder capable of recovering the data. The proposed method achieves successful segmentation on both simulated and real patient data. This approach provides a fast initial segmentation solution, reducing the burden of manual labeling. Additionally, it has the potential to be directly used for 3D printing when combined with hand-tuned post-processing.