Screener: Self-supervised Pathology Segmentation in Medical CT Images

TL;DR

Screener introduces a fully self-supervised method for pathology segmentation in 3D CT images, outperforming existing UVAS techniques and setting new benchmarks in medical image analysis.

Contribution

It presents a novel dense self-supervised learning framework with masking-invariant features, eliminating the need for labeled data and improving pathology detection accuracy.

Findings

Outperforms existing UVAS methods on large-scale datasets

Surpasses current self-supervised pretraining approaches in supervised fine-tuning

Demonstrates state-of-the-art results in pathology segmentation

Abstract

Accurate detection of all pathological findings in 3D medical images remains a significant challenge, as supervised models are limited to detecting only the few pathology classes annotated in existing datasets. To address this, we frame pathology detection as an unsupervised visual anomaly segmentation (UVAS) problem, leveraging the inherent rarity of pathological patterns compared to healthy ones. We enhance the existing density-based UVAS framework with two key innovations: (1) dense self-supervised learning for feature extraction, eliminating the need for supervised pretraining, and (2) learned, masking-invariant dense features as conditioning variables, replacing hand-crafted positional encodings. Trained on over 30,000 unlabeled 3D CT volumes, our fully self-supervised model, Screener, outperforms existing UVAS methods on four large-scale test datasets comprising 1,820 scans with diverse pathologies. Furthermore, in a supervised fine-tuning setting, Screener surpasses existing self-supervised pretraining methods, establishing it as a state-of-the-art foundation for pathology segmentation. The code and pretrained models will be made publicly available.