Hide-and-Seek Attribution: Weakly Supervised Segmentation of Vertebral Metastases in CT

TL;DR

This paper presents a novel weakly supervised method for segmenting vertebral metastases in CT scans using only vertebra-level labels, leveraging generative editing and attribution techniques to achieve high accuracy.

Contribution

It introduces Hide-and-Seek Attribution, a new approach combining a Diffusion Autoencoder and selective occlusion to generate lesion masks without lesion-level annotations.

Findings

Achieves F1 scores of 0.91/0.85 for blastic/lytic lesions, outperforming baselines.

Demonstrates that vertebra-level labels can be effectively transformed into lesion masks.

Shows that generative editing with occlusion supports accurate weakly supervised segmentation.

Abstract

Accurate segmentation of vertebral metastasis in CT is clinically important yet difficult to scale, as voxel-level annotations are scarce and both lytic and blastic lesions often resemble benign degenerative changes. We introduce a 2D weakly supervised method trained solely on vertebra-level healthy/malignant labels, without any lesion masks. The method combines a Diffusion Autoencoder (DAE) that produces a classifier-guided healthy edit of each vertebra with pixel-wise difference maps that propose suspect candidate lesions. To determine which regions truly reflect malignancy, we introduce Hide-and-Seek Attribution: each candidate is revealed in turn while all others are hidden, the edited image is projected back to the data manifold by the DAE, and a latent-space classifier quantifies the isolated malignant contribution of that component. High-scoring regions form the final lytic or blastic segmentation. On held-out radiologist annotations, we achieve strong blastic/lytic performance despite no mask supervision (F1: 0.91/0.85; Dice: 0.87/0.78), exceeding baselines (F1: 0.79/0.67; Dice: 0.74/0.55). These results show that vertebra-level labels can be transformed into reliable lesion masks, demonstrating that generative editing combined with selective occlusion supports accurate weakly supervised segmentation in CT.