From Variability To Accuracy: Conditional Bernoulli Diffusion Models with Consensus-Driven Correction for Thin Structure Segmentation
This work addresses the need for precise segmentation in medical imaging for customized implants and surgical planning, representing an incremental advancement over existing methods.
The paper tackles the problem of accurately segmenting orbital bones in facial CT images, which is challenging due to ambiguous boundaries and thin structures, by proposing a framework that uses a conditional Bernoulli diffusion model and consensus-driven correction to improve segmentation results, achieving significant recall improvements in ambiguous regions.
Accurate segmentation of orbital bones in facial computed tomography (CT) images is essential for the creation of customized implants for reconstruction of defected orbital bones, particularly challenging due to the ambiguous boundaries and thin structures such as the orbital medial wall and orbital floor. In these ambiguous regions, existing segmentation approaches often output disconnected or under-segmented results. We propose a novel framework that corrects segmentation results by leveraging consensus from multiple diffusion model outputs. Our approach employs a conditional Bernoulli diffusion model trained on diverse annotation patterns per image to generate multiple plausible segmentations, followed by a consensus-driven correction that incorporates position proximity, consensus level, and gradient direction similarity to correct challenging regions. Experimental results demonstrate that our method outperforms existing methods, significantly improving recall in ambiguous regions while preserving the continuity of thin structures. Furthermore, our method automates the manual process of segmentation result correction and can be applied to image-guided surgical planning and surgery.