CT Synthesis with Conditional Diffusion Models for Abdominal Lymph Node Segmentation
This work addresses the problem of limited annotated data for abdominal lymph node segmentation in medical imaging, which is crucial for computer-aided diagnosis, though it is incremental as it builds on existing diffusion models and segmentation methods.
The paper tackled the challenge of segmenting abdominal lymph nodes in medical images by developing a pipeline that uses a conditional diffusion model to generate synthetic lymph node data, which improved segmentation performance when combined with nnU-Net, achieving better results than other generative methods in synthesis and downstream segmentation tasks.
Despite the significant success achieved by deep learning methods in medical image segmentation, researchers still struggle in the computer-aided diagnosis of abdominal lymph nodes due to the complex abdominal environment, small and indistinguishable lesions, and limited annotated data. To address these problems, we present a pipeline that integrates the conditional diffusion model for lymph node generation and the nnU-Net model for lymph node segmentation to improve the segmentation performance of abdominal lymph nodes through synthesizing a diversity of realistic abdominal lymph node data. We propose LN-DDPM, a conditional denoising diffusion probabilistic model (DDPM) for lymph node (LN) generation. LN-DDPM utilizes lymph node masks and anatomical structure masks as model conditions. These conditions work in two conditioning mechanisms: global structure conditioning and local detail conditioning, to distinguish between lymph nodes and their surroundings and better capture lymph node characteristics. The obtained paired abdominal lymph node images and masks are used for the downstream segmentation task. Experimental results on the abdominal lymph node datasets demonstrate that LN-DDPM outperforms other generative methods in the abdominal lymph node image synthesis and better assists the downstream abdominal lymph node segmentation task.