Devil is in Details: Locality-Aware 3D Abdominal CT Volume Generation for Self-Supervised Organ Segmentation
Open Access
- 28 October 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 10640-10648
Abstract
In the realm of medical image analysis, self-supervised learning (SSL) techniques have emerged to alleviate labeling demands, while still facing the challenge of training data scarcity owing to escalating resource requirements and privacy constraints. Numerous efforts employ generative models to generate high-fidelity, unlabeled 3D volumes across diverse modalities and anatomical regions. However, the intricate and indistinguishable anatomical structures within the abdomen pose a unique challenge to abdominal CT volume generation compared to other anatomical regions. To address the overlooked challenge, we introduce the Locality-Aware Diffusion (Lad), a novel method tailored for exquisite 3D abdominal CT volume generation. We design a locality loss to refine crucial anatomical regions and devise a condition extractor to integrate abdominal priori into generation, thereby enabling the generation of large quantities of high-quality abdominal CT volumes essential for SSL tasks without the need for additional data such as labels or radiology reports. Volumes generated through our method demonstrate remarkable fidelity in reproducing abdominal structures, achieving a decrease in FID score from 0.0034 to 0.0002 on AbdomenCT-1K dataset, closely mirroring authentic data and surpassing current methods. Extensive experiments demonstrate the effectiveness of our method in self-supervised organ segmentation tasks, resulting in an improvement in mean Dice scores on two abdominal datasets effectively. These results underscore the potential of synthetic data to advance self-supervised learning in medical image analysis.Keywords
All Related Versions
Funding Information
- The National Natural Science Foundation of China (62171325)
- the Hubei Key R&D Project (2022BAA033)
This publication has 21 references indexed in Scilit:
- Generating Realistic Brain MRIs via a Conditional Diffusion Probabilistic Model2023
- Denoising diffusion probabilistic models for 3D medical image generationScientific Reports, 2023
- Long Video Generation with Time-Agnostic VQGAN and Time-Sensitive Transformer2022
- BT-Unet: A self-supervised learning framework for biomedical image segmentation using barlow twins with U-net modelsMachine Learning, 2022
- Self-supervised learning in medicine and healthcareNature Biomedical Engineering, 2022
- Contrastive Self-supervised Representation Learning Using Synthetic DataInternational Journal of Automation and Computing, 2021
- Analyzing and Improving the Image Quality of StyleGAN2020
- Generation of 3D Brain MRI Using Auto-Encoding Generative Adversarial Networks2019
- Multiscale structural similarity for image quality assessment2004
- Multidimensional Scaling1998