A Transformer-Based Network for Anisotropic 3D Medical Image Segmentation
Danfeng Guo, Demetri Terzopoulos
Abstract
Imaging anisotropy poses a critical challenge in applying deep learning models to 3D medical image analysis. Anisotropy downgrades model performance, especially when slice spacing varies significantly between training and clinical datasets. We propose a transformer-based model to tackle the anisotropy problem. It is adaptable to different levels of anisotropy and is computationally efficient. Our model outperforms baseline models in 3D lung cancer segmentation experiments.
Topics & Concepts
AnisotropyTransformerSegmentationComputer scienceArtificial intelligenceImage segmentation3d modelComputer visionEngineeringOpticsPhysicsElectrical engineeringVoltageRadiomics and Machine Learning in Medical ImagingAI in cancer detectionMedical Imaging and Analysis