Related Experiment Video
Updated: May 21, 2025
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
Published on: September 25, 2019
FastSurfer-LIT: Lesion inpainting tool for whole-brain MRI segmentation with tumors, cavities, and abnormalities
Clemens Pollak1, David Kügler1, Tobias Bauer2,3
1AI in Medical Imaging, German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany.
Abstract:
Resection cavities, tumors, and other lesions can fundamentally alter brain structure and present as abnormalities in brain MRI. Specifically, quantifying subtle neuroanatomical changes in other, not directly affected regions of the brain is essential to assess the impact of tumors, surgery, chemo/radiotherapy, or drug treatments. However, only a limited number of solutions address this important task, while many standard analysis pipelines simply do not support abnormal brain images at all. In this paper, we present a method to perform sensitive neuroanatomical analysis of healthy brain regions in the presence of large lesions and cavities. Our approach called "FastSurfer Lesion Inpainting Tool" (FastSurfer-LIT) leverages the recently emerged Denoising Diffusion Probabilistic Models (DDPM) to fill lesion areas with healthy tissue that matches and extends the surrounding tissue. This enables subsequent processing with established MRI analysis methods such as the calculation of adjusted volume and surface measurements using FastSurfer or FreeSurfer. FastSurfer-LIT significantly outperforms previously proposed solutions on a large dataset of simulated brain tumors (N = 100) and synthetic multiple sclerosis lesions (N = 39) with improved Dice and Hausdorff measures, and also on a highly heterogeneous dataset with lesions and cavities in a manual assessment (N = 100). Finally, we demonstrate increased reliability to reproduce pre-operative cortical thickness estimates from corresponding post-operative temporo-mesial resection surgery MRIs. The method is publicly available at https://github.com/Deep-MI/LIT and will be integrated into the FastSurfer toolbox.
Insights
This study introduces the FastSurfer Lesion Inpainting Tool (FastSurfer-LIT), a novel method using Denoising Diffusion Probabilistic Models to realistically fill brain lesions in MRI scans. This enables accurate neuroanatomical analysis even with significant brain abnormalities.
Area of Science:
- Neuroimaging
- Medical Image Analysis
- Artificial Intelligence in Medicine
Background:
- Brain lesions from tumors or surgery significantly alter neuroanatomy, complicating MRI analysis.
- Standard analysis pipelines often fail with abnormal brain images, limiting assessment of treatment effects.
- Accurate quantification of neuroanatomical changes in unaffected brain regions is crucial.
Purpose of the Study:
- To develop a method for sensitive neuroanatomical analysis in the presence of large brain lesions.
- To enable the use of standard MRI analysis tools on images with significant abnormalities.
- To improve the assessment of subtle neuroanatomical changes after interventions or treatments.
Main Methods:
- Developed the FastSurfer Lesion Inpainting Tool (FastSurfer-LIT) utilizing Denoising Diffusion Probabilistic Models (DDPM).
- DDPMs were used to inpaint lesion areas with healthy-appearing brain tissue.
- The inpainted images were processed using established methods like FastSurfer/FreeSurfer for volumetric and surface measurements.
Main Results:
- FastSurfer-LIT outperformed previous methods on simulated tumors and MS lesions, showing improved Dice and Hausdorff measures.
- Manual assessment on a heterogeneous dataset with lesions and cavities also favored FastSurfer-LIT.
- Demonstrated increased reliability in reproducing pre-operative cortical thickness from post-operative MRIs.
Conclusions:
- FastSurfer-LIT effectively reconstructs healthy brain tissue in lesion areas, enabling robust neuroanatomical analysis.
- The method significantly improves the analysis of brain MRI scans affected by lesions and cavities.
- This tool enhances the reliability of neuroimaging assessments in clinical and research settings.
