Fast Multi-Stack Slice-to-Volume Reconstruction via Multi-Scale Unrolled Optimization
This addresses the need for efficient and accurate 3D reconstruction from misaligned 2D acquisitions in medical imaging, particularly for fetal brain MRI, with potential applications in real-time scanner-side feedback.
The paper tackles the problem of slice-to-volume reconstruction (SVR) for medical imaging by introducing a fast convolutional framework that fuses multiple 2D slice stacks to recover 3D structure and refines slice alignment, achieving high-quality 3D volume reconstruction in under 10 seconds with accuracy on par with state-of-the-art methods.
Fully convolutional networks have become the backbone of modern medical imaging due to their ability to learn multi-scale representations and perform end-to-end inference. Yet their potential for slice-to-volume reconstruction (SVR), the task of jointly estimating 3D anatomy and slice poses from misaligned 2D acquisitions, remains underexplored. We introduce a fast convolutional framework that fuses multiple orthogonal 2D slice stacks to recover coherent 3D structure and refines slice alignment through lightweight model-based optimization. Applied to fetal brain MRI, our approach reconstructs high-quality 3D volumes in under 10s, with 1s slice registration and accuracy on par with state-of-the-art iterative SVR pipelines, offering more than speedup. The framework uses non-rigid displacement fields to represent transformations, generalizing to other SVR problems like fetal body and placental MRI. Additionally, the fast inference time paves the way for real-time, scanner-side volumetric feedback during MRI acquisition.