Reproduce “NestedMorph: Enhancing Deformable Medical Image Registration with Nested Attention Mechanisms” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Gurucharan Marthi Krishna Kumar, Janine Mendola, Amir Shmuel

NestedMorph aligns same-subject T1-weighted MRI with diffusion MRI by predicting a deformation field, fusing encoder context with decoder detail at multiple scales. On 25 held-out HCP Aging subjects after affine pre-alignment, it reports SSIM 0.89 ± 0.06, HD95 2.5 ± 0.3, and SDlogJ 0.22 ± 0.02, versus TransMorph at 0.88 ± 0.07, 2.8 ± 0.4, and 0.25 ± 0.02. These results concern intra-subject registration; no significance test was reported.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.