Scholay

学术搜索 · AI 审稿 · LaTeX 协作

QSM-CI method: χ-separation (iLSQR) (v2)

作者:Hyeong-Geol Shin, Jingu Lee, Jongho Lee · 发表于:Open MIND · 年份:2026 · DOI:10.5281/zenodo.21697748 · 研究领域:Computer science、Artificial intelligence、Physics、Mathematics

χ-separation (iLSQR variant) — separates net susceptibility into paramagnetic (χ+, iron) and diamagnetic (χ−, myelin/calcium) sources using the local field and an R2' map to break the para/dia degeneracy. The QSM is reconstructed in-house with STI-Suite QSM_iLSQR from the local field (the toolbox needs a QSM in its own scaling/orientation conventions — see recon.m). From the SNU-LIST chi-separation toolbox. A classical iterative baseline (χ+ xsim ~0.58 / χ− ~0.47 on the phantom). QSM-CI reconstruction method chi-sep-ilsqr. Browse and run it at https://qsmxt.github.io/QSM-CI/submission.html?method=chi-sep-ilsqr.