# Fast and robust quantitative MRI with tissue-susceptibility mapping of the human brain > **NIH NIH K99** · STANFORD UNIVERSITY · 2024 · $103,441 ## Abstract Project Summary/Abstract: T2* mapping and quantitative susceptibility mapping (QSM) are vital for in vivo iron quantification. They can track the subtle changes in tissue compositions during early brain development; they can provide valuable characteristics of lesions in Multiple Sclerosis (MS) for structural, pathological, and dynamic information. T2* and QSM are commonly estimated using multi-echo gradient echo (GRE), which suffers from long scan time (5-10 minutes) and sensitivity to motion and B0 perturbations, significantly limiting its clinical accessibility and impeding its potential for fine tissue characterization towards submillimeter resolution. This project aims to develop and validate a rapid, distortion- and blurring-free, and motion- and B0-robust technique for submillimeter-resolution T2* and QSM quantification, and translate it to applications on challenging populations, including infants and MS patients, demonstrating its potential for wide neuroscientific and clinical applications. The goal is fulfilled through three Specific Aims: In Aim 1, we will develop and optimize a highly efficient spherical-coverage echo-planar time-resolved imaging (sEPTI) framework for whole-brain distortion- and blurring-free T2* and QSM quantification. We will leverage a subspace-based unrolled deep-learning network for SNR-boosting reconstruction to achieve a 4x reduction in scan time on top of the already-fast state-of-the-art EPTI techniques. In Aim 2, we will develop a 3.5-ms multi-channel SPINS-trajectory navigator and combine it with the sEPTI technique as nasEPTI to achieve accurate motion and δB! estimation per TR with 70 ms latency. A supervised deep learning network will be developed to achieve fast and accurate estimation of motion and δB! in <5ms. Taken together, we will develop a synergistic per-TR prospective motion correction and retrospective B0 correction pipeline based on the rapid deep learning inference to accomplish a robust nasEPTI technique for artifact-minimized T2* weighted images, and T2* and QSM quantification. In Aim 3, nasEPTI will first be validated on 20 motion-prone infant subjects on 3-Tesla scanners with 0.7mm3 isotropic resolution and 1-minute scan time to demonstrate its robustness to motion and B0 perturbations. In parallel, this technique will be translated to 7-Tesla systems for a protocol of 0.35 mm isotropic resolution and 6- minute scan time. Finally, we will validate the 7T nasEPTI protocol on clinically suspected MS patients with cortical lesions, where T2* and QSM provide critical pathological information about the lesions. The expected outcome is that the proposed techniques achieve distortion- and blurring-free, motion- and B0-robust high-resolution T2* and QSM quantification. They are consistent with conventional multi-echo GRE in terms of T2* and QSM values, but provide enhanced image quality with significantly reduced artifacts. This will enable mesoscale iron characterization within a scientif... ## Key facts - **NIH application ID:** 10985536 - **Project number:** 1K99EB035178-01A1 - **Recipient organization:** STANFORD UNIVERSITY - **Principal Investigator:** Nan Wang - **Activity code:** K99 (R01, R21, SBIR, etc.) - **Funding institute:** NIH - **Fiscal year:** 2024 - **Award amount:** $103,441 - **Award type:** 1 - **Project period:** 2024-08-01 → 2026-07-31 ## Primary source NIH RePORTER: https://reporter.nih.gov/project-details/10985536 ## Citation > US National Institutes of Health, RePORTER application 10985536, Fast and robust quantitative MRI with tissue-susceptibility mapping of the human brain (1K99EB035178-01A1). Retrieved via AI Analytics 2026-05-24 from https://api.ai-analytics.org/grant/nih/10985536. Licensed CC0. --- *[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*