# Elementary Neuronal Ensembles to Whole Brain Networks:  Ultrahigh Resolution Imaging of Function and Connectivity in Humans

> **NIH NIH U01** · UNIVERSITY OF MINNESOTA · 2022 · $150,381

## Abstract

Project Summary
This supplement aims to disseminate the NORDIC (NOise Reduction with DIstribution
Corrected) PCA method for reducing the influence of thermal noise in neuroimaging data.
NORDIC uses a dedicated processing approach to ensure that the noise component is
additive with independent, identically distributed, zero-mean Gaussian entries. Using this
characterization, results from random matrix theory can be efficiently used to devise a
parameter-free objective threshold. For NORDIC, this threshold value is both numerically
quantifiable and descriptive as the removal of components which cannot be distinguished from
Gaussian noise. NORDIC, unlike other methods, uses known information from the acquisition
to transform the data to fit the algorithm instead of either estimating the necessary information
or adapting the algorithm to fit the data. This approach for denoising is unique from previous
methods as it has negligible, if any, impact on real MR signals and can be more generally
applied to different types of MRI data without re-calibration or optimization. This, in turn, allows
for much higher resolutions and/or reduced scan times of otherwise SNR-starved MRI
protocols. The cost of the method is no more than having a clean sampling of the noise and
modest computational requirements, which could be implemented into an online acquisition
protocol and reconstruction pipeline for any protocol from any MRI scanner. The NORDIC
code will be made available to users and we will solicit feedback from a group of end users
with the goal of further optimizing and testing the denoising technique.

## Key facts

- **NIH application ID:** 10476050
- **Project number:** 3U01EB025144-05S1
- **Recipient organization:** UNIVERSITY OF MINNESOTA
- **Principal Investigator:** KAMIL UGURBIL
- **Activity code:** U01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2022
- **Award amount:** $150,381
- **Award type:** 3
- **Project period:** 2017-09-30 → 2023-06-30

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10476050

## Citation

> US National Institutes of Health, RePORTER application 10476050, Elementary Neuronal Ensembles to Whole Brain Networks:  Ultrahigh Resolution Imaging of Function and Connectivity in Humans (3U01EB025144-05S1). Retrieved via AI Analytics 2026-05-25 from https://api.ai-analytics.org/grant/nih/10476050. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*