# Label-free measurement of blood lipids with hyperspectral short-wave infrared spatial frequency domain imaging to improve cardiovascular disease risk prediction and treatment monitoring

> **NIH NIH R21** · BOSTON UNIVERSITY (CHARLES RIVER CAMPUS) · 2021 · $209,212

## Abstract

ABSTRACT:
Cardiovascular disease (CVD) reportedly causes 31% of global deaths with overwhelming personal and
economic costs, the latter estimated as $400 billion annually in the US. Lack of regular screening contributes to
severe undertreatment for CVD, with some estimates finding only 1/3 of eligible patients taking preventative
medications. One of the most important biomarkers for CVD, elevated blood lipids, requires invasive blood draws
followed by lab-based testing. These requirements limit access to screening for many at-risk patients in low
resource settings, and make frequent longitudinal monitoring impractical for everyone. Worse still, recent data
shows that the temporal dynamics of blood lipids, including postprandial increases, circadian, ultradian (<24 hr
cycles), and fluctuations during menstrual cycle, all negatively affect the accuracy of blood lipid test results. Non-
invasive methods that more immediately and continuously track blood-lipids would transform CVD monitoring for
those at risk. The goal of this project is to develop the first non-invasive optical technology for measuring blood
lipids. To accomplish this goal, we will develop a new imaging technique called Short-Wave Infrared Spatial
Frequency Domain Imaging (SWIR-SFDI). SWIR-SFDI leverages spectroscopic SWIR patterned illumination
combined with model-based analysis to extract tissue optical properties as well as lipid and water concentrations.
Compared to both visible (VIS) and near infrared (NIR) imaging, the SWIR wavelength band potentially provides
better quantification of lipids and deeper imaging. In this project, we propose to develop SWIR-SFDI
instrumentation and processing methodology to demonstrate that triglycerides (TG), cholesterol, LDL-C, and
HDL-C, all of which are strong predictors of CVD risk, can be tracked longitudinally with high accuracy. We will
fabricate a mobile hyperspectral SWIR-SFDI system with an innovative dual digital micromirror device (DMD)
configuration for rapid wavelength tuning and spatial light patterning between 700 – 1600 nm at high speed.
Performance benchmarks include: spectral resolution <6 nm FWHM, 10-𝜆 acquisition < 5 sec, μa and μs´ errors
<3%, drift (7 hr) <2%. We will also develop a 2-layer skin-model to improve in vivo blood lipid quantification and
develop methodology for spectral classification and quantification of TG, cholesterol, LDL-C, and HDL-C and
test the accuracy of these algorithms in simulation and by fabricating 3-D printed vascularized phantoms with a
range of skin types. Finally, we will conduct a normal volunteer feasibility study to assess the ability of SWIR-
SFDI to accurately track postprandial lipids in comparison to traditional laboratory measurement from invasive
blood draws. Completion of these aims will enable our team to progress to a larger R01-funded, hypothesis-
driven clinical study following the period of this project. Over the longer term, SWIR-SFDI has the potential to
transform CVD lipid ...

## Key facts

- **NIH application ID:** 10178014
- **Project number:** 5R21EB030197-02
- **Recipient organization:** BOSTON UNIVERSITY (CHARLES RIVER CAMPUS)
- **Principal Investigator:** Darren Michael Roblyer
- **Activity code:** R21 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2021
- **Award amount:** $209,212
- **Award type:** 5
- **Project period:** 2020-07-01 → 2023-03-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10178014

## Citation

> US National Institutes of Health, RePORTER application 10178014, Label-free measurement of blood lipids with hyperspectral short-wave infrared spatial frequency domain imaging to improve cardiovascular disease risk prediction and treatment monitoring (5R21EB030197-02). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10178014. Licensed CC0.

---

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