Team introduces a noninvasive method to monitor postprandial cardiovascular health

The dynamics of blood nutrient and lipid levels after consuming a high-fat meal are crucial indicators of both current and future cardiovascular health. Traditionally, measuring these circulating substances has involved invasive blood draws, which are not feasible for regular health tracking.

Researchers are exploring noninvasive methods to assess cardiovascular health, which could improve monitoring of postprandial effects and help identify factors that contribute to cardiovascular disease. A promising approach is a noncontact optical imaging technique called “spatial frequency domain imaging” (SFDI), which quantifies tissue properties and hemodynamics.

A recent study by researchers from Boston University, Harvard Medical School, and Brigham and Women’s Hospital has investigated how meal composition affects skin tissue properties shortly after eating. As reported in Biophotonics Discovery, the research team focused on the peripheral tissue of the hand to understand the immediate impacts of low-fat and high-fat meals.

Using SFDI, the researchers monitored 15 subjects who consumed both types of meals on separate days. The team imaged the back of each subject’s hand hourly for five hours post-meal, analyzing three specific wavelengths to evaluate hemoglobin, water, and lipid concentrations.

The results revealed significant differences in tissue responses. The high-fat meal led to an increase in tissue oxygen saturation, while the low-fat meal caused a decrease, suggesting that dietary fat can affect not just overall health but also immediate physiological responses. The peak changes occurred three hours after eating, coinciding with spikes in triglyceride levels.

Alongside imaging, researchers tracked blood pressure and heart rate, and also performed blood draws to measure triglycerides, cholesterol, and glucose levels. The results indicated that the optical absorption changes at specific wavelengths accurately correspond to variations in lipid concentrations.

Building on these insights, the team then trained a machine learning model using SFDI data to predict triglyceride levels, achieving an accuracy within 40 mg/dL. This precision could pave the way for noninvasive monitoring of cardiovascular health.

Senior author Darren Roblyer, professor of biomedical engineering at Boston University, remarks, “The research suggests that SFDI could serve as a promising alternative, allowing for easier monitoring of how meals affect cardiovascular health.” He adds, “Overall, these findings highlight the intricate relationship between diet, body response, and cardiovascular risk, suggesting a need for further exploration of non-invasive assessment methods.”