Caltech’s laser headset becomes 1st-ever device to predict stroke noninvasively

A team of engineers and scientists from Caltech and the Keck School of Medicine at USC has created a noninvasive, headset-based device that can assess a patient’s stroke risk by monitoring blood flow and volume changes during a breath-holding test.

The device uses a laser-based system and has demonstrated promising results in distinguishing between individuals with low and high stroke risk.

“With this device, for the first time, we are going to have a way of knowing if the risk of someone having a stroke in the future is significant or not based on a physiological measurement,” says Simon Mahler, a co-lead author of the study.

Stroke predictions using laser

The Caltech team has developed a compact, wearable device that uses infrared laser light to assess brain blood flow through a technique called speckle contrast optical spectroscopy (SCOS). The device shines laser light through the skull, and a specialized camera captures the scattered light after it interacts with blood vessels in the brain.

SCOS measures the decrease in light intensity from its entry point to where it is collected, determining blood volume in the vessels. Additionally, it tracks how light scatters into speckles, which fluctuate based on the rate of blood flow—the faster the blood flow, the quicker the speckle patterns change.

By analyzing these measurements, the researchers calculate the ratio of blood flow to volume, helping them gauge stroke risk in patients. In a study with 50 participants, the SCOS technique enabled the researchers to assess how much the blood vessels expand and how much blood flow increases during a breath-holding test, offering valuable insights into each participant’s stroke risk.

“These reactive measurements are indicative of vessel stiffness,” Yang says. “Our technology makes it possible to make these type of measurements noninvasively for the first time.”

Integrating machine learning

“What we found is clear, striking evidence of a different reaction of blood flow and blood volume between the two groups,” says Yu Xi Huang, a co-lead author of the study, in the press release .

In the low-stroke-risk group, researchers observed a smaller increase in blood flow during the breath-holding test compared to the high-stroke-risk group, but a larger increase in blood volume. This suggests that the blood vessels in low-risk individuals are more capable of widening to allow greater blood flow.

The team is now expanding their research by testing the current prototype on a larger and more diverse group of patients at a hospital in Visalia, California. They also plan to integrate machine learning into the device to enhance data analysis.

Additionally, they aim to conduct a clinical trial involving long-term patient tracking over two years to further refine the technology. Ultimately, the researchers hope the device will not only prescreen for stroke risk but also help pinpoint the location of strokes within the brain.

The study has been published in Biomedical Optics Express .