SEOUL, March 05 (AJP) - Researchers at South Korea's Advanced Institute of Science and Technology have developed a wireless wearable electronic patch capable of measuring blood flow in real time. The device uses deep learning and multilayer thermal sensing to monitor cardiovascular health without invasive procedures. This technology provides a potential tool for the early detection of cardiovascular diseases and continuous monitoring of patients in clinical settings.
Blood flow serves as a critical indicator of cardiovascular health, with fluctuations often signaling conditions such as hypertension, arteriosclerosis, or even physical shock. While Doppler ultrasound is currently the clinical standard for measuring blood flow, its reliance on bulky equipment and trained medical professionals limits its use for continuous daily monitoring. Existing wearable thermal sensors have struggled with accuracy because the depth of blood vessels varies between individuals, distorting the thermal signals used to calculate flow speed.
The research team, led by Professor Kwon Kyung-ha of the School of Electrical Engineering at the Korea Advanced Institute of Science and Technology (KAIST), addressed this limitation by creating a multilayer thermal gradient sensing structure. The device features temperature sensors placed at different depths to analyze the three-dimensional movement of heat generated by flowing blood. By integrating a deep learning algorithm, the system can distinguish between the depth of the blood vessel and the actual speed of the blood flow simultaneously.
Experimental results showed that the patch can measure blood flow speeds between 1 and 10 millimeters per second with an error margin of less than 0.12 millimeters per second. It also determined blood vessel depths between 1 and 2 millimeters with an accuracy within 0.07 millimeters. When integrated with photoplethysmography (PPG) sensors commonly found in smartwatches, the system reduced blood pressure estimation errors by up to 72.6 percent compared to using PPG sensors alone. This improvement was particularly evident during the Valsalva maneuver, a breathing technique that causes rapid changes in blood pressure.
The platform was tested on human subjects through various physiological interventions, including breath-holding, external vascular compression, and cycling. In all scenarios, the blood flow measurements remained consistent with clinical perfusion index standards. The entire system is implemented as a wearable patch with Bluetooth Low Energy (BLE) capabilities, allowing for wireless transmission of biometric data to external devices for real-time analysis.
"This technology provides a fundamental platform for more accurate measurement of blood flow and blood pressure," said Professor Kwon Kyung-ha. "By combining this with smartwatches, we can significantly improve the quality of daily health monitoring."
The study, led by first author Sim Young-min, a student in the integrated master's and doctoral program, was supported by institutional funding and published in the journal Science Advances on February 6.
(Paper information)
Journal: Science Advances
Title: Deep learning–integrated multilayer thermal gradient sensing platform for real-time blood flow monitoring
DOI: https://doi.org/10.1126/sciadv.aea8902
Copyright ⓒ Aju Press All rights reserved.