Noninvasive continuous blood pressure monitoring using infrared spectroscopy and machine learning

JUL 25, 2025

Diffuse correlation spectroscopy and a machine learning algorithm can provide continuous blood pressure feedback, enabling real-time monitoring during surgery and for at-risk patients.

Avery Thompson

DOI: 10.1063/10.0038652

Noninvasive continuous blood pressure monitoring using infrared spectroscopy and machine learning internal name — Noninvasive continuous blood pressure monitoring using infrared spectroscopy and machine learning lead image

Abnormal blood pressure is one of the primary indicators of many common illnesses and medical conditions, making it crucial data for healthcare providers. For patients suffering from high blood pressure or cardiovascular diseases, or for patients undergoing surgery, continuous blood pressure monitoring can quickly detect dangerous fluctuations and avert serious complications. However, existing continuous blood pressure monitoring is invasive, painful, and can lead to complications of its own.

Li et al. developed an alternative, noninvasive method for continuous blood pressure monitoring using diffuse correlation spectroscopy (DCS) coupled with machine learning.

DCS is a technique that relies on near-infrared light to directly image blood flow. While blood flow is not the same as blood pressure, they are related, and the authors trained a machine learning model to correlate the two. They demonstrated the effectiveness of this method by accurately measuring the blood pressure of 12 human subjects.

“In our tests, the results showed that the continuous blood pressure monitoring achieved Grade A for both systolic and diastolic blood pressure in accordance with the British Hypertension Society standards,” said author Zhe Li.

The authors plan to expand their sample size and continue to develop their machine learning model to further increase its accuracy. They will also explore integrating other physiological data into the analysis to improve their results.

“In the future, this method may be applied to portable devices, enabling patients to monitor their blood pressure in real time in daily life,” said Li.

Source: “Continuous noninvasive blood pressure estimation using tissue blood flow measured by diffuse correlation spectroscopy,” by Zhe Li, Jiangtao Bai, Xiangyu Cao, Xing Chen, Haiqing Song, Peng Tian, Ran Wei, Jinchao Feng, Pengyu Liu, and Kebin Jia, APL Bioengineering (2025). The article can be accessed at https://doi.org/10.1063/5.0266243 .