Wearable electronic device for in-clinic or remote blood pressure monitoring Blood pressure (BP) is a key physiological indicator of health. It reflects the overall physical wellbeing of an individual, and high BP (affecting 32% of the US population) is a major risk factor for various medical conditions. To date, there is only one clinically validated cuff-less technique to measure BP. A small wearable device capable of accurately measuring BP would have a significant impact on the public health by enabling long-term monitoring of individuals at a large scale over the course of decades and identifying those at risk of hypertension early on. In addition, such a technology would allow for targeted monitoring of patients for whom BP fluctuations may be indicative of critical states such as hemodynamic instability.
Existing techniques that attempt to measure BP continuously use the principle of pulse wave velocity (the speed by which blood travels through the arteries) for cuff-less estimation of BP. These techniques suffer from multiple shortcomings that prohibit their use in commercial wearables: 1) they require two sensors far enough apart on different locations of the body to measure the pulse travel time (preventing simplicity and convenience); 2) they need to be frequently re-calibrated at least every few hours (diminishing practicality); and 3) they make inaccurate assumptions about certain physiological factors (e.g., vascular resistance), leading to imprecise estimates (hindering accuracy). Our invention addresses all these shortcomings and provides a practical tool to measure BP accurately, continuously and conveniently.
Researchers at the University of Michigan (UM) have invented a novel non-invasive sensor and analytical algorithms to continuously monitor BP in both healthy and disease patient populations. The sensor is small, inexpensive and low-power. It wraps around the finger like a ring and generates a waveform that is rich in information about the cardiovascular system, including BP. We have designed signal processing and machine learning tools that use the intrinsic features of this waveform to measure BP continuously and accurately. The sensor is also capable of providing information on vascular resistance and cardiac output.
The technology was tested in an animal model and shown to be highly effective for BP monitoring. The sensor was placed on the foreleg of five swine and several maneuvers were performed to induce large amounts of variation in their BP (ranging from 35/20 to 260/150 mmHg). A high average correlation of 0.94 (ranging from 0.90 to 0.98) was found between the true (clinical-grade invasively measured intra-arterial BP) and estimated pressure calculated from the ring data.
UM researchers also tested the technology during pulmonary artery catheterization (PAC) procedures in order to identify patients with low cardiac index (cardiac output normalized by body surface area). A model using animal data was built and tested on 61 patients undergoing PAC. An area under the receiver operating characteristic curve of 0.83 was achieved, indicating the effectiveness of the proposed technology for stratification of patients with low cardiac index.
Monitoring BP continuously during daily activity will have an enormous impact on patient care and experience. Cuff-based BP monitors are bulky and obtrusive, work non-continuously and are inaccurate. The UM ring sensor can transform patient care and have far-reaching impact on public health by enabling on-demand or continuous 24/7 monitoring of BP in a way that is reliable and convenient. This technology will have a positive impact on a large number of health conditions, including stroke, heart failure, hypertension and diabetes. Jeremy Nelson jernelso@umich.edu 734.936.2095