The underlying idea was first laid out in a granted patent by Murray, Dingchang Zheng, Clive Griffiths and Chengyu Liu, and then validated in a 2025 research paper in which micropulses were identified by manual, visual analysis of recorded signals. The new publication takes the crucial next step: full automation. The physiological event at the heart of the technique is subtle but well defined. During a standard cuff measurement, the cuff is inflated above systolic pressure to cut off blood flow and then deflated slowly, at roughly 2 mmHg per second in this study. At the instant when arterial pressure exceeds cuff pressure, the artery snaps open and blood begins to flow downstream. That opening produces a sudden change in the gradient of the rising cuff pulse pressure, and with it a tiny high-frequency signature, the micropulse, that is present only for heartbeats occurring between systole and diastole.

The engineering implementation is elegant in its economy. A conventional blood pressure cuff with two pneumatic tubes is wrapped around the upper left arm exactly as in any routine measurement. One tube serves inflation and deflation and feeds a standard low-frequency pressure sensor that tracks cuff pressure and the slow pressure changes of each heartbeat. The second tube connects to a high-frequency pressure sensor with a flat response between 20 and 300 Hz, dedicated to capturing the micropulse events. The low-frequency signal is sampled at 2 kHz and the micropulse channel at 6 kHz. A stethoscope bound lightly below the cuff records Korotkoff sounds, not for the device itself but to provide the auscultatory reference values against which the new technique is judged. Because the frequency separation is handled in software, the team found that the choice of specific sensors mattered far less than getting the frequency ranges right.

The automation algorithm exploits two consistent features of the micropulse sequence. First, the slowly changing characteristics of the micropulses between systole and diastole are used to identify where the sequence begins and ends, separating genuine events from unwanted noise. Second, each micropulse is timed relative to the start of its simultaneous low-frequency beat pulse. During cuff deflation this timed interval shortens steadily from one beat to the next, falling from about 200 milliseconds to about 50 milliseconds, which gives the algorithm a built-in physiological consistency check. Systole is always detected near the peak of the low-frequency pulse, while diastole appears close to the rising start of the pulse, and therefore at a shorter interval. The first micropulse detected during deflation marks systolic pressure; the last marks diastolic pressure. Every beat between, and inclusive of, those two endpoints is identified, and the cuff pressure at each event is read off directly.

To test the device, the researchers recruited 32 healthy volunteers, 19 women and 13 men aged 24 to 68 with a mean age of 39, none of whom had known cardiovascular disease. These were the same subjects used in the earlier manual-validation study, deliberately chosen so the automated device could be compared like-for-like with the published hand-analysed results. Ethical approval came from the Newcastle & North Tyneside Research Ethics Committee, all participants gave signed informed consent, and the investigation conformed to the Declaration of Helsinki. Reference measurements were taken by auscultation with two trained operators, whose readings were averaged to reduce variability, and each subject’s data were recorded twice with a four-minute gap between sessions.

The results are striking. Correlation between the automated device and the auscultatory reference was 0.976 for systolic pressure and 0.993 for diastolic pressure, both with p values below 0.000001. The mean paired differences were -2.5 plus or minus 3.1 mmHg for systolic and -1.0 plus or minus 1.2 mmHg for diastolic pressure, comfortably within the international standard’s clinical limits of 5 plus or minus 8 mmHg. Bland-Altman limits of agreement ran from -8.60 to 3.58 mmHg for systolic and -3.34 to 1.32 mmHg for diastolic pressure. The device also tracked the small, expected fall in systolic pressure between the two sequential measurements, which the authors attribute to subjects relaxing during the four-minute interval, while diastolic pressure remained unchanged, exactly as auscultation showed.

Perhaps most tellingly, the automated device agreed closely with the team’s own published manual analysis of the same micropulse recordings, differing by -1.2 plus or minus 2.1 mmHg for systolic and 1.1 plus or minus 1.6 mmHg for diastolic pressure. The authors are candid about the limits of their reference method: invasive intra-arterial measurement would ideally be better than auscultation, but auscultation remains the accepted clinical gold standard, is used to establish normal and clinical ranges in drug trials, and no ethical permission or practical environment existed for invasive work in this study. They also stress that this is not a formal validation of a market-ready product, which would require larger and more diverse subject groups under the relevant international standard.

The deeper significance of the work lies in what it could change about how blood pressure devices are built and trusted. Because the technique identifies exactly when systole and diastole occur, and therefore the exact cuff pressure at those moments, the only calibration the device needs is of its cuff pressure sensor against a known reference pressure. No clinical subjects are required for calibration, in contrast to oscillometric devices, which each manufacturer must calibrate against its own private set of auscultatory measurements. That removes the proprietary-algorithm problem at a stroke and allows users, including clinics, to check and recalibrate a device themselves on a regular schedule. The authors argue that their technique therefore delivers true measurements rather than estimates, overcoming the documented unreliability of oscillometric devices and answering calls from reviewers, such as the 2017 meta-analysis of 76 studies by Picone and colleagues, for stronger accuracy standards to improve cardiovascular risk management.

The clinical stakes are high. The UK Medicines and Healthcare products Regulatory Agency has acknowledged the limitations of automated oscillometric devices and recommended that calibrated, non-mercury devices which do not rely on oscillometry be made available in all clinical areas. Advocates of manual measurement, including Grim and Grim, have warned that failure to assure the most accurate possible blood pressure reading could expose practitioners to charges of medical negligence. The Newcastle team, whose work was funded by the UK Engineering and Physical Sciences Research Council, says it is now ready to collaborate with industry, echoing the view of Ringrose and colleagues that greater collaboration between academia and industry is an important first step. The prototype itself is a research instrument, not a commercial product, and future work will include invasive studies of the micropulse origin, which may be as brief as a few milliseconds for systole at the peak of the arterial pressure pulse, and trials in larger and more varied clinical populations. But the core demonstration stands: a first automated implementation of micropulse detection has measured blood pressure directly, within a single routine cuff measurement, with errors smaller than the pressure change of a single heartbeat.

Subject of Research: Automated micropulse-based non-invasive blood pressure measurement technology

Article Title: Implementation of the first practical automatic micropulse blood pressure measurement device using the published revolutionary Newcastle technique

Article References: Implementation of the first practical automatic micropulse blood pressure measurement device using the published revolutionary Newcastle technique. (n.d.). https://doi.org/10.1007/s11517-026-03687-w

Image Credits: AI Generated

DOI: 10.1007/s11517-026-03687-w

Keywords: blood pressure, micropulse technique, oscillometric devices, medical device, hypertension, auscultation, biomedical engineering, Newcastle University, cuff measurement, diagnostic devices, clinical accuracy, ISO standard