Application of neural networks for estimation of aortic systlic pressure from peripheral systolic and diastolic pressure

H. G. Xiao, A. Qasem, M. Butlin, A. P. Avolio

    Research output: Contribution to journalMeeting abstractpeer-review


    Objective: Methods for estimation of aortic systolic blood pressure (SBP) require recording of a peripheral pressure waveform, which requires specialised devices for brachial volumetric or radial tonometric pulse measurement. This study investigates the possibility of aortic SBP estimation from peripheral SBP and diastolic blood pressure (DBP) using artificial neural networks (ANN) with (ANN[SBP,DBP,HR]) and without (ANN[SBP,DBP]) heart rate (HR). These parameters were investigated as they are readily available in conventional brachial sphygmomanometry. As a proof of concept and to remove measurement error, the theory was investigated using invasive measurements of blood pressure. Design and method: Ten-fold cross validation was applied to invasive, simultaneously recorded aortic and radial pressure during rest and vasodilation following nitroglycerin (6g/kg/min)) infusion in subjects (n=62) drawn from a patient cohort previously reported. The results of the ANN models were compared to an ANN model using additional waveform features (ANNwaveform), to an N-point moving average method (NPMA) and to an existing, validated generalized transfer function (GTF). Results: Estimated aortic SBP for all methods was on average 1 mmHg different to measured aortic SBP with the exception of NPMA (difference 2.0 ± 3.5 mmHg, p=0.62). Variability of the difference was significantly greater in ANN[SBP,DBP,HR] and ANN[SBP,DBP] (both S.D. of ± 5.9 mmHg, p0.001 compared to the GTF method, ± 4.0 mmHg, p0.001). Inclusion of waveform features decreased the variability (ANNwaveform ± 3.9 mmHg, p=0.264). Estimated aortic SBP in all models was correlated with measured SBP, with ANN models providing statistically similar results to the GTF method, only the NPMA being statistically different (p=0.031). Conclusions These findings indicate that ANN applied to peripheral SBP, DBP and HR alone can provide aortic SBP estimation comparable to the GTF, albeit with slightly greater variance. Pending noninvasive validation using brachial sphygmomanometry techniques, the ANN method provides plausible aortic SBP estimation without waveform analysis, allowing potential inclusion in conventional brachial sphygmomanometer devices.
    Original languageEnglish
    Article numberPP.19.27
    Pages (from-to)e247
    Number of pages1
    JournalJournal of Hypertension
    Issue numbere-Supplement 2
    Publication statusPublished - 2017
    EventEuropean Meeting on Hypertension and Cardiovascular Protection (27th : 2017) - Milan, Italy
    Duration: 16 Jun 201719 Jun 2017


    Dive into the research topics of 'Application of neural networks for estimation of aortic systlic pressure from peripheral systolic and diastolic pressure'. Together they form a unique fingerprint.

    Cite this