Non-invasive waveform analysis for emergency triage via simulated hemorrhage: An experimental study using novel dynamic lower body negative pressure model

• Autorzy: Nesaragi Naimahmed , Høiseth Lars Øivind , Qadir Hemin Ali , Rosseland Leiv Arne , Halvorsen Per Steinar , Balasingham Ilangko

Identyfikatory

DOI

10.1016/j.bbe.2023.06.002

• Języki publikacji: EN

Abstrakty

EN

The extent to which advanced waveform analysis of non-invasive physiological signals can diagnose levels of hypovolemia remains insufficiently explored. The present study explores the discriminative ability of a deep learning (DL) framework to classify levels of ongoing hypovolemia, simulated via novel dynamic lower body negative pressure (LBNP) model among healthy volunteers. We used a dynamic LBNP protocol as opposed to the traditional model, where LBNP is applied in a predictable step-wise, progressively descending manner. This dynamic LBNP version assists in circumventing the problem posed in terms of time dependency, as in real-life pre-hospital settings intravascular blood volume may fluctuate due to volume resuscitation. A supervised DL-based framework for ternary classification was realized by segmenting the underlying noninvasive signal and labeling segments with corresponding LBNP target levels. The proposed DL model with two inputs was trained with respective time–frequency representations extracted on waveform segments to classify each of them into blood volume loss: Class 1 (mild); Class 2 (moderate); or Class 3 (severe). At the outset, the latent space derived at the end of the DL model via late fusion among both inputs assists in enhanced classification performance. When evaluated in a 3-fold cross-validation setup with stratified subjects, the experimental findings demonstrated PPG to be a potential surrogate for variations in blood volume with average classification performance, AUROC: 0.8861, AUPRC: 0.8141, F1-score:72.16%, Sensitivity:79.06%, and Specificity:89.21%. Our proposed DL algorithm on PPG signal demonstrates the possibility to capture the complex interplay in physiological responses related to both bleeding and fluid resuscitation using this challenging LBNP setup.

Słowa kluczowe

EN

lower body negative pressure; blood loss; photoplethysmography; noninvasive arterial waveform analysis; deep learning; time frequency analysis

PL

strata krwi; fotopletyzmografia; uczenie głębokie; analiza czasowo-częstotliwościowa

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2023
• Tom: Vol. 43, no. 3
• Strony: 551--567
• Opis fizyczny: Bibliogr. 68 poz., rys., tab., wykr.

Twórcy

autor

Nesaragi Naimahmed

• The Intervention Centre, Oslo University Hospital, Rikshospitelet Sognsvannsveien 20, 0372 Oslo, Norway

autor

Høiseth Lars Øivind

• Division of Emergencies and Critical Care, Department of Anesthesiology, Oslo University Hospital, Norway

autor

Qadir Hemin Ali

• The Intervention Centre, Oslo University Hospital, Rikshospitalet, Oslo, Norway

autor

Rosseland Leiv Arne

• Division of Emergencies and Critical Care, Department of Anesthesiology, Oslo University Hospital, Norway
• Institute of Clinical Medicine, Faculty of Medicine, University of Oslo, Oslo, Norway
• Department of Research and Development, Division of Emergencies and Critical Care, Oslo University Hospital, Norway

autor

Halvorsen Per Steinar

• The Intervention Centre, Oslo University Hospital, Rikshospitalet, Oslo, Norway
• Institute of Clinical Medicine, Faculty of Medicine, University of Oslo, Oslo, Norway

autor

Balasingham Ilangko

• The Intervention Centre, Oslo University Hospital, Rikshospitalet, Oslo, Norway
• Department of Electronic Systems, Norwegian University of Science and Technology Trondheim, Norway

Bibliografia

• [1] Convertino VA, Cardin S. Advanced medical monitoring for the battlefield: a review on clinical applicability of compensatory reserve measurements for early and accurate hemorrhage detection. J Trauma Acute Care Surg 2022;93(2S): S147-54.
• [2] Chen Y, Hong C, Pinsky MR, Ma T, Clermont G. Estimating surgical blood loss volume using continuously monitored vital signs. Sensors 2020;20(22):6558.
• [3] Convertino VA, Techentin RW, Poole RJ, Dacy AC, Carlson AN, Cardin S, et al. AI-enabled advanced development for assessing low circulating blood volume for emergency medical care: comparison of compensatory reserve machine-learning algorithms. Sensors 2022;22(7):2642.
• [4] Convertino VA, Moulton SL, Grudic GZ, Rickards CA, Hinojosa-Laborde C, Gerhardt RT, et al. Use of advanced machine-learning techniques for noninvasive monitoring of hemorrhage. J Trauma Acute Care Surg 2011;71(1):S25-32.
• [5] Van Der Ster BJ, Bennis FC, Delhaas T, Westerhof BE, Stok WJ, Van Lieshout JJ. Support vector machine based monitoring of cardio-cerebrovascular reserve during simulated hemorrhage. Front Physiol 2018;8:1057.
• [6] Ji SY, Belle A, Ward KR, Ryan KL, Rickards CA, Convertino VA, et al. Heart rate variability analysis during central hypovolemia using wavelet transformation. J Clini Monitor Comput 2013;27(3):289-302.
• [7] Chew MS, Ðneman A. Haemodynamic monitoring using arterial waveform analysis. Curr Opin Critical Care 2013;19 (3):234-41.
• [8] Convertino VA, Grudic G, Mulligan J, Moulton S. Estimation of individual-specific progression to impending cardiovascular instability using arterial waveforms. J Appl Physiol 2013;115 (8):1196-202.
• [9] Convertino VA, Wirt MD, Glenn JF, Lein BC. The compensatory reserve for early and accurate prediction of hemodynamic compromise: a review of the underlying physiology. Shock 2016;45(6):580-90.
• [10] Suresh MR. The early detection of hypovolemic shock and shifting the focus to compensation. Los Angeles, CA: SAGE Publications Sage CA; 2022.
• [11] Davies SJ, Vistisen ST, Jian Z, Hatib F, Scheeren TW. Ability of an arterial waveform analysis–derived hypotension prediction index to predict future hypotensive events in surgical patients. Anesthesia Analgesia 2020;130(2):352-9.
• [12] Hatib F, Jian Z, Buddi S, Lee C, Settels J, Sibert K, et al. Machine-learning algorithm to predict hypotension based on high-fidelity arterial pressure waveform analysis. Anesthesiology 2018;129(4):663-74.
• [13] Dziuda Ł, Krej M, Śmietanowski M, Sobotnicki A, Sobiech M, Kwaśny P, et al. Development and evaluation of a novel system for inducing orthostatic challenge by tilt tests and lower body negative pressure. Sci Rep 2018;8(1):7793.
• [14] Hansen LH, Bülow K. Feasibility of a continuous non-invasive arterial pressure (CNAP) device in a prehospital setting. Resuscitation 2014;85:S88-9.
• [15] Scully CG, Selvaraj N, Romberg FW, Wardhan R, Ryan J, Florian JP, et al. Using time-frequency analysis of the photoplethysmographic waveform to detect the withdrawal of 900 mL of blood. Anesthesia Analgesia 2012;115(1):74-81.
• [16] Selvaraj N, Scully CG, Shelley KH, Silverman DG, Chon KH. Early detection of spontaneous blood loss using amplitude modulation of photoplethysmogram. 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE; 2011. p. 5499-502.
• [17] Liang Y, Chen Z, Ward R, Elgendi M. Hypertension assessment via ECG and PPG signals: An evaluation using MIMIC database. Diagnostics 2018;8(3):65.
• [18] Mukkamala R, Hahn JO, Inan OT, Mestha LK, Kim CS, Tö reyin H, et al. Toward ubiquitous blood pressure monitoring via pulse transit time: theory and practice. IEEE Trans Biomed Eng 2015;62(8):1879-901.
• [19] Djupedal H, Nøstdahl T, Hisdal J, Landsverk SA, Høiseth LØ. Effects of experimental hypovolemia and pain on preejection period and pulse transit time in healthy volunteers. Physiol Rep 2022;10(12):e15355.
• [20] Elgendi M, Liang Y, Ward R. Toward generating more diagnostic features from photoplethysmogram waveforms. Diseases 2018;6(1):20.
• [21] Chen Y, Yoon JH, Pinsky MR, Ma T, Clermont G. Development of hemorrhage identification model using non-invasive vital signs. Physiol Measur 2020;41(5):055010.
• [22] Pinsky MR, Wertz A, Clermont G, Dubrawski A. Parsimony of hemodynamic monitoring data sufficient for the detection of hemorrhage. Anesthesia Analgesia 2020;130(5):1176.
• [23] Elgendi M, Norton I, Brearley M, Abbott D, Schuurmans D. Systolic peak detection in acceleration photoplethysmograms measured from emergency responders in tropical conditions. PloS One 2013;8(10):e76585.
• [24] Elgendi M, Norton I, Brearley M, Abbott D, Schuurmans D. Detection of a and b waves in the acceleration photoplethysmogram. Biomed Eng Online 2014;13(1):1-18.
• [25] Elgendi M. Optimal signal quality index for photoplethysmogram signals. Bioengineering 2016;3(4):21.
• [26] Mejía-Mejía E, Budidha K, Kyriacou PA, Mamouei M. Comparison of pulse rate variability and morphological features of photoplethysmograms in estimation of blood pressure. Biomed Signal Process Control 2022;78:103968.
• [27] Pybus DA. Real-time, spectral analysis of the arterial pressure waveform using a wirelessly-connected, tablet computer: a pilot study. J Clin Monitor Comput 2019;33(1):53-63.
• [28] Goswami N, Blaber AP, Hinghofer-Szalkay H, Convertino VA. Lower body negative pressure: physiological effects, applications, and implementation. Physiol Rev 2019;99 (1):807-51.
• [29] Goswami N. Compensatory hemodynamic changes in response to central hypovolemia in humans: lower body negative pressure: updates and perspectives. J Muscle Res Cell Motility 2022:1-6.
• [30] Park FS, Kay VL, Sprick JD, Rosenberg AJ, Anderson GK, Mallet RT, et al. Hemorrhage simulated by lower body negative pressure provokes an oxidative stress response in healthy young adults. Exp Biol Med 2019;244(3):272-8.
• [31] Fagoni N, Bruseghini P, Adami A, Capelli C, Lador F, Moia C, et al. Effect of lower body negative pressure on phase I cardiovascular responses at exercise onset. Int J Sports Med 2020;41(04):209-18.
• [32] Neumann S, Burchell AE, Rodrigues JC, Lawton CB, Burden D, Underhill M, et al. Cerebral blood flow response to simulated hypovolemia in essential hypertension: a magnetic resonance imaging study. Hypertension 2019;74(6):1391-8.
• [33] Cai M, Wang H, Kline G, Ding Y, Ross SE, Davis S, et al. Habitual physical activity improves vagal cardiac modulation and carotid baroreflex function in elderly women. Exp Biol Med 2023. 15353702231160334.
• [34] Wang L, Li Z, Tan C, Wang H, Zhou X, He S, et al. Lower-body negative pressure/ergometer exercise in bed rest: Effects on female orthostatic tolerance. J Novel Physiother Rehabili 2020;4(2):040-8.
• [35] Jain M, Chitturi V, Chandran DS, Jaryal AK, Deepak K. Vasoconstriction during non-hypotensive hypovolemia is not associated with activation of baroreflex: A causality-based approach. Pflügers Archiv-European. J Physiol 2023:1-9.
• [36] Jonsson H, Henricson J, Saager RB, Wilhelms D. Microcirculatory response to lower body negative pressure and the association to large vessel function. In: Photonics in Dermatology and Plastic Surgery 2023. vol. 12352. SPIE; 2023. p. 49-53.
• [37] Harris KM, Petersen LG, Weber T. Reviving lower body negative pressure as a countermeasure to prevent pathological vascular and ocular changes in microgravity. npj Microgravity 2020;6(1):38.
• [38] Van Akin M, Lantz O, Fellows A, Toutain-Kidd C, Zegans M, Buckey JC, et al. Acute effects of postural changes and lower body positive and negative pressure on the eye. Front Physiol 2022:1685.
• [39] Gallardo P, Giannakopoulos I, Romare M, Karanika P, Elcadi G, et al. The Effect of Upright Lower Body Negative Pressure on Muscle Activity and Hemodynamics during Exercise. Austin Sports Med 2023;8(1):1053.
• [40] Kenny JÉS, Elfarnawany M, Yang Z, Eibl AM, Eibl JK, Kim CH, et al. A wireless ultrasound patch detects mild-to-moderate central hypovolemia during lower body negative pressure. J Trauma Acute Care Surg 2022;93(2):S35.
• [41] Kenny JÉS, Barjaktarevic I, Mackenzie DC, Elfarnawany M, Yang Z, Eibl AM, et al. Carotid artery velocity time integral and corrected flow time measured by a wearable Doppler ultrasound detect stroke volume rise from simulated hemorrhage to transfusion. BMC Res Notes 2022;15(1):1-5.
• [42] Kenny JE´S, Elfarnawany M, Yang Z, Myers M, Eibl AM, Eibl JK, et al. The Doppler shock index measured by a wearable ultrasound patch accurately detects moderate-to-severe central hypovolemia during lower body negative pressure. J Am College Emergency Physic Open 2021;2(4):e12533.
• [43] Raj KV, Nabeel P, Chandran D, Sivaprakasam M, Joseph J. High-frame-rate A-mode ultrasound for calibration-free cuffless carotid pressure: feasibility study using lower body negative pressure intervention. Blood Pressure 2022;31 (1):19-30.
• [44] Kimball JP, Inan OT, Convertino VA, Cardin S, Sawka MN. Wearable Sensors and Machine Learning for Hypovolemia Problems in Occupational, Military and Sports Medicine: Physiological Basis, Hardware and Algorithms. Sensors 2022;22(2):442.
• [45] Berboth L, Zirngast B, Manninger M, Steendijk P, Tschöpe C, Scherr D, et al. Graded lower body negative pressure induces intraventricular negative pressures and incremental diastolic suction: a pressure-volume study in a porcine model. J Appl Physiol 2022;133(1):20-6.
• [46] Rosenberg AJ, Kay VL, Anderson GK, Sprick JD, Rickards CA. A comparison of protocols for simulating hemorrhage in humans: step versus ramp lower body negative pressure. J Appl Physiol 2021;130(2):380-9.
• [47] Techentin RW, Felton CL, Schlotman TE, Gilbert BK, Joyner MJ, Curry TB, et al. 1D Convolutional neural networks for estimation of compensatory reserve from blood pressure waveforms. 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE; 2019. p. 2169-73.
• [48] Killick R, Fearnhead P, Eckley IA. Optimal detection of changepoints with a linear computational cost. J Am Stat Assoc 2012;107(500):1590-8.
• [49] Ji SY. Computer-aided Trauma Decision Making Using Machine Learning and Signal Processing. 2008.
• [50] Nesaragi N, Sharma A, Patidar S, Majumder S, Tavakolian K. Application of recurrent neural network for the prediction of target non-apneic arousal regions in physiological signals. In: 2018 Computing in Cardiology Conference (CinC). vol. 45; 2018. p. 1-4.
• [51] Elola A, Aramendi E, Oliveira J, Renna F, Coimbra MT, Reyna MA, et al. Beyond Heart Murmur Detection: Automatic Murmur Grading from Phonocardiogram. arXiv; 2022. Available from: https://arxiv.org/abs/2209.13385.
• [52] Pham TD. Time-frequency time-space LSTM for robust classification of physiological signals. Sci Rep 2021;11(1):6936.
• [53] Buttkus B. Spectral analysis and filter theory in applied geophysics. Springer Science & Business Media; 2012.
• [54] Salvi M, Acharya UR, Molinari F, Meiburger KM. The impact of pre-and post-image processing techniques on deep learning frameworks: A comprehensive review for digital pathology image analysis. Comput Biol Med 2021;128:104129.
• [55] Shaffer F, Ginsberg JP. An overview of heart rate variability metrics and norms. Front Public Health 2017;258.
• [56] Flood MW, Grimm B. EntropyHub: An open-source toolkit for entropic time series analysis. Plos One 2021;16(11):e0259448.
• [57] Li T, Zhou M. ECG classification using wavelet packet entropy and random forests. Entropy 2016;18(8):285.
• [58] Miao F, Fu N, Zhang YT, Ding XR, Hong X, He Q, et al. A novel continuous blood pressure estimation approach based on data mining techniques. IEEE J Biomed Health Informat 2017;21(6):1730-40.
• [59] Miao F, Liu ZD, Liu JK, Wen B, He QY, Li Y. Multi-sensor fusion approach for cuff-less blood pressure measurement. IEEE J Biomed Health Informat 2019;24(1):79-91.
• [60] van der Ster BJ, Westerhof BE, Stok WJ, van Lieshout JJ. Detecting central hypovolemia in simulated hypovolemic shock by automated feature extraction with principal component analysis. Physiol Rep 2018;6(22):e13895.
• [61] Fadil R, Verma AK, Sadeghian F, Blaber AP, Tavakolian K. Cardio-respiratory interactions in response to lower-body negative pressure. Physiol Measur 2023;44(2):025005.
• [62] Kugener G, Zhu Y, Pangal DJ, Sinha A, Markarian N, Roshannai A, et al. Deep neural networks can accurately detect blood loss and hemorrhage control task success from video. Neurosurgery 2022;90(6):823-9.
• [63] Gupta JF, Arshad SH, Telfer BA, Snider EJ, Convertino VA. Noninvasive monitoring of simulated hemorrhage and whole blood resuscitation. Biosensors 2022;12(12):1168.
• [64] Bedolla CN, Gonzalez JM, Vega SJ, Convertino VA, Snider EJ. An explainable machine-learning model for compensatory reserve measurement: methods for feature selection and the effects of subject variability. Bioengineering 2023;10(5):612.
• [65] Chalumuri YR, Kimball JP, Mousavi A, Zia JS, Rolfes C, Parreira JD, et al. Classification of blood volume decompensation state via machine learning analysis of multi-modal wearable-compatible physiological signals. Sensors 2022;22(4):1336.
• [66] Finnegan E, Davidson S, Harford M, Watkinson P, Tarassenko L, Villarroel M. Features from the photoplethysmogram and the electrocardiogram for estimating changes in blood pressure. Sci Rep 2023;13(1):986.
• [67] Eid AM, Elgamal M, Gonzalez-Fiol A, Shelley KH, Wu HT, Alian AA. Using the ear photoplethysmographic waveform as an early indicator of central hypovolemia in healthy volunteers utilizing LBNP induced hypovolemia model. Physiol Meas 2023.
• [68] Alian A, Shelley K, Wu HT. Amplitude and phase measurements from harmonic analysis may lead to new physiologic insights: lower body negative pressure photoplethysmographic waveforms as an example. J Clin Monitor Comput 2023;37(1):127-37.