Predicting Volume Responsiveness Among Sepsis Patients Using Clinical Data and Continuous Physiological Waveforms

Overview

Journal AMIA Annu Symp Proc

Specialty Medical Informatics

Date 2021 May 3

PMID 33936436

Citations 4

Authors

Rishikesan Kamaleswaran

Jiaoying Lian

Dong-Lien Lin

Himasagar Molakapuri

SriManikanth Nunna

Parth Shah

Shiv Dua

Rema Padman

Affiliations

Soon will be listed here.

Abstract

The efficacy of early fluid treatment in patients with sepsis is unclear and may contribute to serious adverse events due to fluid non-responsiveness. The current method of deciding if patients are responsive to fluid administration is often subjective and requires manual intervention. This study utilizes MIMIC III and associated matched waveform datasets across the entire ICU stay duration of each patient to develop prediction models for assessing fluid responsiveness in sepsis patients. We developed a pipeline to extract high frequency continuous waveform data and included waveform features in the prediction models. Comparing across five machine learning models, random forest performed the best when no waveform information is added (AUC = 0.84), with mean arterial blood pressure and age identified as key factors. After incorporation of features from physiologic waveforms, logistic regression with L1 penalty provided consistent performance and high interpretability, achieving an accuracy of 0.89 and F1 score of 0.90.

Citing Articles

Machine learning derived serum creatinine trajectories in acute kidney injury in critically ill patients with sepsis.

Takkavatakarn K, Oh W, Chan L, Hofer I, Shawwa K, Kraft M Crit Care. 2024; 28(1):156.

PMID: 38730421 PMC: 11084026. DOI: 10.1186/s13054-024-04935-x.

Improving the precision of shock resuscitation by predicting fluid responsiveness with machine learning and arterial blood pressure waveform data.

Gupta C, Basu D, Williams T, Neff L, Johnson M, Patel N Sci Rep. 2024; 14(1):2227.

PMID: 38278825 PMC: 10817926. DOI: 10.1038/s41598-023-50120-5.

Predicting patient decompensation from continuous physiologic monitoring in the emergency department.

Sundrani S, Chen J, Jin B, Shakeri Hossein Abad Z, Rajpurkar P, Kim D NPJ Digit Med. 2023; 6(1):60.

PMID: 37016152 PMC: 10073111. DOI: 10.1038/s41746-023-00803-0.

Evaluating machine learning models for sepsis prediction: A systematic review of methodologies.

Deng H, Sun M, Wang Y, Zeng J, Yuan T, Li T iScience. 2022; 25(1):103651.

PMID: 35028534 PMC: 8741489. DOI: 10.1016/j.isci.2021.103651.

References

Christov I . Real time electrocardiogram QRS detection using combined adaptive threshold. Biomed Eng Online. 2004; 3(1):28. PMC: 516783. DOI: 10.1186/1475-925X-3-28. View

Kamaleswaran R, Akbilgic O, Hallman M, West A, Davis R, Shah S . Applying Artificial Intelligence to Identify Physiomarkers Predicting Severe Sepsis in the PICU. Pediatr Crit Care Med. 2018; 19(10):e495-e503. DOI: 10.1097/PCC.0000000000001666. View

Singer M, Deutschman C, Seymour C, Shankar-Hari M, Annane D, Bauer M . The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016; 315(8):801-10. PMC: 4968574. DOI: 10.1001/jama.2016.0287. View

Komorowski M, Celi L, Badawi O, Gordon A, Faisal A . The Artificial Intelligence Clinician learns optimal treatment strategies for sepsis in intensive care. Nat Med. 2018; 24(11):1716-1720. DOI: 10.1038/s41591-018-0213-5. View

Thiel S, Kollef M, Isakow W . Non-invasive stroke volume measurement and passive leg raising predict volume responsiveness in medical ICU patients: an observational cohort study. Crit Care. 2009; 13(4):R111. PMC: 2750155. DOI: 10.1186/cc7955. View

Boyd J, Forbes J, Nakada T, Walley K, Russell J . Fluid resuscitation in septic shock: a positive fluid balance and elevated central venous pressure are associated with increased mortality. Crit Care Med. 2010; 39(2):259-65. DOI: 10.1097/CCM.0b013e3181feeb15. View

Azur M, Stuart E, Frangakis C, Leaf P . Multiple imputation by chained equations: what is it and how does it work?. Int J Methods Psychiatr Res. 2011; 20(1):40-9. PMC: 3074241. DOI: 10.1002/mpr.329. View

Lamia B, Ochagavia A, Monnet X, Chemla D, Richard C, Teboul J . Echocardiographic prediction of volume responsiveness in critically ill patients with spontaneously breathing activity. Intensive Care Med. 2007; 33(7):1125-1132. DOI: 10.1007/s00134-007-0646-7. View

Peng X, Ding Y, Wihl D, Gottesman O, Komorowski M, Lehman L . Improving Sepsis Treatment Strategies by Combining Deep and Kernel-Based Reinforcement Learning. AMIA Annu Symp Proc. 2019; 2018:887-896. PMC: 6371300. View

10.

Mohammed A, Cui Y, Mas V, Kamaleswaran R . Differential gene expression analysis reveals novel genes and pathways in pediatric septic shock patients. Sci Rep. 2019; 9(1):11270. PMC: 6677896. DOI: 10.1038/s41598-019-47703-6. View

11.

Nemati S, Holder A, Razmi F, Stanley M, Clifford G, Buchman T . An Interpretable Machine Learning Model for Accurate Prediction of Sepsis in the ICU. Crit Care Med. 2017; 46(4):547-553. PMC: 5851825. DOI: 10.1097/CCM.0000000000002936. View

12.

Johnson A, Pollard T, Shen L, Lehman L, Feng M, Ghassemi M . MIMIC-III, a freely accessible critical care database. Sci Data. 2016; 3:160035. PMC: 4878278. DOI: 10.1038/sdata.2016.35. View

13.

Monnet X, Bataille A, Magalhaes E, Barrois J, Le Corre M, Gosset C . End-tidal carbon dioxide is better than arterial pressure for predicting volume responsiveness by the passive leg raising test. Intensive Care Med. 2012; 39(1):93-100. DOI: 10.1007/s00134-012-2693-y. View

14.

Johnson A, Aboab J, Raffa J, Pollard T, Deliberato R, Celi L . A Comparative Analysis of Sepsis Identification Methods in an Electronic Database. Crit Care Med. 2018; 46(4):494-499. PMC: 5851804. DOI: 10.1097/CCM.0000000000002965. View

15.

Shashikumar S, Stanley M, Sadiq I, Li Q, Holder A, Clifford G . Early sepsis detection in critical care patients using multiscale blood pressure and heart rate dynamics. J Electrocardiol. 2017; 50(6):739-743. PMC: 5696025. DOI: 10.1016/j.jelectrocard.2017.08.013. View

16.

Wu M, Ghassemi M, Feng M, Celi L, Szolovits P, Doshi-Velez F . Understanding vasopressor intervention and weaning: risk prediction in a public heterogeneous clinical time series database. J Am Med Inform Assoc. 2016; 24(3):488-495. PMC: 6455107. DOI: 10.1093/jamia/ocw138. View

17.

Monnet X, Marik P, Teboul J . Prediction of fluid responsiveness: an update. Ann Intensive Care. 2016; 6(1):111. PMC: 5114218. DOI: 10.1186/s13613-016-0216-7. View

18.

Joynes E . More challenges around sepsis: definitions and diagnosis. J Thorac Dis. 2017; 8(11):E1467-E1469. PMC: 5179457. DOI: 10.21037/jtd.2016.11.10. View

19.

Goldberger A, Amaral L, Glass L, Hausdorff J, Ivanov P, Mark R . PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals. Circulation. 2000; 101(23):E215-20. DOI: 10.1161/01.cir.101.23.e215. View

20.

van Wyk F, Khojandi A, Mohammed A, Begoli E, Davis R, Kamaleswaran R . A minimal set of physiomarkers in continuous high frequency data streams predict adult sepsis onset earlier. Int J Med Inform. 2019; 122:55-62. DOI: 10.1016/j.ijmedinf.2018.12.002. View