• Home
  • Search
  • Advancing shock prediction: leveraging prior knowledge and self-controlled data for enhanced model accuracy and generalizability.
  • https://doi.org/10.1186/s12911-025-03108-2Copy DOI Icon

Advancing shock prediction: leveraging prior knowledge and self-controlled data for enhanced model accuracy and generalizability.

Show More
  • Abstract
  • Literature Map
  • References
  • Similar Papers
Abstract

Timely intervention in shock is vital, as delays over one hour greatly increase mortality. This study aims to develop an enhanced machine learning model that improves predictive performance by utilizing self-controlled data and applying feature engineering informed by medical knowledge to physiological waveforms, enabling the prediction of shock one hour in advance without relying on blood tests. Patient data and physiological waveforms were obtained from the Medical Information Mart for Intensive Care III (MIMIC-3) database. Shock was defined as a mean arterial pressure ≤ 65 mmHg for more than one minute, combined with serum lactate levels ≥ 2 mmol/L within 12h before or after the hypotension event. Waveforms used for prediction were extracted from 30min time-segment before a 1-hour period prior to the event. Self-controlled waveforms were obtained from the same patient either one day before or up to seven days after the shock event. The study included 389 ICU patients who met the shock criteria and had complete physiological waveform data available for analysis. A total of 299 features were derived: 90 from arterial blood pressure (ABP), 89 from electrocardiogram (ECG), 112 from respiratory waveforms (RESP), and 8 from blood oxygen saturation (SpO2). The weighted ensemble model showed the best performance with an AUC of 0.93 and accuracy of 84.15%, and sensitivity of 79.64% in the testing set. The most predictive features included ECG_HRV_pNN50 (proportion of successive heartbeat intervals differing by more than 50 ms), RESP_Width_Mean (mean width of respiratory waveform), RESP_Cycle_Rate_Mean (mean respiratory cycle rate), ABP_TimeSBP2DBP_SampEn (sample entropy of systolic-diastolic intervals), and ABP_AmplitudeDBP_Median (median amplitude of diastolic peaks). This study demonstrated the feasibility of predicting shock one hour before its onset using only four physiological waveforms, combined with feature engineering based on physiological concepts and self-sampling data. The model achieved a strong AUC and a high sensitivity. Not applicable.

Similar Papers
  • Research Article
  • Citations2

Assessment of the effect of two regimens of milrinone infusion in paediatric patients with pulmonary artery hypertension undergoing corrective cardiac procedure: A prospective observational study

  • Jan 01, 2022
  • Annals of Pediatric Cardiology
  • Mrugesh Prajapati +5
  • Research Article
  • Citations52

Blood pressure management in patients with acute ischemic stroke.

  • Jan 12, 2004
  • Hypertension (Dallas, Tex. : 1979)
  • Larry B Goldstein
  • PDF
  • Research Article
  • Citations4

A comparison of invasive arterial blood pressure measurement with oscillometric non-invasive blood pressure measurement in patients with sepsis.

  • Feb 02, 2024
  • Journal of Anesthesia
  • Ziqing Jiang +8
  • Discussion
  • Citations16

Hypertension Optimal Treatment (HOT) trial

  • Aug 01, 1998
  • The Lancet
  • Lennart Hansson +1
  • PDF
  • Research Article
  • Citations69

Imputation of the continuous arterial line blood pressure waveform from non-invasive measurements using deep learning

  • Aug 03, 2021
  • Scientific Reports
  • Brian L Hill +7
  • Research Article
  • Citations15

Comparison between hemodynamic changes after single-dose and incremental subarachnoid anesthesia.

  • Jul 01, 1996
  • Regional Anesthesia: The Journal of Neural Blockade in Obstetrics, Surgery, & Pain Control
  • A Casati +3
  • Research Article
  • Citations317

Arterial stiffness and hypertension: chicken or egg?

  • May 05, 2014
  • Hypertension
  • Gary F Mitchell
  • Research Article

The effective application of ventilation at prone position on patients with severe acute respiratory distress syndrome at high altitude

  • Mar 21, 2019
  • The Journal of practical nursing
  • Xiaolin Sun +3
  • Front Matter
  • Citations1

Hypertension Editors' Picks: Preeclampsia, Pregnancy, and Hypertension.

  • Jul 01, 2018
  • Hypertension
  • The Editors
  • Research Article
  • Citations7

Cardiopulmonary Hemodynamic Changes During Acute Subdural Hematoma Evacuation

  • Jan 01, 2006
  • Neurologia medico-chirurgica
  • Tomonori Tamaki +3
  • Research Article
  • Citations1

Short-Term Angiotensin Subtype 1 Receptor Blockade Does Not Alter the Circulatory Responses to Sympathetic Nervous System Modulation in Healthy Volunteers Before and During Sevoflurane Anesthesia: Results of a Pilot Study

  • Aug 11, 2016
  • Journal of Cardiothoracic and Vascular Anesthesia
  • Shahbaz R Arain +4
  • Research Article
  • Citations46

Uterine exteriorisation at caesarean section: influence on maternal morbidity.

  • Oct 01, 1998
  • BJOG: An International Journal of Obstetrics & Gynaecology
  • Edmond C O Edi‐Osagie +6
  • Research Article
  • Citations1

Identifying key covariates of clinical outcomes for critically ill patients with Parkinson's disease: analysis of the MIMIC-IV database.

  • Jul 03, 2025
  • Frontiers in neurology
  • Lihua Luo +4
  • Research Article
  • Citations10

Utility of microdialysis to detect the lactate/pyruvate ratio in subcutaneous tissue for the reliable monitoring of hemorrhagic shock

  • Jan 01, 2009
  • Journal of Smooth Muscle Research
  • Hitoshi Ohashi +7
  • Research Article
  • Citations18

Effects of labetalol on uterine blood flow and cardiovascular hemodynamics in the hypertensive gravid baboon

  • May 01, 1993
  • American Journal of Obstetrics and Gynecology
  • Mark A Morgan +4
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.