Cardiac output estimation using ballistocardiography: a feasibility study in healthy subjects

Q: What is the clinical evidence from this study?

Study design: Observational. Population: Healthy subjects (n=20). Intervention: Ballistocardiography (BCG) based machine learning model vs. Invasive arterial blood pressure based CO (pulse contour analysis). Primary outcome: Mean absolute error of cardiac output estimation in the validation set (L/min) (95% CI -2.18-1.89).

Q: What does this research mean for the field?

A machine learning model utilizing morphological features from ballistocardiography signals can estimate cardiac output non-invasively with a mean absolute error of 0.83 L/min and a percentage error of 30.2% compared to invasive monitoring. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Johannes Nordsteien Svensøy; Erik Alonso; Andoni Elola; Reidar Bjørnerheim; Johan Ræder; Elisabete Aramendi; Lars Wik

doi:10.1038/s41598-024-52300-3

January 19, 2024Scientific ReportsOpen Access

Cardiac output estimation using ballistocardiography: a feasibility study in healthy subjects

View Full Paper

Ask AI

Bookmark

Share

Key result

Ballistocardiography machine learning estimates cardiac output with ~30% error compared to invasive monitoring.

95% CI -2.18-1.89
n=20

Figures

Why the study?

Prehospital emergency medicine lacks a reliable automated non-invasive method for monitoring circulation, and conventional cardiac output monitors are not feasible or practical in that setting.

Does non-invasive ballistocardiography accurately estimate cardiac output compared to invasive arterial blood pressure monitoring in healthy subjects?

Study Design

Type

Observational (n=20)

Multicenter

No

Structured PICO

Does non-invasive ballistocardiography accurately estimate cardiac output compared to invasive arterial blood pressure monitoring in healthy subjects?

P

Population

20 healthy adult subjects, median age 25, 70% female.

I

Intervention

Non-invasive ballistocardiography (BCG) using piezoelectric biosensors placed on the skin over the carotid artery and abdominal aorta, processed with a machine learning model to estimate cardiac output.

C

Comparator

Invasive arterial blood pressure-based cardiac output monitoring (FloTrac sensor connected to a radial arterial line and HemoSphere monitor).

O

Outcome

Accuracy of cardiac output estimation measured by mean absolute error, percentage error, and 95% limits of agreement in the validation set.surrogate

Non-invasive ballistocardiography combined with machine learning shows feasibility for estimating cardiac output, offering a potential future tool for prehospital emergency monitoring.

Limitations

Small sample size of 20 healthy subjects without compromised blood flow.
Measurements during the hyperventilation phase were excluded due to movement artifacts.
The gold standard used was pulse wave contour analysis rather than thermodilution.
Small sample size (20 subjects, with only 5 in the validation set)
Model trained on normal cardiac output values, leading to overestimation of low CO and underestimation of high CO
Susceptibility to movement artifacts, which negatively affected recordings during hyperventilation and ambulance transport phases

Implication

Supports further development of non-invasive cardiac output tools; leaves open validation in patients and clinical settings.

Abstract

There is no reliable automated non-invasive solution for monitoring circulation and guiding treatment in prehospital emergency medicine. Cardiac output (CO) monitoring might provide a solution, but CO monitors are not feasible/practical in the prehospital setting. Non-invasive ballistocardiography (BCG) measures heart contractility and tracks CO changes. This study analyzed the feasibility of estimating CO using morphological features extracted from BCG signals. In 20 healthy subjects ECG, carotid/abdominal BCG, and invasive arterial blood pressure based CO were recorded. BCG signals were adaptively processed to isolate the circulatory component from carotid (CCc) and abdominal (CCa) BCG. Then, 66 features were computed on a beat-to-beat basis to characterize amplitude/duration/area/length of the fluctuation in CCc and CCa. Subjects' data were split into development set (75%) to select the best feature subset with which to build a machine learning model to estimate CO and validation set (25%) to evaluate model's performance. The model showed a mean absolute error, percentage error and 95% limits of agreement of 0.83 L/min, 30.2% and - 2.18-1.89 L/min respectively in the validation set. BCG showed potential to reliably estimate/track CO. This method is a promising first step towards an automated, non-invasive and reliable CO estimator that may be tested in prehospital emergencies.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Svensøy et al. (2024) conducted an observational in Healthy subjects (n=20). Ballistocardiography (BCG) based machine learning model vs. Invasive arterial blood pressure based CO (pulse contour analysis) was evaluated on Mean absolute error of cardiac output estimation in the validation set (L/min) (95% CI -2.18-1.89). A machine learning model using ballistocardiography features estimated cardiac output with a mean absolute error of 0.83 L/min and a percentage error of 30.2% compared to invasive monitoring.

synapsesocial.com/papers/6a189355985da83d54918215 https://doi.org/10.1038/s41598-024-52300-3

View Full Paper

Ask AI

Bookmark

Share

Cardiac output estimation using ballistocardiography: a feasibility study in healthy subjects

Figures

Study Design

Structured PICO

Limitations

Implication

Abstract

Discussion

Cite This Study

Cardiac output estimation using ballistocardiography: a feasibility study in healthy subjects

Figures

Study Design

Structured PICO

Limitations

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading