Non-invasive measurement of intracardiac pressures in heart failure patients using machine learning techniques

Non-invasive measurement of intracardiac pressures in heart failure patients using machine learning techniques

van Ravensberg, Annemiek (TU Delft Mechanical, Maritime and Materials Engineering)

van der Boon, Robert (mentor)
Bruining, Nico (mentor)
Scholte, Niels (mentor)
van der Velde, Enno (graduation committee)
van Deutekom, Arend (graduation committee)

Delft University of Technology
Universiteit Leiden
Erasmus Universiteit Rotterdam

Technical Medicine | Sensing and Stimulation

2023-04-18

Background: Solutions targeting early recognition of congestion in heart failure (HF) patients have the potential to prevent readmissions and can thus significantly reduce the burden on HF care. The gold standard measure of congestion is invasively measured pulmonary capillary wedge pressure (PCWP). However, the invasive nature and accessibility of this measurement limits its clinical use. Non-invasive approximation of the PCWP using biosensing wearables could be a promising replacement for HF monitoring.
Purpose: The primary aim of this retrospective study was to create a model that estimates the PCWP based on non-invasive measurements of vital signs using both traditional statistics and machine learning (ML) techniques.

Methods: The study cohort comprised right-sided heart catheterisations between 23/6/2017 and 19/8/2022 performed in the Erasmus MC, Rotterdam, The Netherlands. The following models were used: linear regression or classification, k-nearest neighbours, random forest, gradient boosting, and multilayer perceptron. The outcome measure for the regression models was the continuous PCWP as measured during the catheterisation. The two outcome classes for the classification models were low (<12 mmHg) and high (≥12 mmHg) PCWP. Non-invasive mean arterial blood pressure (MAP), saturation, heart rate, weight and temperature measured at most 72 hours before or after the catheterisation were collected as the features for the models, as well as the age and gender of the patient. Additionally, ECG-signals acquired during the catheterisation were used to calculate the heart rate variability (HRV). The data was split into a validation (20%) and training (80%) data set. The models were built based on the training set and then applied on the validation set to determine the coefficients of determination (R2) for the regression models and the area under the curve (AUC) for the classification models.

Results: A total of 853 catheterisation patients were included of which 31.3% had HF as primary diagnosis and 48.7% had a PCWP of 12 mmHg or higher. The average age of the cohort was 58.9 ± 13.7 years and 52.1% were male. The HRV had the highest correlation with the PCWP with a correlation of 0.16. All the regression models resulted in low R2 values of up to 0.11 and the classification models in AUC values of up to 0.59.

Conclusion: In the current study, PCWP could not be approximated with non-invasive measurements using traditional statistics and ML techniques. These findings support the notion that traditional measures for monitoring HF are poorly correlated with hemodynamic parameters. Perhaps repeated measurements over time, e.g. trends, or continuously measured signals such as photoplethysmography (PPG) could overcome the shortcomings of a single vital signs measurement for congestion evaluation. Therefore, future prospective research is needed to evaluate the potential of wearable devices measuring trends based on PPG signals in predicting hemodynamics.

Heart failure
Wedge pressure
Non-invasive
Machine Learning

http://resolver.tudelft.nl/uuid:c6674517-a0c8-42fd-81a1-1be6323dd60b

Student theses

master thesis

© 2023 Annemiek van Ravensberg