Cancer classification using machine learning and HRV analysis: preliminary evidence from a pilot study

Marta Vigier,Benjamin Vigier,Elisabeth Andritsch,Andreas R Schwerdtfeger

doi:10.1038/s41598-021-01779-1

Marta Vigier, Benjamin Vigier + Show 2 more

Open Access

https://doi.org/10.1038/s41598-021-01779-1

Copy DOI

Abstract

Most cancer patients exhibit autonomic dysfunction with attenuated heart rate variability (HRV) levels compared to healthy controls. This research aimed to create and evaluate a machine learning (ML) model enabling discrimination between cancer patients and healthy controls based on 5-min-ECG recordings. We selected 12 HRV features based on previous research and compared the results between cancer patients and healthy individuals using Wilcoxon sum-rank test. Recursive Feature Elimination (RFE) identified the top five features, averaged over 5 min and employed them as input to three different ML. Next, we created an ensemble model based on a stacking method that aggregated the predictions from all three base classifiers. All HRV features were significantly different between the two groups. SDNN, RMSSD, pNN50%, HRV triangular index, and SD1 were selected by RFE and used as an input to three different ML. All three base-classifiers performed above chance level, RF being the most efficient with a testing accuracy of 83%. The ensemble model showed a classification accuracy of 86% and an AUC of 0.95. The results obtained by ML algorithms suggest HRV parameters could be a reliable input for differentiating between cancer patients and healthy controls. Results should be interpreted in light of some limitations that call for replication studies with larger sample sizes.

Highlights

Most cancer patients exhibit autonomic dysfunction with attenuated heart rate variability (HRV) levels compared to healthy controls
This pilot study evaluated the possibility of machine learning-based discrimination between cancer patients and healthy controls based on five-minute-ECG recordings
Despite the moderate improvement in classification accuracy of the ensemble model, it should be noted that the performance of our model proved satisfactory as compared with previous research, showing 86% accuracy, 93% specificity and 77% sensitivity while classifying cancer vs healthy individuals on unseen data

Summary

Introduction

Most cancer patients exhibit autonomic dysfunction with attenuated heart rate variability (HRV) levels compared to healthy controls. Abbreviations AUC Area under the curve ECG Electrocardiogram FFT Fast Fourier transform RFE Recursive features elimination RF Random forest HRV Heart rate variability IBI Interbeat interval ML Machine learning XGB EXtreme gradient boosting NB Naïve Bayes LDA Linear discrimination analysis. It has been reported that the vagus nerve may exercise a neuromodulatory influence on cancer by slowing tumour development and p rogression[5,9]. These authors inferred that vagal influences might reduce oxidative stress, modulate inflammation, and inhibit sympathetic activity. HRV parameters can be described by linear (time- and frequency-domain) and nonlinear measures[17]

Objectives

Methods

Results

Discussion

Conclusion