Leveraging the potential of machine learning for assessing vascular ageing: state-of-the-art and future research

Abstract

Vascular ageing biomarkers have been found to be predictive of cardiovascular risk independently of classical risk factors, yet are not widely used in clinical practice. In this review, we present two basic approaches for using machine learning (ML) to assess vascular age: parameter estimation and risk classification. We then summarize their role in developing new techniques to assess vascular ageing quickly and accurately. We discuss the methods used to validate ML-based markers, the evidence for their clinical utility, and key directions for future research. The review is complemented by case studies of the use of ML in vascular age assessment which can be replicated using freely available data and code.

Keywords: Arterial stiffness; Blood pressure; Cardiovascular; Central blood pressure; Machine learning; Pulse wave velocity.

Graphical abstract

Figure 1

Using machine learning to assess vascular ageing biomarkers from more easily obtained measurements. BMI, body mass index; CV, cardiovascular; , presence of CV event; , absence of CV event. Adapted from: ‘Adult male with organs’, Wikimedia Commons, under CC0 1.0.

Figure 2

Schematic representation of a random forest regression prediction.

Figure 3

A case study of estimating central systolic blood pressure and central diastolic blood pressure from age, brachial systolic and diastolic blood pressures, and heart rate using a random forest regressor. CDBP, central diastolic blood pressure; CSBP, central systolic blood pressure; LOA, limit of agreement.

Figure 4

Pulse wave analysis of exemplary photoplethysmography and radial blood pressure waveforms. Adapted from: ‘Photoplethysmogram pulse wave composition’, under CC BY 4.0. BP, blood pressure; PPG, photoplethysmography.