This abstract is published in the Journal of the American Society of Echocardiography (volume 33, number 6, year 2020), as part of the proceedings of the American Society Echocardiography Conference 2020.

Title: Automated Wall Motion Scoring from Two-Dimensional Echocardiography Data using Combinations of Novel Echocardiographic Biomarkers

Authors: Parker, Andrew and Markham, Deborah and Woodward, Gary and Upton, Ross and Leeson, Paul and Beqiri, Arian and Mumith, Angela and Gao, Shan and Porumb, Mihaela and Hawkes, William and others

Wall motion scoring is a standardised semi-quantitative approach to record regional left ventricular (LV) function at rest and during stress echocardiography. Assessment depends on expert visual review of the echo images. We hypothesised automated scoring of wall motion would be possible using novel biomarkers known to describe variation in cardiac geometry over the cardiac cycle.

Traditional wall motion scores were determined by an expert cardiologist for 105 patients for the two-chamber view in each of the 6 segments, for both the rest and stress acquisitions. Further, the LV was contoured by an echocardiographer at end-diastole and end-systole, with thirteen points placed around the inner wall such that segments were approximately isolated by adjacent clusters of three points. In each manually contoured segment, geometric features were computed such as area, solidity, rectangularity and displacement. These features were then regressed into a single continuous value using principle component analysis, termed the Oxford Score.

In each manually contoured segment, the geometric features computed were seen to correlate with the assessed wall motion score, as shown in Figure 1. Further, a highly negative Oxford Score in each segment correlated well with a positive CAD clinical outcome, as shown in Figure 2.

A continuous wall motion score can be calculated based on novel geometric features derived for each LV segment that correlates with traditional wall motion scores, and with clinical outcomes. Automated quantitative approaches to wall motion scoring could remove subjectivity of commonly used clinical parameters.

Figure 1. The derived wall motion metrics correlate with traditional wall motion scores in resting two chamber data. (A) Resting data wall motion scores for each segment vs normalised segment. (B) Resting data wall motion scores for each segment vs normalised mean distances. (C) Resting data wall motion scores for each segment vs solidity. (D) Resting data wall motion scores for each segment vs rectangularity.



Figure 2. Oxford scores correlate with clinical outcome. (A) Oxford scores vs clinical outcomes for the resting data. (B) Oxford scores vs clinical outcomes for the stress data.

Share on

Share on facebook
Share on twitter
Share on linkedin

Further reading

Ultromics and Mayo Clinic to Use AI to Improve Diagnosis of Heart Failure

●  Heart failure is the leading cause of hospitalization for people over the age of 65 – but diagnosis can be challenging and is often delayed through lack of …

AI Solution that Predicts Heart Disease To Be Implemented In NHS Hospitals

Ultromics’ EchoGo system selected among first winners of new £140m NHS AI in Health and Care Award  A pioneering AI platform that can predict heart disease will be implemented in NHS hospitals as …

What’s to Gain from Strain Imaging

Strain imaging techniques have shown great promise in echocardiography to support early signs of cardiac disease and is considered a more accurate and reproducible quantitative analysis of ventricular function and …