Machine learning based automated dynamic quantification of left heart chamber volumes

Il futuro dell'imaging è nel machine learning? Per l'ecocardiografia 3D parrebbe di sì. (Eur Heart J Cardiovasc Imaging 2018 Oct 9)

17 October Oct 2018 one month ago
  • Tamborini G, Fusini L, Pepi M

Studies have demonstrated the ability of a new automated algorithm for volumetric analysis of 3D echocardiographic (3DE) datasets to provide accurate and reproducible measurements of left ventricular and left atrial (LV, LA) volumes at end-systole and end-diastole. Recently, this methodology was expanded using a machine learning (ML) approach to automatically measure chamber volumes throughout the cardiac cycle, resulting in LV and LA volume-time curves. This paper aimed to validate ejection and filling parameters obtained from these curves by comparing them to independent well-validated reference techniques.

Studying 20 patients referred for cardiac magnetic resonance (CMR) examinations, the Authors obtained volume–time curves for both LV and LA chambers using the ML algorithm, and independently conventional 3DE volumetric analysis, and CMR images. Automatically derived LV and LA volumes and ejection/filling parameters were compared against both reference techniques.

In the first study testing a new automated ML-based analysis of 3DE datasets, the Authors conclude that automated ML algorithm can quickly measure dynamic LV and LA volumes and accurately analyse ejection/filling parameters. Incorporation of this algorithm into the clinical workflow may increase the utilization of 3DE imaging.


Reference
1. Narang A, Mor-Avi V, Prado A, Volpato V, Prater D, Tamborini G, Fusini L, Pepi M, Goyal N, Addetia K, Gonçalves A, Patel AR, Lang RM. Machine learning based automated dynamic quantification of left heart chamber volumes. Eur Heart J Cardiovasc Imaging 2018 Oct 9. doi: 10.1093/ehjci/jey137. [Epub ahead of print] Go to PubMed