Automatic selection of optimal Savitzky-Golay filter parameters for Coronary Wave Intensity Analysis

Research output: Chapter in Book/Report/Conference proceedingConference paperpeer-review

10 Citations (Scopus)

Abstract

Coronary Wave Intensity Analysis (cWIA) is a technique capable of separating the effects of proximal arterial haemodynamics from cardiac mechanics. The cWIA ability to establish a mechanistic link between coronary haemodynamics measurements and the underlying pathophysiology has been widely demonstrated. Moreover, the prognostic value of a cWIA-derived metric has been recently proved. However, the clinical application of cWIA has been hindered due to the strong dependence on the practitioners, mainly ascribable to the cWIA-derived indices sensitivity to the pre-processing parameters. Specifically, as recently demonstrated, the cWIA-derived metrics are strongly sensitive to the Savitzky-Golay (S-G) filter, typically used to smooth the acquired traces. This is mainly due to the inability of the S-G filter to deal with the different timescale features present in the measured waveforms. Therefore, we propose to apply an adaptive S-G algorithm that automatically selects pointwise the optimal filter parameters. The newly proposed algorithm accuracy is assessed against a cWIA gold standard, provided by a newly developed in-silico cWIA modelling framework, when physiological noise is added to the simulated traces. The adaptive S-G algorithm, when used to automatically select the polynomial degree of the S-G filter, provides satisfactory results with ≤ 10% error for all the metrics through all the levels of noise tested. Therefore, the newly proposed method makes cWIA fully automatic and independent from the practitioners, opening the possibility to multi-centre trials.
Original languageEnglish
Title of host publicationEngineering in Medicine and Biology Society (EMBC), 2014 36th Annual International Conference of the IEEE
PublisherIEEE
Pages5056-5059
Number of pages4
DOIs
Publication statusPublished - 1 Aug 2014
Event2014 36th Annual International Conference of the IEEEE - IL, Chicago, United States
Duration: 26 Aug 201430 Aug 2014

Publication series

NameEngineering in Medicine and Biology Society
PublisherIEEE
ISSN (Print)1557-180X

Conference

Conference2014 36th Annual International Conference of the IEEEE
Country/TerritoryUnited States
CityChicago
Period26/08/201430/08/2014

Keywords

  • Accuracy,Algorithm design and analysis,Measurement,Myocardium,Noise,Polynomials,Standards

Fingerprint

Dive into the research topics of 'Automatic selection of optimal Savitzky-Golay filter parameters for Coronary Wave Intensity Analysis'. Together they form a unique fingerprint.

Cite this