Application of an automatic adaptive filter for Heart Rate Variability analysis.

Area of Science:

• Biomedical Engineering
• Cardiology
• Signal Processing

Background:

• Analysis of Heart Rate Variability (HRV) is crucial for assessing cardiac health.
• Artifacts and noise in temporal series data, such as tachograms, can lead to erroneous interpretations.
• Manual elimination of ectopic beats and artifacts by specialists is time-consuming and labor-intensive.

Purpose of the Study:

• To develop and evaluate an automated data filtering method for tachogram pre-analysis in HRV.
• To compare the efficacy of an adaptive filtering technique against conventional manual editing methods.
• To assess the reliability of the proposed method across diverse patient populations and clinical conditions.

Main Methods:

• An adaptive filtering method was employed for automated data filtering of temporal series.
• The method was applied to 229 time series from patient groups including newborns, healthy adults, and those with specific diets or awaiting surgery.
• Correlation coefficients were calculated between the adaptive filter results and manual filtering by specialists for various HRV indices.

Main Results:

• A high correlation was observed between the adaptive filtering method and manual editing across most patient groups.
• Highly significant p values confirmed the reliability of the automated method for HRV analysis.
• Some parameter discrepancies were noted in the premature newborns group, warranting further investigation.

Conclusions:

• The proposed adaptive filtering method significantly enhances the efficiency of temporal series editing for HRV analysis.
• This automated approach provides a reliable alternative to manual artifact correction, saving valuable physician time.
• The method demonstrates broad applicability across different clinical scenarios, with specific considerations for vulnerable populations like premature infants.