Principal component analysis of heart rate variability data in assessing cardiac autonomic neuropathy

Area of Science:

• Cardiology
• Neurology
• Biomedical Engineering

Background:

• Heart rate variability (HRV) is a key indicator for early diagnosis of cardiac autonomic neuropathy (CAN).
• Existing HRV analysis methods offer limited, differing insights into HRV time series.
• A comprehensive, multi-dimensional HRV analysis may improve CAN assessment.

Purpose of the Study:

• To evaluate the efficacy of a multi-dimensional HRV analysis approach for diagnosing CAN.
• To determine if principal component analysis (PCA) can effectively differentiate patients with CAN from healthy individuals using HRV data.

Main Methods:

• Collected multi-dimensional HRV data from 11 patients diagnosed with CAN and 71 control subjects.
• Applied principal component analysis (PCA) to analyze the HRV data.
• Utilized the two most significant principal components for classification.

Main Results:

• Principal component analysis (PCA) effectively separated patients with CAN from control subjects.
• The classification accuracy using the two principal components reached 87%.

Conclusions:

• A multi-dimensional HRV analysis, particularly using PCA, shows significant potential for accurate CAN diagnosis.
• This approach offers a more robust method for assessing HRV changes associated with CAN compared to traditional single-feature analyses.