Enhanced ECG signal based on heart rate variability: validation of the direct TR correction sequence in diabetes

Abstract

Background Heart rate variability (HRV) is a critical biomarker for assessing autonomic dysfunction, particularly in elderly and type 2 diabetes mellitus (T2DM) populations. However, traditional HRV indices, derived from R-R intervals (RRIs), often fail to detect subtle autonomic dysfunction in complex clinical scenarios. This study proposes a novel Direct TRI Correction (DTRC) sequence to optimize HRV assessment by directly extracting TR intervals (DTRI) and applying a single-step correction. The aim is to enhance accuracy in evaluating HRV, particularly in elderly and T2DM cohorts, where early detection is critical for preventing cardiovascular complications. Methods The study enrolled 124 participants (60 T2DM patients and 64 healthy controls) stratified into three groups based on glycemic control. HRV indices were derived from traditional RRI sequences, two-step corrected TRC sequences, and the novel single-step corrected DTRC sequences, respectively. Statistical analyses were performed on the HRV indices of three sequences to evaluate the comparative performance of DTRC sequences. Results DTRC-based HRV indices demonstrated superior sensitivity and discriminative capability compared to both RRI-based and TRC-based HRV indices. Significant improvements were observed in RMSSD (p = 0.007), pNN50 (p = 0.043), SDSD (p = 0.007), LHR (p = 0.016), SSR (p = 0.010), BEI (p = 0.002), and MSELS (p = 0.009). Visualization analysis and ROC curves confirmed clearer inter-group separation, particularly for T2DM patients with poor glycemic control. Conclusions The DTRC sequence significantly enhances HRV assessment accuracy, offering a reliable tool for early detection of autonomic dysfunction in high-risk populations. Its simplified correction process and improved sensitivity hold promise for clinical diagnostics and wearable health monitoring applications.