Skip to content
DSML.KZ
Publications

Code · Zenodo · July 17, 2026

A stability-first dual-Lorentzian HRV analysis pipeline with dual-Lyapunov guarantee: Parametric efficiency, robust deployability and in silico pharmacological validation

Xiwen sun, Azhar Tursynova, Мадина Мансурова

Azhar Tursynova
Open via DOI

Abstract

Lightweight open-source Python toolbox for interpretable heart rate variability (HRV) analysis, constructed based on Lindblad open quantum dissipative dynamics and dual-Lyapunov stability theory. Core advantages of this toolbox: 1. Only 2 physiologically interpretable dynamical biomarkers (γ, κ, |ξ|), 75% parameter reduction compared with 2-State HMM baseline. 2. Up to 117× faster spectral fitting speed, optimized for low-power edge wearable deployment. 3. Built-in manifold physiological constraints to avoid divergent fitting outputs, guarantees 100% valid biomarker extraction from clinical RR sequences. 4. Included reproduce_paper.py to fully replicate the spectral fitting benchmark results in Table 1 of the corresponding submitted manuscript. This software package contains 5 core files: - lindblad_hrv.py: Core algorithm implementation - example.py: Quick demo for single-sample HRV analysis - reproduce_paper.py: Benchmark reproduction script - LICENSE: MIT open-source license text - readme.md: User guide, dependency list and citation template Released under MIT License. This software is the supplementary implementation for the manuscript submitted to Computer Methods and Programs in Biomedicine.

Heart Rate Variability and Autonomic ControlNon-Invasive Vital Sign MonitoringECG Monitoring and Analysisheart rate variabilityspectral fittingpharmacological validation

More by Azhar Tursynova