KardioAS Research

Publications & Research

Technical research behind the measurement systems

Preprints by Safin Aydar covering photoplethysmography, pulse-to-pulse interval measurement, HRV/PRV analysis and RR-interval-derived respiratory monitoring.



Preprint Published October 26, 2026 Version v1

Agreement Between ECG-Derived Heart Rate Variability and Finger-PPG Pulse Rate Variability During Rest: A Dual-Sensor Study Using Polar H10 and LifePulsePro

Safin Aydar

01

This dual-sensor study assesses agreement between ECG-derived RR intervals from a Polar H10 and finger-PPG pulse intervals from LifePulsePro during quiet seated rest. It also examines time-domain HRV/PRV, repeatability, pulse arrival time (PAT) and an exploratory velocity index.


Study designThirteen volunteers completed 25 three-minute resting recordings, yielding 5,244 accepted ECG–PPG interval pairs.
Interval agreementVolunteer-level mean absolute error was 5.23 ms, with mean within-record correlation r = 0.965 and signed bias of +0.010 ms.
HRV vs PRVPPG-derived SDNN and RMSSD were 4.73% and 15.22% higher, respectively, than the corresponding ECG measures.

Key interpretation

Finger PPG closely tracked beat-to-beat interval timing at rest, but PRV indices were not identical to ECG-derived HRV, particularly RMSSD. PAT and the derived velocity index remain exploratory and require independent vascular validation before clinical interpretation.


Read full text on Zenodo ↗ Zenodo record 23660057 · DOI 10.5281/zenodo.23660057

Preprint Published October 22, 2026 Version 1.0

RR-Interval-Derived Respiration From a Polar H10 Chest Sensor: Agreement With Simultaneous Chest Accelerometry During Fixed and Dynamically Changing Breathing

Safin Aydar

02

This methodological study evaluates respiratory rate derived from beat-to-beat RR intervals acquired with a Polar H10 chest sensor against simultaneously recorded chest accelerometry during controlled fixed and dynamically changing breathing.


Study designSixteen adult volunteers completed 31 recordings using a standardized 210-s controlled-breathing protocol.
Stable breathingAt 6 breaths/min, participant-level mean absolute error was 0.38 breaths/min with no significant systematic bias.
Dynamic breathingError increased during the respiratory-rate sweep, with rate-dependent underestimation at higher breathing rates and measurable temporal lag.

Read full text on Zenodo ↗ Zenodo record 23568023 · DOI 10.5281/zenodo.23568023

Preprint Published June 15, 2026 Version 1

LifePulsePro as an intelligent transmissive PPG sensor for heart rate variability assessment

Safin Aydar

03

This paper describes LifePulsePro as a transmissive photoplethysmographic system for pulse-wave acquisition, pulse-interval extraction and HRV analysis under standardized resting conditions. The device combines optical sensing, mechanical finger fixation, embedded signal processing, quality control and wired communication with external software.


Methodological focusExplains the physiological difference between ECG RR intervals and PPG-derived pulse intervals.
System conceptEmbedded pulse detection, interval calculation, quality control and structured data transfer.
Practical conclusionUnder stable resting conditions, pulse interval dynamics can closely reproduce ECG-derived timing for HRV assessment.

Read full text on Zenodo ↗ Zenodo record 20707180

Preprint Published June 15, 2026 Version v2

Accuracy assessment of the LifePulsePro photoplethysmographic sensor for pulse interval measurement compared with reference electrocardiography

Safin Aydar

04

A pilot technical validation compared consecutive LifePulsePro pulse intervals with ECG RR intervals recorded using the BTL Flexi 12 ECG system and BTL CardioPoint software. Recordings were made at rest in the supine position and synchronized interval by interval.


3subjects
6paired recordings
139matched intervals
1.90–8.97 msmean absolute error
0.20–1.03%mean absolute percentage error
0.9726–0.9978Pearson correlation

Key result

The largest individual absolute difference was 25.35 ms. Bland–Altman analysis reported a mean bias of 0.00 ms with 95% limits of agreement from −10.77 to +10.77 ms. The study presents these results as pilot evidence of good agreement under stable resting conditions, not as evidence that PPG replaces diagnostic ECG.


Read full text on Zenodo ↗ Zenodo record 20706051

Preprint Published June 8, 2026 Version 1

Comparison of Transmissive and Reflective Photoplethysmography for Pulse Wave Recording and Pulse Variability Analysis

Safin Aydar

05

The paper compares transmissive and reflective PPG with emphasis on a demanding use case: obtaining a reliable sequence of pulse-to-pulse intervals for subsequent HRV/PRV analysis, rather than merely displaying a visible pulse waveform.


Transmissive clip

Mechanical finger fixation and USB data transfer provide a more stable primary signal for interval-based analysis.

Reflective sensor

Suitable for approximate pulse monitoring in the described implementation, but scientific pulse-variability use requires further validation, strict signal-quality control and artifact handling.


Read full text on Zenodo ↗ Zenodo record 20594659


Research context

Why these studies matter

Together, the five papers address complementary questions: how a transmissive PPG system can be designed for interval analysis, how transmissive and reflective acquisition differ for pulse-variability work, how LifePulsePro interval measurements compare with reference ECG in a pilot resting study, how RR-interval-derived respiration from a Polar H10 tracks stable and dynamically changing breathing against simultaneous chest accelerometry, and how ECG-derived HRV compares with finger-PPG-derived PRV during seated rest using Polar H10 and LifePulsePro.