Atrial Fibrillation (AF), Heart Failure
Conditions
Keywords
Photoplethysmography, Machine learning, Remote telemonitoring, Wearable device, Digital biomarker, Arrhythmia detection
Brief summary
This is a prospective study validating a new machine-learning algorithm that detects atrial fibrillation (AF) from photoplethysmography (PPG) signals, developed for integration into the Seerlinq remote monitoring platform. This algorithm builds on the same core PPG signal-processing technology as Seerlinq's HeartCore device, a CE-certified (Class IIb, MDR) device that monitors left ventricular filling pressures in heart failure patients. The algorithm will be validated through internal cross-validation, external validation against an independent cohort with paired PPG-ECG recordings, and validation in a cohort of patients with paroxysmal atrial fibrillation and frequent sinus-AF transitions.
Detailed description
Atrial fibrillation (AF) and heart failure (HF) frequently coexist and share a bidirectional causal relationship; their concurrence is associated with worse clinical outcomes. Early detection of AF may enable timely intervention and improve outcomes. This study is prospectively validating a machine-learning algorithm for AF detection from PPG signals, intended for integration into the Seerlinq remote monitoring platform. This algorithm builds on the same core PPG signal-processing technology as Seerlinq's HeartCore device (a CE-certified, Class IIb device under the EU MDR that monitors left ventricular filling pressures in heart failure patients). It is a stand-alone algorithm designed specifically to detect clinically relevant (≥ 30s) atrial fibrillation. Validation of the algorithm will proceed in three stages: (1) internal cross-validation; (2) external validation against an independent cohort with paired PPG-ECG recordings, to confirm generalizability; and (3) validation in a cohort of patients with paroxysmal atrial fibrillation and frequent sinus-AF transitions, to assess performance during clinically challenging rhythm changes. The study is enrolling toward an estimated 1,000 unique PPG recordings. A 12-lead ECG is used to confirm cardiac rhythm classification (gold standard) as the reference for evaluating algorithm performance.
Interventions
OTHERPPG-based AF detection algorithm
The PPG-based atrial fibrillation detection algorithm is a non-invasive signal processing approach that analyzes photoplethysmographic waveforms obtained during remote monitoring. The algorithm evaluates pulse-to-pulse variability, waveform characteristics, and signal quality parameters to identify irregular rhythm patterns associated with atrial fibrillation and provide early detection of potential arrhythmic events.
Sponsors
Seerlinq s. r. o.
ACADEMY - občianske združenie
Premedix Academy
Study design
Observational model
COHORT
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Adults ≥18 years with a diagnosis of heart failure (HFrEF, HFmrEF, or HFpEF) * 12-lead ECG performed to confirm cardiac rhythm classification (AF vs. non-AF)
Exclusion criteria
* Missing a valid PPG recording
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Diagnostic accuracy (area under the ROC curve) of the PPG-based machine-learning algorithm for detecting clinically relevant AF (≥ 30s), compared with gold-standard 12-lead ECG | Through study completion (estimated November 2026) |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Sensitivity and specificity of the algorithm at the Youden-optimal threshold | Through study completion (estimated November 2026) | — |
| Positive predictive value and negative predictive value | Through study completion (estimated November 2026) | — |
| Average precision | Through study completion (estimated November 2026) | area under the precision-recall curve |
| Model calibration | Through study completion (estimated November 2026) | e.g., calibration curve / Brier score |
| Matthews correlation coefficient | Through study completion (estimated November 2026) | — |
| Overall classification accuracy | Through study completion (estimated November 2026) | — |
| Specificity and false-positive rate in the subgroup with frequent atrial/ventricular extrasystoles | Through study completion (estimated November 2026) | — |
| Accuracy of AF detection during sinus-AF transitions at the individual patient level | Through study completion (estimated November 2026) | — |
Countries
Slovakia
Contacts
CONTACTMarta Kollárová, MSc., PhD.
Outcome results
None listed