A Prospective Study for Early Automated Detection of Cardiovascular Events Using CPAP Remote Monitoring Data in Patients with OSA

A Prospective Study on Early Detection of Cardiovascular Events via Remote Monitoring of Cheyne-Stokes Respiration - Abnormality Detection Using CSR Signals

Status

Active, not recruiting

Phases

Unknown

Study type

Observational

Source

JPRN

Registry ID

JPRN-UMIN000058672

Enrollment

1300

Registered

2025-08-02

Start date

2026-03-01

Completion date

Unknown

Last updated

2026-06-29

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Sleep apnea

Interventions

None listed

Eligibility

Sex/Gender

All

Inclusion criteria

Inclusion criteria: All patients currently undergoing treatment with the same CPAP device model at our institution.

Exclusion criteria

Exclusion criteria: Average CPAP usage rate less than 50% Average CPAP usage time less than 4 hours

Design outcomes

Primary

Measure	Time frame
The study monitors three-day moving averages of parameters derived from remote CPAP monitoring data, such as the percentage of Cheyne-Stokes respiration. When predefined threshold criteria are exceeded, an anomaly alert is triggered. These alerts are then compared with the actual onset of cardiovascular events to evaluate the performance of the anomaly detection system. Primary outcome measures include: Accuracy, Precision, F measure.	—

Measure

Time frame

The study monitors three-day moving averages of parameters derived from remote CPAP monitoring data, such as the percentage of Cheyne-Stokes respiration. When predefined threshold criteria are exceeded, an anomaly alert is triggered. These alerts are then compared with the actual onset of cardiovascular events to evaluate the performance of the anomaly detection system. Primary outcome measures include: Accuracy, Precision, F measure.

—

Secondary

Measure	Time frame
The predefined threshold criteria will incorporate not only the absolute values of three-day moving averages (e.g., CSR%) but also the growth rate of these averages. Additional parameters such as air leakage and other relevant indicators will also be considered to identify optimal alert conditions. Optimal cutoff values will be determined using receiver operating characteristic (ROC) curve analysis and the area under the curve (AUC). These methods will guide the selection of thresholds that maximize the accuracy of anomaly detection in relation to actual cardiovascular events.	—

Measure

Time frame

The predefined threshold criteria will incorporate not only the absolute values of three-day moving averages (e.g., CSR%) but also the growth rate of these averages. Additional parameters such as air leakage and other relevant indicators will also be considered to identify optimal alert conditions. Optimal cutoff values will be determined using receiver operating characteristic (ROC) curve analysis and the area under the curve (AUC). These methods will guide the selection of thresholds that maximize the accuracy of anomaly detection in relation to actual cardiovascular events.

—

Countries

Japan

Contacts

Public ContactKimimasa Saito

Medical Corporation MSC Miesleep Clinic

k1saito@carrot.ocn.ne.jp0596291159

Outcome results