Heart failure & cardiomyopathiesDiagnostic or prognostic study
Smartphone-Based Recognition of Heart Failure by Means of Microelectromechanical Sensors
- Diagnostic accuracy study
- multicentre
- prespecified pooled data analysis
- bootstrap aggregation validation
Authors' conclusion
Smartphone-based assessment of cardiac function using seismocardiography is feasible and differentiates patients with HF from control subjects with high diagnostic accuracy.
Study context
- Population
- Inpatients and outpatients with stage C heart failure (HF) and control subjects, Finland and US174 inpatients · 172 outpatients
- Participants
- 217 vs 786
- Comparison
- Smartphone seismocardiography for detection of HF vs control subjects
- HFn = 217
- Controln = 786reference
- Outcome
- Diagnostic accuracy of smartphone seismocardiography algorithms for detection of HF
Performance
As reported in the abstract
Detection of HF (both HF cohorts, decision threshold 0.5)
- AUC
- 0.95
- Sensitivity
- 85%
- Specificity
- 90%
- Accuracy
- 89%
- Positive likelihood ratio
- 8.50
- Negative likelihood ratio
- 0.17
Further findings
Accuracy by subgroups (age, sex, BMI, AF)
- Accuracy not significantly different in subgroups based on age, sex, BMI and AF
Original publication
Smartphone-Based Recognition of Heart Failure by Means of Microelectromechanical Sensors. JACC Heart Fail. Published 3 April 2024. doi:10.1016/j.jchf.2024.01.022
Summarised from the published abstract; every number is checked against it