Évaluation initiale d’une méthode de dépistage de la fibrillation atriale paroxystique

Abstract : Introduction: screening and early diagnosis of atrial fibrillation (AF) is an attractive solution to reduce the public health burden of embolic stroke. AF is often under diagnosed and asymptomatic. Previous work has demonstrated that long-duration non-invasive heart rate monitoring is the best diagnostic tool. Connected objects provide unprecedented opportunities in this scope. Materials and methods: a low-cost device using a heart rate sensor strap and a smartphone application collected RR intervals in hospitalized patients with paroxysmal AF undergoing simultaneous telemetry monitoring. RR intervals were then analyzed offline by 3 different automatic AF detection algorithms. The results from the analyzes were compared with the results of telemetry for the diagnosis of AF. Results and analysis: fifty patients were recorded using both modalities for an average of 18 hours. Four presented an episode of paroxysmal AF diagnosed by telemetry. The most efficient automatic AF detection algorithm based on our recording method had a sensitivity of 100 % and a specificity of 87 % in our population. Discussion: RR intervals derived from heart rate sensors enable long-term recordings and appear to provide an effective screening tool for AF episodes. Paroxysmal AF detection via automatic RR analysis is feasible in clinical practice, with high performance and low cost. A similar device with an access to the ECG data would be an interesting screening tool for AF in primary care.
Document type :
Master Thesis
Complete list of metadatas

Cited literature [101 references]  Display  Hide  Download

Contributor : Bu Carreire Université de Bordeaux <>
Submitted on : Thursday, December 14, 2017 - 4:45:27 PM
Last modification on : Wednesday, December 20, 2017 - 1:20:40 AM


Files produced by the author(s)


  • HAL Id : dumas-01664295, version 1



Jérémy Claude Jacques Papin. Évaluation initiale d’une méthode de dépistage de la fibrillation atriale paroxystique . Médecine humaine et pathologie. 2017. ⟨dumas-01664295⟩



Record views


Files downloads