Automatic Detection of Atrial Fibrillation Using Basic Shannon Entropy of RR Interval Feature

International Conference on Science and Applied Science (Engineering and Educational Science) 2016

物理学;工业技术;教育

Afdala, Adfal^1 ; Nuryani, Nuryani^2 ; Nugroho, Anto Satriyo^2

Physics Department, Post Graduate Program, Sebelas Maret University, Indonesia^1

Informatics Department, Under Graduate Program, Sebelas Maret University, Jl. Ir. Sutami 36A, KentinganJebres Surakarta

57126, Indonesia^2

关键词: Atrial fibrillation;    Automatic Detection;    Heart beats;    Heart disease;    PhysioNet;    RR intervals;    Shannon entropy;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/795/1/012038/pdf
DOI  :  10.1088/1742-6596/795/1/012038

来源: IOP

【 摘 要 】

Atrial Fibrillation is one of heart disease, that common characterized by irregularity heart beat. Atrial fibrillation leads to severe complications such as cardiac failure with the subsequent risk of a stroke. A method to detect atrial fibrillation is needed to prevent a risk of atrial fibrillation. This research uses data from physionet in atrial fibrillation database category. The performance of Shannon entropy has the highest accuracy if a threshold is 0.5 with accuracy 89.79%, sensitivity 91.04% and specificity 89.01%. Based on the result we get a conclusion, the ability of Shannon entropy to detect atrial fibrillation is good.