Automated detection of atrial fibrillation using R-R intervals and multivariate-based classification

详细信息    查看全文

• 作者：Alan Kennedy ; BSc^a ; ^{kennedy-a23@email.ulster.ac.uk" class="auth_mail" title="E-mail the corresponding author} ; Dewar D. Finlay ; PhD^a ; Daniel Guldenring ; PhD^a ; Raymond R. Bond ; PhD^b ; Kieran Moran ; PhD^c ; James McLaughlin ; PhD^a
• 关键词：Atrial fibrillation ; R-R intervals ; Algorithms
• 刊名：Journal of Electrocardiology
• 出版年：2016
• 出版时间：November-December 2016
• 年：2016
• 卷：49
• 期：6
• 页码：871-876
• 全文大小：397 K

文摘

Automated detection of AF from the electrocardiogram (ECG) still remains a challenge. In this study, we investigated two multivariate-based classification techniques, Random Forests (RF) and k-nearest neighbor (k-nn), for improved automated detection of AF from the ECG. We have compiled a new database from ECG data taken from existing sources. R-R intervals were then analyzed using four previously described R-R irregularity measurements: (1) the coefficient of sample entropy (CoSEn), (2) the coefficient of variance (CV), (3) root mean square of the successive differences (RMSSD), and (4) median absolute deviation (MAD). Using outputs from all four R-R irregularity measurements, RF and k-nn models were trained. RF classification improved AF detection over CoSEn with overall specificity of 80.1% vs. 98.3% and positive predictive value of 51.8% vs. 92.1% with a reduction in sensitivity, 97.6% vs. 92.8%. k-nn also improved specificity and PPV over CoSEn; however, the sensitivity of this approach was considerably reduced (68.0%).